Search results for: artificial intellogence

Authors: Gyu-Dong Kim, Wuk-Jae Yoo, Sang-Youl Lee

Abstract:

Composite structures offer numerous advantages over conventional structural systems in the form of higher specific stiffness and strength, lower life-cycle costs, and benefits such as easy installation and improved safety. Recently, there has been a considerable increase in the use of composites in engineering applications and as wraps for seismic upgrading and repairs. However, these composites deteriorate with time because of outdated materials, excessive use, repetitive loading, climatic conditions, manufacturing errors, and deficiencies in inspection methods. In particular, damaged fibers in a composite result in significant degradation of structural performance. In order to reduce the failure probability of composites in service, techniques to assess the condition of the composites to prevent continual growth of fiber damage are required. Condition assessment technology and nondestructive evaluation (NDE) techniques have provided various solutions for the safety of structures by means of detecting damage or defects from static or dynamic responses induced by external loading. A variety of techniques based on detecting the changes in static or dynamic behavior of isotropic structures has been developed in the last two decades. These methods, based on analytical approaches, are limited in their capabilities in dealing with complex systems, primarily because of their limitations in handling different loading and boundary conditions. Recently, investigators have introduced direct search methods based on metaheuristics techniques and artificial intelligence, such as genetic algorithms (GA), simulated annealing (SA) methods, and neural networks (NN), and have promisingly applied these methods to the field of structural identification. Among them, GAs attract our attention because they do not require a considerable amount of data in advance in dealing with complex problems and can make a global solution search possible as opposed to classical gradient-based optimization techniques. In this study, we propose an alternative damage-detection technique that can determine the degraded stiffness distribution of vibrating laminated composites made of Glass Fiber-reinforced Polymer (GFRP). The proposed method uses a modified form of the bivariate Gaussian distribution function to detect degraded stiffness characteristics. In addition, this study presents a method to detect the fiber property variation of laminated composite plates from the micromechanical point of view. The finite element model is used to study free vibrations of laminated composite plates for fiber stiffness degradation. In order to solve the inverse problem using the combined method, this study uses only first mode shapes in a structure for the measured frequency data. In particular, this study focuses on the effect of the interaction among various parameters, such as fiber angles, layup sequences, and damage distributions, on fiber-stiffness damage detection.

Keywords: stiffness detection, fiber damage, genetic algorithm, layup sequences

Procedia PDF Downloads 272

Authors: Eiji Nakamachi, Tsuyoshi Eguchi, Sayo Yamamoto, Yusuke Morita, H. Sakamoto

Abstract:

In this study, we aim to reveal the relationship between the meniscal tearing and the articular cartilage injury of knee joint by using the dynamic explicit finite element (FE) method. Meniscal injuries reduce its functional ability and consequently increase the load on the articular cartilage of knee joint. In order to prevent the induction of osteoarthritis (OA) caused by meniscal injuries, many medical treatment techniques, such as artificial meniscus replacement and meniscal regeneration, have been developed. However, it is reported that these treatments are not the comprehensive methods. In order to reveal the fundamental mechanism of OA induction, the mechanical characterization of meniscus under the condition of normal and injured states is carried out by using FE analyses. At first, a FE model of the human knee joint in the case of normal state – ‘intact’ - was constructed by using the magnetron resonance (MR) tomography images and the image construction code, Materialize Mimics. Next, two types of meniscal injury models with the radial tears of medial and lateral menisci were constructed. In FE analyses, the linear elastic constitutive law was adopted for the femur and tibia bones, the visco-hyperelastic constitutive law for the articular cartilage, and the visco-anisotropic hyperelastic constitutive law for the meniscus, respectively. Material properties of articular cartilage and meniscus were identified using the stress-strain curves obtained by our compressive and the tensile tests. The numerical results under the normal walking condition revealed how and where the maximum compressive stress occurred on the articular cartilage. The maximum compressive stress and its occurrence point were varied in the intact and two meniscal tear models. These compressive stress values can be used to establish the threshold value to cause the pathological change for the diagnosis. In this study, FE analyses of knee joint were carried out to reveal the influence of meniscal injuries on the cartilage injury. The following conclusions are obtained. 1. 3D FE model, which consists femur, tibia, articular cartilage and meniscus was constructed based on MR images of human knee joint. The image processing code, Materialize Mimics was used by using the tetrahedral FE elements. 2. Visco-anisotropic hyperelastic constitutive equation was formulated by adopting the generalized Kelvin model. The material properties of meniscus and articular cartilage were determined by curve fitting with experimental results. 3. Stresses on the articular cartilage and menisci were obtained in cases of the intact and two radial tears of medial and lateral menisci. Through comparison with the case of intact knee joint, two tear models show almost same stress value and higher value than the intact one. It was shown that both meniscal tears induce the stress localization in both medial and lateral regions. It is confirmed that our newly developed FE analysis code has a potential to be a new diagnostic system to evaluate the meniscal damage on the articular cartilage through the mechanical functional assessment.

Keywords: finite element analysis, hyperelastic constitutive law, knee joint injury, meniscal tear, stress concentration

Procedia PDF Downloads 246

Authors: Ja-Keoung Koo, Kensuke Nakamura, Hyohun Kim, Dongwha Shin, Yeonseok Kim, Ji-Su Ahn, Byung-Woo Hong

Abstract:

The machine learning techniques based on a convolutional neural network (CNN) have been actively developed and successfully applied to a variety of image analysis tasks including reconstruction, noise reduction, resolution enhancement, segmentation, motion estimation, object recognition. The classical visual information processing that ranges from low level tasks to high level ones has been widely developed in the deep learning framework. It is generally considered as a challenging problem to derive visual interpretation from high dimensional imagery data. A CNN is a class of feed-forward artificial neural network that usually consists of deep layers the connections of which are established by a series of non-linear operations. The CNN architecture is known to be shift invariant due to its shared weights and translation invariance characteristics. However, it is often computationally intractable to optimize the network in particular with a large number of convolution layers due to a large number of unknowns to be optimized with respect to the training set that is generally required to be large enough to effectively generalize the model under consideration. It is also necessary to limit the size of convolution kernels due to the computational expense despite of the recent development of effective parallel processing machinery, which leads to the use of the constantly small size of the convolution kernels throughout the deep CNN architecture. However, it is often desired to consider different scales in the analysis of visual features at different layers in the network. Thus, we propose a CNN model where different sizes of the convolution kernels are applied at each layer based on the random projection. We apply random filters with varying sizes and associate the filter responses with scalar weights that correspond to the standard deviation of the random filters. We are allowed to use large number of random filters with the cost of one scalar unknown for each filter. The computational cost in the back-propagation procedure does not increase with the larger size of the filters even though the additional computational cost is required in the computation of convolution in the feed-forward procedure. The use of random kernels with varying sizes allows to effectively analyze image features at multiple scales leading to a better generalization. The robustness and effectiveness of the proposed CNN based on random kernels are demonstrated by numerical experiments where the quantitative comparison of the well-known CNN architectures and our models that simply replace the convolution kernels with the random filters is performed. The experimental results indicate that our model achieves better performance with less number of unknown weights. The proposed algorithm has a high potential in the application of a variety of visual tasks based on the CNN framework. Acknowledgement—This work was supported by the MISP (Ministry of Science and ICT), Korea, under the National Program for Excellence in SW (20170001000011001) supervised by IITP, and NRF-2014R1A2A1A11051941, NRF2017R1A2B4006023.

Keywords: deep learning, convolutional neural network, random kernel, random projection, dimensionality reduction, object recognition

Procedia PDF Downloads 289

Authors: Duncan Wallace, M-Tahar Kechadi

Abstract:

In recent years, Machine Learning (ML) approaches have been successfully applied to an analysis of patient symptom data in the context of disease diagnosis, at least where such data is well codified. However, much of the data present in Electronic Health Records (EHR) are unlikely to prove suitable for classic ML approaches. Furthermore, as scores of data are widely spread across both hospitals and individuals, a decentralized, computationally scalable methodology is a priority. The focus of this paper is to develop a method to predict outliers in an out-of-hours healthcare provision center (OOHC). In particular, our research is based upon the early identification of patients who have underlying conditions which will cause them to repeatedly require medical attention. OOHC act as an ad-hoc delivery of triage and treatment, where interactions occur without recourse to a full medical history of the patient in question. Medical histories, relating to patients contacting an OOHC, may reside in several distinct EHR systems in multiple hospitals or surgeries, which are unavailable to the OOHC in question. As such, although a local solution is optimal for this problem, it follows that the data under investigation is incomplete, heterogeneous, and comprised mostly of noisy textual notes compiled during routine OOHC activities. Through the use of Deep Learning methodologies, the aim of this paper is to provide the means to identify patient cases, upon initial contact, which are likely to relate to such outliers. To this end, we compare the performance of Long Short-Term Memory, Gated Recurrent Units, and combinations of both with Convolutional Neural Networks. A further aim of this paper is to elucidate the discovery of such outliers by examining the exact terms which provide a strong indication of positive and negative case entries. While free-text is the principal data extracted from EHRs for classification, EHRs also contain normalized features. Although the specific demographical features treated within our corpus are relatively limited in scope, we examine whether it is beneficial to include such features among the inputs to our neural network, or whether these features are more successfully exploited in conjunction with a different form of a classifier. In this section, we compare the performance of randomly generated regression trees and support vector machines and determine the extent to which our classification program can be improved upon by using either of these machine learning approaches in conjunction with the output of our Recurrent Neural Network application. The output of our neural network is also used to help determine the most significant lexemes present within the corpus for determining high-risk patients. By combining the confidence of our classification program in relation to lexemes within true positive and true negative cases, with an inverse document frequency of the lexemes related to these cases, we can determine what features act as the primary indicators of frequent-attender and non-frequent-attender cases, providing a human interpretable appreciation of how our program classifies cases.

Keywords: artificial neural networks, data-mining, machine learning, medical informatics

Procedia PDF Downloads 131

Authors: A. Chávez, A. Rodríguez, F. Pinzón

Abstract:

Society is concerned about measures of environmental, economic and social impacts generated in the solid waste disposal. These places of confinement, also known as landfills, are locations where problems of pollution and damage to human health are reduced. They are technically designed and operated, using engineering principles, storing the residue in a small area, compact it to reduce volume and covering them with soil layers. Problems preventing liquid (leachate) and gases produced by the decomposition of organic matter. Despite planning and site selection for disposal, monitoring and control of selected processes, remains the dilemma of the leachate as extreme concentration of pollutants, devastating soil, flora and fauna; aggressive processes requiring priority attention. A biological technology is the activated sludge system, used for tributaries with high pollutant loads. Since transforms biodegradable dissolved and particulate matter into CO2, H2O and sludge; transform suspended and no Settleable solids; change nutrients as nitrogen and phosphorous; and degrades heavy metals. The microorganisms that remove organic matter in the processes are in generally facultative heterotrophic bacteria, forming heterogeneous populations. Is possible to find unicellular fungi, algae, protozoa and rotifers, that process the organic carbon source and oxygen, as well as the nitrogen and phosphorus because are vital for cell synthesis. The mixture of the substrate, in this case sludge leachate, molasses and wastewater is maintained ventilated by mechanical aeration diffusers. Considering as the biological processes work to remove dissolved material (< 45 microns), generating biomass, easily obtained by decantation processes. The design consists of an artificial support and aeration pumps, favoring develop microorganisms (denitrifying) using oxygen (O) with nitrate, resulting in nitrogen (N) in the gas phase. Thus, avoiding negative effects of the presence of ammonia or phosphorus. Overall the activated sludge system includes about 8 hours of hydraulic retention time, which does not prevent the demand for nitrification, which occurs on average in a value of MLSS 3,000 mg/L. The extended aeration works with times greater than 24 hours detention; with ratio of organic load/biomass inventory under 0.1; and average stay time (sludge age) more than 8 days. This project developed a pilot system with sludge leachate from Doña Juana landfill - RSDJ –, located in Bogota, Colombia, where they will be subjected to a process of activated sludge and extended aeration through a sequential Bach reactor - SBR, to be dump in hydric sources, avoiding ecological collapse. The system worked with a dwell time of 8 days, 30 L capacity, mainly by removing values of BOD and COD above 90%, with initial data of 1720 mg/L and 6500 mg/L respectively. Motivating the deliberate nitrification is expected to be possible commercial use diffused aeration systems for sludge leachate from landfills.

Keywords: sludge, landfill, leachate, SBR

Procedia PDF Downloads 270

Authors: D. Malouch, M. Berchel, C. Dreanno, S. Stachowski-Haberkorn, P-A. Jaffres

Abstract:

Biofouling is a complex natural phenomenon that involves biological, physical and chemical properties related to the environment, the submerged surface and the living organisms involved. Bio-colonization of artificial structures can cause various economic and environmental impacts. The increase in costs associated with the over-consumption of fuel from biocolonized vessels has been widely studied. Measurement drifts from submerged sensors, as well as obstructions in heat exchangers, and deterioration of offshore structures are major difficulties that industries are dealing with. Therefore, surfaces that inhibit biocolonization are required in different areas (water treatment, marine paints, etc.) and many efforts have been devoted to produce efficient and eco-compatible antifouling agents. The different steps of surface fouling are widely described in literature. Studying the biofilm and its stages provides a better understanding of how to elaborate more efficient antifouling strategies. Several approaches are currently applied, such as the use of biocide anti-fouling paint (mainly with copper derivatives) and super-hydrophobic coatings. While these two processes are proving to be the most effective, they are not entirely satisfactory, especially in a context of a changing legislation. Nowadays, the challenge is to prevent biofouling with non-biocide compounds, offering a cost effective solution, but with no toxic effects on marine organisms. Since the micro-fouling phase plays an important role in the regulation of the following steps of biofilm formation, it is desired to reduce or delate biofouling of a given surface by inhibiting the micro-fouling at its early stages. In our recent works, we reported that some amphiphilic compounds exhibited bacteriostatic or bactericidal properties at a concentration that did not affect mammalian eukaryotic cells. These remarkable properties invited us to assess this type of bio-inspired phospholipids to prevent the colonization of surfaces by marine bacteria. Of note, other studies reported that amphiphilic compounds interacted with bacteria leading to a reduction of their development. An amphiphilic compound is a molecule consisting of a hydrophobic domain and a polar head (ionic or non-ionic). These compounds appear to have interesting antifouling properties: some ionic compounds have shown antimicrobial activity, and zwitterions can reduce nonspecific adsorption of proteins. Herein, we investigate the potential of amphiphilic compounds as inhibitors of bacterial growth and marine biofilm formation. The aim of this study is to compare the efficacy of four synthetic phospholipids that features a cationic charge or a zwitterionic polar-head group to prevent microfouling with marine bacteria. Toxicity of these compounds was also studied in order to identify the most promising compounds that inhibit biofilm development and show low cytotoxicity on two links representative of coastal marine food webs: phytoplankton and oyster larvae.

Keywords: amphiphilic phospholipids, biofilm, marine fouling, non-toxique assays

Procedia PDF Downloads 134

Authors: Till Gramberg, Erwin Gross, Christoph Birenbaum

Abstract:

This paper presents the approach of the Easy2IoT research project. Easy2IoT aims to enable companies in the prefabrication sheet metal and sheet metal processing industry to enter the Industrial Internet of Things (IIoT) with a low-threshold and cost-effective approach. It focuses on the development of physical hardware and software to easily capture machine activities from on a sawing machine, benefiting various stakeholders in the SME value chain, including machine operators, tool manufacturers and service providers. The methodological approach of Easy2IoT includes an in-depth requirements analysis and customer interviews with stakeholders along the value chain. Based on these insights, actions, requirements and potential solutions for smart services are derived. The focus is on providing actionable recommendations, competencies and easy integration through no-/low-code applications to facilitate implementation and connectivity within production networks. At the core of the project is a novel, non-invasive measurement and analysis system that can be easily deployed and made IIoT-ready. This system collects machine data without interfering with the machines themselves. It does this by non-invasively measuring the tension on a sawing machine. The collected data is then connected and analyzed using artificial intelligence (AI) to provide smart services through a platform-based application. Three Smart Services are being developed within Easy2IoT to provide immediate benefits to users: Wear part and product material condition monitoring and predictive maintenance for sawing processes. The non-invasive measurement system enables the monitoring of tool wear, such as saw blades, and the quality of consumables and materials. Service providers and machine operators can use this data to optimize maintenance and reduce downtime and material waste. Optimize Overall Equipment Effectiveness (OEE) by monitoring machine activity. The non-invasive system tracks machining times, setup times and downtime to identify opportunities for OEE improvement and reduce unplanned machine downtime. Estimate CO2 emissions for connected machines. CO2 emissions are calculated for the entire life of the machine and for individual production steps based on captured power consumption data. This information supports energy management and product development decisions. The key to Easy2IoT is its modular and easy-to-use design. The non-invasive measurement system is universally applicable and does not require specialized knowledge to install. The platform application allows easy integration of various smart services and provides a self-service portal for activation and management. Innovative business models will also be developed to promote the sustainable use of the collected machine activity data. The project addresses the digitalization gap between large enterprises and SME. Easy2IoT provides SME with a concrete toolkit for IIoT adoption, facilitating the digital transformation of smaller companies, e.g. through retrofitting of existing machines.

Keywords: smart services, IIoT, IIoT-platform, industrie 4.0, big data

Procedia PDF Downloads 73

Authors: Orit Wolf

Abstract:

Classical music performance is undergoing a profound transformation, marked by a confluence of technological advancements and evolving cultural dynamics. This academic paper explores the multifaceted changes and challenges faced by classical music performance, considering the impact of artificial intelligence (AI) along with other vital factors shaping this evolution. In the contemporary era, classical music is experiencing shifts in performance practices. This paper delves into these changes, emphasizing the need for adaptability within the classical music world. From repertoire selection and concert formats to artistic expression, performers and institutions navigate a delicate balance between tradition and innovation. We explore how these changes impact the authenticity and vitality of classical music performances. Furthermore, the influence of AI in the classical music concert world cannot be underestimated. AI technologies are making inroads into various aspects, from composition assistance to rehearsal and live performances. This paper examines the transformative effects of AI, considering how it enhances precision, adaptability, and creative exploration for musicians. We explore the implications for composers, performers, and the overall concert experience while addressing ethical concerns and creative opportunities. In addition to AI, there is the importance of cross-genre interactions within the classical music sphere. Mash-ups and collaborations with artists from diverse musical backgrounds are redefining the boundaries of classical music and creating works that resonate with a wider and more diverse audience. The benefits of cross-pollination in classical music seem crucial, offering a fresh perspective to listeners. As an active concert artist, Orit Wolf will share how the expectations of classical music audiences are evolving. Modern concertgoers seek not only exceptional musical performances but also immersive experiences that may involve technology, multimedia, and interactive elements. This paper examines how classical musicians and institutions are adapting to these changing expectations, using technology and innovative concert formats to deliver a unique and enriched experience to their audiences. As these changes and challenges reshape the classical music world, the need for a harmonious coexistence of tradition, technology, and innovation becomes evident. Musicians, composers, and institutions are striving to find a balance that ensures classical music remains relevant in a rapidly changing cultural landscape while maintaining the value it brings to compositions and audiences. This paper, therefore, aims to explore the evolving trends in classical music performance. It considers the influence of AI as one element within the broader context of change, highlighting the necessity of adaptability, cross-genre interactions, and a response to evolving audience expectations. By doing so, the classical music world can navigate this transformative period while preserving its timeless traditions and adding value to both performers and listeners. Orit Wolf, an international concert pianist, fulfils her vision to bring this music in new ways to mass audiences and will share her personal and professional experience as an artist who goes on stage and makes disruptive concerts.

Keywords: cross culture collaboration, music performance and ai, classical music in the digital age, classical concerts, innovation and technology, performance innovation, audience engagement in classical concerts

Procedia PDF Downloads 64

Authors: Catherine McWhorter

Abstract:

Robotics is a fast-evolving field lacking comprehensive and harm-mitigating regulation; it also lacks critical data on how human-robot interaction (HRI) may affect human psychology. As most anthropomorphic robots are intended as substitutes for humans, this paper asserts that the commercial robotics industry should be preemptively regulated at the federal level such that robots capable of embodying a victim role in criminal scenarios (“vicbots”) are prohibited until clinical studies determine their effects on the user and society. The results of these studies should then inform more permanent legislation that strives to mitigate risks of harm without infringing upon fundamental rights or stifling innovation. This paper explores these concepts through the lens of the sex robot industry. The sexbot industry offers some of the most realistic, interactive, and customizable robots for sale today. From approximately 2010 until 2017, some sex robot producers, such as True Companion, actively promoted ‘vicbot’ culture with personalities like “Frigid Farrah” and “Young Yoko” but received significant public backlash for fetishizing rape and pedophilia. Today, “Frigid Farrah” and “Young Yoko” appear to have vanished. Sexbot producers have replaced preprogrammed vicbot personalities in favor of one generic, customizable personality. According to the manufacturer ainidoll.com, when asked, there is only one thing the user won’t be able to program the sexbot to do – “…give you drama”. The ability to customize vicbot personas is possible with today’s generic personality sexbots and may undermine the intent of some current legislative efforts. Current debate on the effects of vicbots indicates a lack of consensus. Some scholars suggest vicbots may reduce the rate of actual sex crimes, and some suggest that vicbots will, in fact, create sex criminals, while others cite their potential for rehabilitation. Vicbots may have value in some instances when prescribed by medical professionals, but the overall uncertainty and lack of data further underscore the need for preemptive regulation and clinical research. Existing literature on exposure to media violence and its effects on prosocial behavior, human aggression, and addiction may serve as launch points for specific studies into the hyperrealism of vicbots. Of course, the customization, anthropomorphism and artificial intelligence of sexbots, and therefore more mainstream robots, will continue to evolve. The existing sexbot industry offers an opportunity to preemptively regulate and to research answers to these and many more questions before this type of technology becomes even more advanced and mainstream. Robots pose complicated moral, ethical, and legal challenges, most of which are beyond the scope of this paper. By examining the possibility for custom vicbots via the sexbots industry, reviewing existing literature on regulation, media violence, and vicbot user effects, this paper strives to underscore the need for preemptive federal regulation prohibiting vicbot capabilities in robots while advocating for further research into the potential for the user and societal harm by the same.

Keywords: human-robot interaction effects, regulation, research, robots

Procedia PDF Downloads 206

Authors: Maryam Fallahpoor, Biswajeet Pradhan

Abstract:

Background: Artificial intelligence (AI) serves as a valuable tool in mitigating the scarcity of human resources required for the evaluation and categorization of vast quantities of medical imaging data. When AI operates with optimal precision, it minimizes the demand for human interpretations and, thereby, reduces the burden on radiologists. Among various AI approaches, deep learning (DL) stands out as it obviates the need for feature extraction, a process that can impede classification, especially with intricate datasets. The advent of DL models has ushered in a new era in medical imaging, particularly in the context of COVID-19 detection. Traditional 2D imaging techniques exhibit limitations when applied to volumetric data, such as Computed Tomography (CT) scans. Medical images predominantly exist in one of two formats: neuroimaging informatics technology initiative (NIfTI) and digital imaging and communications in medicine (DICOM). Purpose: This study aims to employ DL for the classification of COVID-19-infected pulmonary patients and normal cases based on 3D CT scans while investigating the impact of image format. Material and Methods: The dataset used for model training and testing consisted of 1245 patients from IranMehr Hospital. All scans shared a matrix size of 512 × 512, although they exhibited varying slice numbers. Consequently, after loading the DICOM CT scans, image resampling and interpolation were performed to standardize the slice count. All images underwent cropping and resampling, resulting in uniform dimensions of 128 × 128 × 60. Resolution uniformity was achieved through resampling to 1 mm × 1 mm × 1 mm, and image intensities were confined to the range of (−1000, 400) Hounsfield units (HU). For classification purposes, positive pulmonary COVID-19 involvement was designated as 1, while normal images were assigned a value of 0. Subsequently, a U-net-based lung segmentation module was applied to obtain 3D segmented lung regions. The pre-processing stage included normalization, zero-centering, and shuffling. Four distinct 3D CNN models (ResNet152, ResNet50, DensNet169, and DensNet201) were employed in this study. Results: The findings revealed that the segmentation technique yielded superior results for DICOM images, which could be attributed to the potential loss of information during the conversion of original DICOM images to NIFTI format. Notably, ResNet152 and ResNet50 exhibited the highest accuracy at 90.0%, and the same models achieved the best F1 score at 87%. ResNet152 also secured the highest Area under the Curve (AUC) at 0.932. Regarding sensitivity and specificity, DensNet201 achieved the highest values at 93% and 96%, respectively. Conclusion: This study underscores the capacity of deep learning to classify COVID-19 pulmonary involvement using real 3D hospital data. The results underscore the significance of employing DICOM format 3D CT images alongside appropriate pre-processing techniques when training DL models for COVID-19 detection. This approach enhances the accuracy and reliability of diagnostic systems for COVID-19 detection.

Keywords: deep learning, COVID-19 detection, NIFTI format, DICOM format

Procedia PDF Downloads 88

Authors: Ganesh V., Asit Kumar Khanra

Abstract:

An in-situ powder metallurgy technique was employed to create Ni-Al3Ni/Al3Ni2 core-shell-shaped aluminum-based intermetallic reinforced composites. The impact of Ni addition on the phase composition, microstructure, and mechanical characteristics of the Al-4Cu-xNi (x = 0, 2, 4, 6, 8, 10 wt.%) in relation to various sintering temperatures was investigated. Microstructure evolution was extensively examined using X-ray diffraction (XRD), scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDX), and transmission electron microscopy (TEM) techniques. Initially, under sintering conditions, the formation of "Single Core-Shell" structures was observed, consisting of Ni as the core with Al3Ni2 intermetallic, whereas samples sintered at 620°C exhibited both "Single Core-Shell" and "Double Core-Shell" structures containing Al3Ni2 and Al3Ni intermetallics formed between the Al matrix and Ni reinforcements. The composite achieved a high compressive yield strength of 198.13 MPa and ultimate strength of 410.68 MPa, with 24% total elongation for the sample containing 10 wt.% Ni. Additionally, there was a substantial increase in hardness, reaching 124.21 HV, which is 2.4 times higher than that of the base aluminum. Nanoindentation studies showed hardness values of 1.54, 4.65, 21.01, 13.16, 5.52, 6.27, and 8.39GPa corresponding to α-Al matrix, Ni, Al3Ni2, Ni and Al3Ni2 interface, Al3Ni, and their respective interfaces. Even at 200°C, it retained 54% of its room temperature strength (90.51 MPa). To investigate the deformation behavior of the composite material, experiments were conducted at deformation temperatures ranging from 300°C to 500°C, with strain rates varying from 0.0001s-1 to 0.1s-1. A sine-hyperbolic constitutive equation was developed to characterize the flow stress of the composite, which exhibited a significantly higher hot deformation activation energy of 231.44 kJ/mol compared to the self-diffusion of pure aluminum. The formation of Al2Cu intermetallics at grain boundaries and Al3Ni2/Al3Ni within the matrix hindered dislocation movement, leading to an increase in activation energy, which might have an adverse effect on high-temperature applications. Two models, the Strain-compensated Arrhenius model and the Artificial Neural Network (ANN) model, were developed to predict the composite's flow behavior. The ANN model outperformed the Strain-compensated Arrhenius model with a lower average absolute relative error of 2.266%, a smaller root means square error of 1.2488 MPa, and a higher correlation coefficient of 0.9997. Processing maps revealed that the optimal hot working conditions for the composite were in the temperature range of 420-500°C and strain rates between 0.0001s-1 and 0.001s-1. The changes in the composite microstructure were successfully correlated with the theory of processing maps, considering temperature and strain rate conditions. The uneven distribution in the shape and size of Core-shell/Al3Ni intermetallic compounds influenced the flow stress curves, leading to Dynamic Recrystallization (DRX), followed by partial Dynamic Recovery (DRV), and ultimately strain hardening. This composite material shows promise for applications in the automobile and aerospace industries.

Keywords: core-shell structure, hot deformation, intermetallic compounds, powder metallurgy

Procedia PDF Downloads 19

Authors: Iman Farasat, Howard M. Salis

Abstract:

Engineered genetic circuits reprogram cellular behavior to act as living computers with applications in detecting cancer, creating self-controlling artificial tissues, and dynamically regulating metabolic pathways. Phenemenological models are often used to simulate and design genetic circuit behavior towards a desired behavior. While such models assume that each circuit component’s function is modular and independent, even small changes in a circuit (e.g. a new promoter, a change in transcription factor expression level, or even a new media) can have significant effects on the circuit’s function. Here, we use statistical thermodynamics to account for the several factors that control transcriptional regulation in bacteria, and experimentally demonstrate the model’s accuracy across 825 measurements in several genetic contexts and hosts. We then employ our first principles model to design, experimentally construct, and characterize a family of signal amplifying genetic circuits (genetic OpAmps) that expand the dynamic range of cell sensors. To develop these models, we needed a new approach to measuring the in vivo binding free energies of transcription factors (TFs), a key ingredient of statistical thermodynamic models of gene regulation. We developed a new high-throughput assay to measure RNA polymerase and TF binding free energies, requiring the construction and characterization of only a few constructs and data analysis (Figure 1A). We experimentally verified the assay on 6 TetR-homolog repressors and a CRISPR/dCas9 guide RNA. We found that our binding free energy measurements quantitatively explains why changing TF expression levels alters circuit function. Altogether, by combining these measurements with our biophysical model of translation (the RBS Calculator) as well as other measurements (Figure 1B), our model can account for changes in TF binding sites, TF expression levels, circuit copy number, host genome size, and host growth rate (Figure 1C). Model predictions correctly accounted for how these 8 factors control a promoter’s transcription rate (Figure 1D). Using the model, we developed a design framework for engineering multi-promoter genetic circuits that greatly reduces the number of degrees of freedom (8 factors per promoter) to a single dimensionless unit. We propose the Ptashne (Pt) number to encapsulate the 8 co-dependent factors that control transcriptional regulation into a single number. Therefore, a single number controls a promoter’s output rather than these 8 co-dependent factors, and designing a genetic circuit with N promoters requires specification of only N Pt numbers. We demonstrate how to design genetic circuits in Pt number space by constructing and characterizing 15 2-repressor OpAmp circuits that act as signal amplifiers when within an optimal Pt region. We experimentally show that OpAmp circuits using different TFs and TF expression levels will only amplify the dynamic range of input signals when their corresponding Pt numbers are within the optimal region. Thus, the use of the Pt number greatly simplifies the genetic circuit design, particularly important as circuits employ more TFs to perform increasingly complex functions.

Keywords: transcription factor, synthetic biology, genetic circuit, biophysical model, binding energy measurement

Procedia PDF Downloads 473

Authors: Elmira Aliyeva, Dadash Huseynov, Robert Hoogendoorn, Salomon Kroonenberg

Abstract:

The Kura River is the major water artery in the Southern Caucasus; it is a third river in the Caspian Sea basin in terms of length and size of the catchment area, the second in terms of the water budget, and the first in the volume of sediment load. Understanding of major controls on the Kura fluvial- deltaic system is valuable for efficient management of the highly populated river basin and coastal zone. We have studied grain size of sediments accumulated in the river channels and delta and dated by 210Pb method, astrophotographs, old topographic and geological maps, and archive data. At present time sediments are supplied by the Kura River to the Caspian Sea through three distributary channels oriented north-east, south-east, and south-west. The river is dominated by the suspended load - mud, silt, very fine sand. Coarse sediments are accumulated in the distributaries, levees, point bar, and delta front. The annual suspended sediment budget in the time period 1934-1952 before construction of the Mingechavir water reservoir in 1953 in the Kura River midstream area was 36 mln.t/yr. From 1953 to 1964, the suspended load has dropped to 12 mln.t/yr. After regulation of the Kura River discharge the volume of suspended load transported via north-eastern channel reduced from 35% of the total sediment amount to 4%, and through the main south-eastern channel increased from 65% to 96% with further fall to 56% due to creation of new south-western channel in 1964. Between 1967-1976 the annual sediment budget of the Kura River reached 22,5 mln. t/yr. From 1977 to 1986, the sediment load carried by the Kura River dropped to 17,6 mln.t/yr. The historical data show that between 1860 and 1907, during relatively stable Caspian Sea level two channels - N and SE, appear to have distributed an equal amount of sediments as seen from the bilateral geometry of the delta. In the time period 1907-1929, two new channels - E and NE, appeared. The growth of three delta lobes - N, NE, and SE, and rapid progradation of the delta has occurred on the background of the Caspian Sea level rise as a result of very high sediment supply. Since 1929 the Caspian Sea level decline was followed by the progradation of the delta occurring along the SE channel. The eastern and northern channels have been silted up. The slow rate of progradation at its initial stage was caused by the artificial reduction in the sediment budget. However, the continuous sea-level fall has brought to this river bed gradient increase, high erosional rate, increase in the sediment supply, and more rapid progradation. During the subsequent sea-level rise after 1977 accompanied by the decrease in the sediment budget, the southern part of the delta has turned into a complex of small, shallow channels oriented to the south. The data demonstrate that behaviour of the Kura fluvial – deltaic system and variations in the sediment budget besides anthropogenic regulation are strongly governed by the Caspian Sea level very rapid changes.

Keywords: anthropogenic control on sediment budget, Caspian sea-level variations, Kura river sediment load, morphology of the Kura river delta, sedimentation in the Kura river delta

Procedia PDF Downloads 153

Authors: Serhat Tüzün, Tufan Demirel

Abstract:

Decision Support Systems (DSS) have been investigated by researchers and technologists for more than 35 years. This paper analyses the developments in the architecture and software of these systems, provides a systematic analysis for different Web-based DSS approaches and Intelligent Decision-making Technologies (IDT), with the suggestion for future studies. Decision Support Systems literature begins with building model-oriented DSS in the late 1960s, theory developments in the 1970s, and the implementation of financial planning systems and Group DSS in the early and mid-80s. Then it documents the origins of Executive Information Systems, online analytic processing (OLAP) and Business Intelligence. The implementation of Web-based DSS occurred in the mid-1990s. With the beginning of the new millennia, intelligence is the main focus on DSS studies. Web-based technologies are having a major impact on design, development and implementation processes for all types of DSS. Web technologies are being utilized for the development of DSS tools by leading developers of decision support technologies. Major companies are encouraging its customers to port their DSS applications, such as data mining, customer relationship management (CRM) and OLAP systems, to a web-based environment. Similarly, real-time data fed from manufacturing plants are now helping floor managers make decisions regarding production adjustment to ensure that high-quality products are produced and delivered. Web-based DSS are being employed by organizations as decision aids for employees as well as customers. A common usage of Web-based DSS has been to assist customers configure product and service according to their needs. These systems allow individual customers to design their own products by choosing from a menu of attributes, components, prices and delivery options. The Intelligent Decision-making Technologies (IDT) domain is a fast growing area of research that integrates various aspects of computer science and information systems. This includes intelligent systems, intelligent technology, intelligent agents, artificial intelligence, fuzzy logic, neural networks, machine learning, knowledge discovery, computational intelligence, data science, big data analytics, inference engines, recommender systems or engines, and a variety of related disciplines. Innovative applications that emerge using IDT often have a significant impact on decision-making processes in government, industry, business, and academia in general. This is particularly pronounced in finance, accounting, healthcare, computer networks, real-time safety monitoring and crisis response systems. Similarly, IDT is commonly used in military decision-making systems, security, marketing, stock market prediction, and robotics. Even though lots of research studies have been conducted on Decision Support Systems, a systematic analysis on the subject is still missing. Because of this necessity, this paper has been prepared to search recent articles about the DSS. The literature has been deeply reviewed and by classifying previous studies according to their preferences, taxonomy for DSS has been prepared. With the aid of the taxonomic review and the recent developments over the subject, this study aims to analyze the future trends in decision support systems.

Keywords: decision support systems, intelligent decision-making, systematic analysis, taxonomic review

Procedia PDF Downloads 279

Authors: Sylvie Michel, Emmanuelle Gagnou, Joanne Hamet

Abstract:

Researchers in the fields of AI and Robotics frequently encounter ethical dilemmas throughout their research endeavors. Various ethical challenges have been pinpointed in the existing literature, including biases and discriminatory outcomes, diffusion of responsibility, and a deficit in transparency within AI operations. This research aims to pinpoint these ethical quandaries faced by researchers and shed light on the mechanisms behind ethical decision-making in the research process. By synthesizing insights from existing literature and acknowledging prevalent shortcomings, such as overlooking the heterogeneous nature of decision-making, non-accumulative results, and a lack of consensus on numerous factors due to limited empirical research, the objective is to conceptualize and validate a model. This model will incorporate influences from individual perspectives and situational contexts, considering potential moderating factors in the ethical decision-making process. Qualitative analyses were conducted based on direct observation of an AI/Robotics research team focusing on collaborative robotics for several months. Subsequently, semi-structured interviews with 16 team members were conducted. The entire process took place during the first semester of 2023. Observations were analyzed using an analysis grid, and the interviews underwent thematic analysis using Nvivo software. An initial finding involves identifying the ethical challenges that AI/robotics researchers confront, underlining a disparity between practical applications and theoretical considerations regarding ethical dilemmas in the realm of AI. Notably, researchers in AI prioritize the publication and recognition of their work, sparking the genesis of these ethical inquiries. Furthermore, this article illustrated that researchers tend to embrace a consequentialist ethical framework concerning safety (for humans engaging with robots/AI), worker autonomy in relation to robots, and the societal implications of labor (can robots displace jobs?). A second significant contribution entails proposing a model for ethical decision-making within the AI/Robotics research sphere. The model proposed adopts a process-oriented approach, delineating various research stages (topic proposal, hypothesis formulation, experimentation, conclusion, and valorization). Across these stages and the ethical queries, they entail, a comprehensive four-point comprehension of ethical decision-making is presented: recognition of the moral quandary; moral judgment, signifying the decision-maker's aptitude to discern the morally righteous course of action; moral intention, reflecting the ability to prioritize moral values above others; and moral behavior, denoting the application of moral intention to the situation. Variables such as political inclinations ((anti)-capitalism, environmentalism, veganism) seem to wield significant influence. Moreover, age emerges as a noteworthy moderating factor. AI and robotics researchers are continually confronted with ethical dilemmas during their research endeavors, necessitating thoughtful decision-making. The contribution involves introducing a contextually tailored model, derived from meticulous observations and insightful interviews, enabling the identification of factors that shape ethical decision-making at different stages of the research process.

Keywords: ethical decision making, artificial intelligence, robotics, research

Procedia PDF Downloads 79

Authors: Nodar Kekelidze, David Kekelidze, Elza Khutsishvili, Bela Kvirkvelia

Abstract:

The semiconductor devices are irreplaceable elements for investigations in Space (artificial Earth satellite, interplanetary space craft, probes, rockets) and for investigation of elementary particles on accelerators, for atomic power stations, nuclear reactors, robots operating on heavily radiation contaminated territories (Chernobyl, Fukushima). Unfortunately, the most important parameters of semiconductors dramatically worsen under irradiation. So creation of radiation-resistant semiconductor materials for opto and microelectronic devices is actual problem, as well as investigation of complicated processes developed in irradiated solid states. Homogeneous single crystals of InP-InAs solid solutions were grown with zone melting method. There has been studied the dependence of the optical absorption coefficient vs photon energy near fundamental absorption edge. This dependence changes dramatically with irradiation. The experiments were performed on InP, InAs and InP-InAs solid solutions before and after irradiation with electrons and fast neutrons. The investigations of optical properties were carried out on infrared spectrophotometer in temperature range of 10K-300K and 1mkm-50mkm spectral area. Radiation fluencies of fast neutrons was equal to 2·1018neutron/cm2 and electrons with 3MeV, 50MeV up to fluxes of 6·1017electron/cm2. Under irradiation, there has been revealed the exponential type of the dependence of the optical absorption coefficient vs photon energy with energy deficiency. The indicated phenomenon takes place at high and low temperatures as well at impurity different concentration and practically in all cases of irradiation by various energy electrons and fast neutrons. We have developed the common mechanism of this phenomenon for unirradiated materials and implemented the quantitative calculations of distinctive parameter; this is in a satisfactory agreement with experimental data. For the irradiated crystals picture get complicated. In the work, the corresponding analysis is carried out. It has been shown, that in the case of InP, irradiated with electrons (Ф=1·1017el/cm2), the curve of optical absorption is shifted to lower energies. This is caused by appearance of the tails of density of states in forbidden band due to local fluctuations of ionized impurity (defect) concentration. Situation is more complicated in the case of InAs and for solid solutions with composition near to InAs when besides noticeable phenomenon there takes place Burstein effect caused by increase of electrons concentration as a result of irradiation. We have shown, that in certain conditions it is possible the prevalence of Burstein effect. This causes the opposite effect: the shift of the optical absorption edge to higher energies. So in given solid solutions there take place two different opposite directed processes. By selection of solid solutions composition and doping impurity we obtained such InP-InAs, solid solution in which under radiation mutual compensation of optical absorption curves displacement occurs. Obtained result let create on the base of InP-InAs, solid solution radiation-resistant optical materials. Conclusion: It was established the nature of optical absorption near fundamental edge in semiconductor materials and it was created radiation-resistant optical material.

Keywords: InAs-InP, electrons concentration, irradiation, solid solutions

Procedia PDF Downloads 201

Authors: Alessandra Swiny, Yiorgos Hadjichristou

Abstract:

There exists an on-going geographical scar creating a division through the Island of Cyprus and its capital, Nicosia. The currently amputated city center is accessed legally by the United Nations convoys, infiltrated only by Turkish and Greek Cypriot army scouts and illegal traders and scavengers. On Christmas day 1963 in Nicosia, Captain M. Hobden of the British Army took a green chinagraph pencil and on a large scale Joint Army-RAF map ‘marked’ the division. From then on this ‘buffer zone’ was called the ‘green line.' This once dividing form, separating the main communities of Greek and Turkish Cypriots from one another, has now been fully reclaimed by an autonomous intruder. It's currently most captivating inhabitant is nature. She keeps taking over, for the past fifty years indigenous and introduced fauna and flora thrive; trees emerge from rooftops and plants, bushes and flowers grow randomly through the once bustling market streets, allowing this ‘no man’s land’ to teem with wildlife. And where are its limits? The idea of fluidity is ever present; it encroaches into the urban and built environment that surrounds it, and notions of ownership and permanence are questioned. Its qualities have contributed significantly in the search for new ‘identities,' expressed in the emergence of new living conditions, be they real or surreal. Without being physically reachable, it can be glimpsed at through punctured peepholes, military bunker windows that act as enticing portals into an emotional and conceptual level of inhabitation. The zone is mystical and simultaneously suspended in time, it triggers people’s imagination, not just that of the two prevailing communities but also of immigrants, refugees, and visitors; it mesmerizes all who come within its proximity. The paper opens a discussion on the issues and the binary questions raised. What is natural and artificial; what is private and public; what is ephemeral and permanent? The ‘green line’ exists in a central fringe condition and can serve in mixing generations and groups of people; mingling functions of living with work and social interaction; merging nature and the human being in a new-found synergy of human hope and survival, allowing thus for new notions of place to be introduced. Questions seek to be answered, such as, “Is the impossibility of dwelling made possible, by interweaving these ‘in-between conditions’ into eloquently traced spaces?” The methodologies pursued are developed through academic research, professional practice projects, and students’ research/design work. Realized projects, case studies and other examples cited both nationally and internationally hold global and local applications. Both paths of the research deal with the explorative understanding of the impossibility of dwelling, testing the limits of its autonomy. The expected outcome of the experience evokes in the user a sense of a new urban landscape, created from human topographies that echo the voice of an emerging identity.

Keywords: urban wildlife, human topographies, buffer zone, no man’s land

Procedia PDF Downloads 198

Authors: Iuri Salukvadze

Abstract:

Construction takes importance role in the history of mankind, there is not a single thing-product in our lives in which the builder’s work was not to be materialized, because to create all of it requires setting up factories, roads, and bridges, etc. The function of the Republic of Georgia, as part of the connecting Europe-Asia transport corridor, is significantly increased. In the context of transit function a large part of the cargo traffic belongs to motor transport, hence the improvement of motor roads transport infrastructure is rather important and rise the new, increased operational demands for existing as well as new motor roads. Construction of the durable road surface is related to rather large values, but because of high transport-operational properties, such as high-speed, less fuel consumption, less depreciation of tires, etc. If the traffic intensity is high, therefore the reimbursement of expenses occurs rapidly and accordingly is increasing income. If the traffic intensity is relatively small, it is recommended to use lightened structures of road carpet in order to pay for capital investments amounted to no more than normative one. The road carpet is divided into the following basic types: asphaltic concrete and cement concrete. Asphaltic concrete is the most perfect type of road carpet. It is arranged in two or three layers on rigid foundation and will be compacted. Asphaltic concrete is artificial building material, which due stratum will be selected and measured from stone skeleton and sand, interconnected by bitumen and a mixture of mineral powder. Less strictly selected similar material is called as bitumen-mineral mixture. Asphaltic concrete is non-rigid building material and well durable on vertical loadings; it is less resistant to the impact of horizontal forces. The cement concrete is monolithic and durable material, it is well durable the horizontal loads and is less resistant related to vertical loads. The cement concrete consists from strictly selected, measured stone material and sand, the binder is cement. The cement concrete road carpet represents separate slabs of sizes from 3 ÷ 5 op to 6 ÷ 8 meters. The slabs are reinforced by a rather complex system. Between the slabs are arranged seams that are designed for avoiding of additional stresses due temperature fluctuations on the length of slabs. For the joint behavior of separate slabs, they are connected by metal rods. Rods provide the changes in the length of slabs and distribute to the slab vertical forces and bending moments. The foundation layers will be extremely durable, for that is required high-quality stone material, cement, and metal. The qualification work aims to: in order for improvement of traffic conditions on motor roads to prolong operational conditions and improving their characteristics. The work consists from three chapters, 80 pages, 5 tables and 5 figures. In the work are stated general concepts as well as carried out by various companies using modern methods tests and their results. In the chapter III are stated carried by us tests related to this issue and specific examples to improving the operational characteristics.

Keywords: asphalt, cement, cylindrikal sample of asphalt, building

Procedia PDF Downloads 223

Authors: Rajkumar Ghosh

Abstract:

Water scarcity is a pressing issue in urban areas, even in smart cities where efficient resource management is a priority. This scarcity is mainly caused by factors such as lifestyle changes, excessive groundwater extraction, over-usage of water, rapid urbanization, and uncontrolled population growth. In the specific case of Gomti Nagar, Lucknow, Uttar Pradesh, India, the depletion of groundwater resources is particularly severe, leading to a water imbalance and posing a significant challenge for the region's sustainable development. The aim of this study is to address the water shortage in the Gomti Nagar region by focusing on the implementation of artificial groundwater recharge methods. Specifically, the research aims to investigate the effectiveness of rainwater collection through rooftop rainwater harvesting systems (RTRWHs) as a sustainable approach to reduce aquifer depletion and bridge the gap between groundwater recharge and extraction. The research methodology for this study involves the utilization of RTRWHs as the main method for collecting rainwater. This approach is considered effective in managing and conserving water resources in a sustainable manner. The focus is on implementing RTRWHs in residential and commercial buildings to maximize the collection of rainwater and its subsequent utilization for various purposes in the Gomti Nagar region. The study reveals that the installation of RTRWHs in the Gomti Nagar region has a positive impact on addressing the water scarcity issue. Currently, RTRWHs cover only a small percentage (0.04%) of the total rainfall collected in the region. However, when RTRWHs are installed in all buildings, their influence on increasing water availability and reducing aquifer depletion will be significantly greater. The study also highlights the significant water imbalance of 24519 ML/yr in the region, emphasizing the urgent need for sustainable water management practices. This research contributes to the theoretical understanding of sustainable water management systems in smart cities. By highlighting the effectiveness of RTRWHs in reducing aquifer depletion, it emphasizes the importance of implementing such systems in urban areas. The findings of this study can serve as a basis for policymakers, urban planners, and developers to prioritize and incentivize the installation of RTRWHs as a potential solution to the water shortage crisis. The data for this study were collected through various sources such as government reports, surveys, and existing groundwater abstraction patterns. The collected data were then analysed to assess the current water situation, groundwater depletion rate, and the potential impact of implementing RTRWHs. Statistical analysis and modelling techniques were employed to quantify the water imbalance and evaluate the effectiveness of RTRWHs. The findings of this study demonstrate that the implementation of RTRWHs can effectively mitigate the water scarcity crisis in Gomti Nagar. By reducing aquifer depletion and bridging the gap between groundwater recharge and extraction, RTRWHs offer a sustainable solution to the region's water scarcity challenges. The study highlights the need for widespread adoption of RTRWHs in all buildings and emphasizes the importance of integrating such systems into the urban planning and development process. By doing so, smart cities like Gomti Nagar can achieve efficient water management, ensuring a better future with improved water availability for its residents.

Keywords: rooftop rainwater harvesting, rainwater, water management, aquifer

Procedia PDF Downloads 95

Authors: Babak Forouraghi

Abstract:

A genetic circuit is a collection of interacting genes and proteins that enable individual cells to implement and perform vital biological functions such as cell division, growth, death, and signaling. In cell engineering, synthetic gene circuits are engineered networks of genes specifically designed to implement functionalities that are not evolved by nature. These engineered networks enable scientists to tackle complex problems such as engineering cells to produce therapeutics within the patient's body, altering T cells to target cancer-related antigens for treatment, improving antibody production using engineered cells, tissue engineering, and production of genetically modified plants and livestock. Construction of computational models to realize genetic circuits is an especially challenging task since it requires the discovery of flow of genetic information in complex biological systems. Building synthetic biological models is also a time-consuming process with relatively low prediction accuracy for highly complex genetic circuits. The primary goal of this study was to investigate the utility of a pre-trained bidirectional encoder transformer that can accurately predict gene expressions in genetic circuit designs. The main reason behind using transformers is their innate ability (attention mechanism) to take account of the semantic context present in long DNA chains that are heavily dependent on spatial representation of their constituent genes. Previous approaches to gene circuit design, such as CNN and RNN architectures, are unable to capture semantic dependencies in long contexts as required in most real-world applications of synthetic biology. For instance, RNN models (LSTM, GRU), although able to learn long-term dependencies, greatly suffer from vanishing gradient and low-efficiency problem when they sequentially process past states and compresses contextual information into a bottleneck with long input sequences. In other words, these architectures are not equipped with the necessary attention mechanisms to follow a long chain of genes with thousands of tokens. To address the above-mentioned limitations of previous approaches, a transformer model was built in this work as a variation to the existing DNA Bidirectional Encoder Representations from Transformers (DNABERT) model. It is shown that the proposed transformer is capable of capturing contextual information from long input sequences with attention mechanism. In a previous work on genetic circuit design, the traditional approaches to classification and regression, such as Random Forrest, Support Vector Machine, and Artificial Neural Networks, were able to achieve reasonably high R2 accuracy levels of 0.95 to 0.97. However, the transformer model utilized in this work with its attention-based mechanism, was able to achieve a perfect accuracy level of 100%. Further, it is demonstrated that the efficiency of the transformer-based gene expression classifier is not dependent on presence of large amounts of training examples, which may be difficult to compile in many real-world gene circuit designs.

Keywords: transformers, generative ai, gene expression design, classification

Procedia PDF Downloads 59

Authors: Md. Abu Sufian, Robiqul Islam, Imam Hossain Shajid, Mahesh Hanumanthu, Jarasree Varadarajan, Md. Sipon Miah, Mingbo Niu

Abstract:

Coronary Heart Disease (CHD) remains a principal cause of global morbidity and mortality, characterized by atherosclerosis—the build-up of fatty deposits inside the arteries. The study introduces an innovative methodology that leverages cloud-based platforms like AWS Live Streaming and Artificial Intelligence (AI) to early detect and prevent CHD symptoms in web applications. By employing novel simulation processes and AI algorithms, this research aims to significantly mitigate the health and societal impacts of CHD. Methodology: This study introduces a novel simulation process alongside a multi-phased model development strategy. Initially, health-related data, including heart rate variability, blood pressure, lipid profiles, and ECG readings, were collected through user interactions with web-based applications as well as API Integration. The novel simulation process involved creating synthetic datasets that mimic early-stage CHD symptoms, allowing for the refinement and training of AI algorithms under controlled conditions without compromising patient privacy. AWS Live Streaming was utilized to capture real-time health data, which was then processed and analysed using advanced AI techniques. The novel aspect of our methodology lies in the simulation of CHD symptom progression, which provides a dynamic training environment for our AI models enhancing their predictive accuracy and robustness. Model Development: it developed a machine learning model trained on both real and simulated datasets. Incorporating a variety of algorithms including neural networks and ensemble learning model to identify early signs of CHD. The model's continuous learning mechanism allows it to evolve adapting to new data inputs and improving its predictive performance over time. Results and Findings: The deployment of our model yielded promising results. In the validation phase, it achieved an accuracy of 92% in predicting early CHD symptoms surpassing existing models. The precision and recall metrics stood at 89% and 91% respectively, indicating a high level of reliability in identifying at-risk individuals. These results underscore the effectiveness of combining live data streaming with AI in the early detection of CHD. Societal Implications: The implementation of cloud-based AI for CHD symptom detection represents a significant step forward in preventive healthcare. By facilitating early intervention, this approach has the potential to reduce the incidence of CHD-related complications, decrease healthcare costs, and improve patient outcomes. Moreover, the accessibility and scalability of cloud-based solutions democratize advanced health monitoring, making it available to a broader population. This study illustrates the transformative potential of integrating technology and healthcare, setting a new standard for the early detection and management of chronic diseases.

Keywords: coronary heart disease, cloud-based ai, machine learning, novel simulation techniques, early detection, preventive healthcare

Procedia PDF Downloads 64

Authors: Beatriz González-rodrigo, Diego Hidalgo-leiva, Omar Flores, Claudia Germoso, Maribel Jiménez-martínez, Laura Navas-sánchez, Belén Orta, Nicola Tarque, Orlando Hernández- Rubio, Miguel Marchamalo, Juan Gregorio Rejas, Belén Benito-oterino

Abstract:

Cities facing seismic hazard necessitate detailed risk assessments for effective urban planning and vulnerability identification, ensuring the safety and sustainability of urban infrastructure. Comprehensive studies involving seismic hazard, vulnerability, and exposure evaluations are pivotal for estimating potential losses and guiding proactive measures against seismic events. However, broad-scale traditional risk studies limit consideration of specific local threats and identify vulnerable housing within a structural typology. Achieving precise results at neighbourhood levels demands higher resolution seismic hazard exposure, and vulnerability studies. This research aims to bolster sustainability and safety against seismic disasters in three Central American and Caribbean capitals. It integrates geospatial techniques and artificial intelligence into seismic risk studies, proposing cost-effective methods for exposure data collection and damage prediction. The methodology relies on prior seismic threat studies in pilot zones, utilizing existing exposure and vulnerability data in the region. Emphasizing detailed building attributes enables the consideration of behaviour modifiers affecting seismic response. The approach aims to generate detailed risk scenarios, facilitating prioritization of preventive actions pre-, during, and post-seismic events, enhancing decision-making certainty. Detailed risk scenarios necessitate substantial investment in fieldwork, training, research, and methodology development. Regional cooperation becomes crucial given similar seismic threats, urban planning, and construction systems among involved countries. The outcomes hold significance for emergency planning and national and regional construction regulations. The success of this methodology depends on cooperation, investment, and innovative approaches, offering insights and lessons applicable to regions facing moderate seismic threats with vulnerable constructions. Thus, this framework aims to fortify resilience in seismic-prone areas and serves as a reference for global urban planning and disaster management strategies. In conclusion, this research proposes a comprehensive framework for seismic risk assessment in high-risk urban areas, emphasizing detailed studies at finer resolutions for precise vulnerability evaluations. The approach integrates regional cooperation, geospatial technologies, and adaptive fragility curve adjustments to enhance risk assessment accuracy, guiding effective mitigation strategies and emergency management plans.

Keywords: assessment, behaviour modifiers, emergency management, mitigation strategies, resilience, vulnerability

Procedia PDF Downloads 67

Authors: Nurkhairul Bariyah Baharun, Afzan Adam, Reena Rahayu Md Zin

Abstract:

Tumor microenvironment (TME) in breast cancer is mainly composed of cancer cells, immune cells, and stromal cells. The interaction between cancer cells and their microenvironment plays an important role in tumor development, progression, and treatment response. The TME in breast cancer includes tumor-infiltrating lymphocytes (TILs) that are implicated in killing tumor cells. TILs can be found in tumor stroma (sTILs) and within the tumor (iTILs). TILs in triple negative breast cancer (TNBC) have been demonstrated to have prognostic and potentially predictive value. The international Immune-Oncology Biomarker Working Group (TIL-WG) had developed a guideline focus on the assessment of sTILs using hematoxylin and eosin (H&E)-stained slides. According to the guideline, the pathologists use “eye balling” method on the H&E stained- slide for sTILs assessment. This method has low precision, poor interobserver reproducibility, and is time-consuming for a comprehensive evaluation, besides only counted sTILs in their assessment. The TIL-WG has therefore recommended that any algorithm for computational assessment of TILs utilizing the guidelines provided to overcome the limitations of manual assessment, thus providing highly accurate and reliable TILs detection and classification for reproducible and quantitative measurement. This study is carried out to develop a TNBC digital whole slide image (WSI) dataset from H&E-stained slides and IHC (CD4+ and CD8+) stained slides. TNBC cases were retrieved from the database of the Department of Pathology, Hospital Canselor Tuanku Muhriz (HCTM). TNBC cases diagnosed between the year 2010 and 2021 with no history of other cancer and available block tissue were included in the study (n=58). Tissue blocks were sectioned approximately 4 µm for H&E and IHC stain. The H&E staining was performed according to a well-established protocol. Indirect IHC stain was also performed on the tissue sections using protocol from Diagnostic BioSystems PolyVue™ Plus Kit, USA. The slides were stained with rabbit monoclonal, CD8 antibody (SP16) and Rabbit monoclonal, CD4 antibody (EP204). The selected and quality-checked slides were then scanned using a high-resolution whole slide scanner (Pannoramic DESK II DW- slide scanner) to digitalize the tissue image with a pixel resolution of 20x magnification. A manual TILs (sTILs and iTILs) assessment was then carried out by the appointed pathologist (2 pathologists) for manual TILs scoring from the digital WSIs following the guideline developed by TIL-WG 2014, and the result displayed as the percentage of sTILs and iTILs per mm² stromal and tumour area on the tissue. Following this, we aimed to develop an automated digital image scoring framework that incorporates key elements of manual guidelines (including both sTILs and iTILs) using manually annotated data for robust and objective quantification of TILs in TNBC. From the study, we have developed a digital dataset of TNBC H&E and IHC (CD4+ and CD8+) stained slides. We hope that an automated based scoring method can provide quantitative and interpretable TILs scoring, which correlates with the manual pathologist-derived sTILs and iTILs scoring and thus has potential prognostic implications.

Keywords: automated quantification, digital pathology, triple negative breast cancer, tumour infiltrating lymphocytes

Procedia PDF Downloads 116

Authors: Nahla Elsherbiny, Ahmed Ahmed, Hamada Mohammed, Mohamed Ali

Abstract:

Background: The enterococci may be considered as opportunistic agents particularly in immunocompromised patients. It is one of the top three pathogens causing many healthcare associated infections (HAIs). Resistance to several commonly used antimicrobial agents is a remarkable characteristic of most species which may carry various genes contributing to virulence. Objectives: to determine the prevalence of enterococci species in different intensive care units (ICUs) causing health care-associated infections (HAIs), intestinal carriage and environmental contamination. Also, to study the antimicrobial susceptibility pattern of the isolates with special reference to vancomycin resistance. In addition to phenotypic and genotypic detection of gelatinase, cytolysin and biofilm formation among isolates. Patients and Methods: This study was carried out in the infection control laboratory at Assiut University Hospitals over a period of one year. Clinical samples were collected from 285 patients with various (HAIs) acquired after admission to different ICUs. Rectal swabs were taken from 14 cases for detection of enterococci carriage. In addition, 1377 environmental samples were collected from the surroundings of the patients. Identification was done by conventional bacteriological methods and confirmed by analytical profile index (API). Antimicrobial sensitivity testing was performed by Kirby Bauer disc diffusion method and detection of vancomycin resistance was done by agar screen method. For the isolates, phenotypic detection of cytolysin, gelatinase production and detection of biofilm by tube method, Congo red method and microtiter plate. We performed polymerase chain reaction (PCR) for detection of some virulence genes (gelE, cylA, vanA, vanB and esp). Results: Enterococci caused 10.5% of the HAIs. Respiratory tract infection was the predominant type (86.7%). The commonest species were E.gallinarum (36.7%), E.casseliflavus (30%), E.faecalis (30%), and E.durans (3.4 %). Vancomycin resistance was detected in a total of 40% (12/30) of those isolates. The risk factors associated with acquiring vancomycin resistant enterococci (VRE) were immune suppression (P= 0.031) and artificial feeding (P= 0.008). For the rectal swabs, enterococci species were detected in 71.4% of samples with the predominance of E. casseliflavus (50%). Most of the isolates were vancomycin resistant (70%). Out of a total 1377 environmental samples, 577 (42%) samples were contaminated with different microorganisms. Enterococci were detected in 1.7% (10/577) of total contaminated samples, 50% of which were vancomycin resistant. All isolates were resistant to penicillin, ampicillin, oxacillin, ciprofloxacin, amikacin, erythromycin, clindamycin and trimethoprim-sulfamethaxazole. For the remaining antibiotics, variable percentages of resistance were reported. Cytolysin and gelatinase were detected phenotypically in 16% and 48 % of the isolates respectively. The microtiter plate method showed the highest percentages of detection of biofilm among all isolated species (100%). The studied virulence genes gelE, esp, vanA and vanB were detected in 62%, 12%, 2% and 12% respectively, while cylA gene was not detected in any isolates. Conclusions: A significant percentage of enterococci was isolated from patients and environments in the ICUs. Many virulence factors were detected phenotypically and genotypically among isolates. The high percentage of resistance, coupled with the risk of cross transmission to other patients make enterococci infections a significant infection control issue in hospitals.

Keywords: antimicrobial resistance, enterococci, ICUs, virulence factors

Procedia PDF Downloads 285

Authors: Byron Baron

Abstract:

Typical of the Mediterranean region, the Maltese islands exhibit calcareous soils containing low organic carbon content and high salinity, in addition to being relatively shallow. This has lead to the common practice of applying copious amounts of artificial fertilisers as well as other chemical inputs, together with the use of ground water having high salinity. Such intensive agricultural activities, over a prolonged time period, on such land has lead further to the loss of any soil fertility, together with direct negative impacts on the quality of fresh water reserves and the local ecosystem. The aim of this study was to investigate whether the natural resources on the island would be sufficient to apply ecological intensification i.e. the use of natural processes to replace anthropological inputs without any significant loss in food production. This was implementing through a sustainable agricultural system based on permaculture practices. Ecological intensification following permaculture principles was implemented for two years in order to capture the seasonal changes in duplicate. The areas dedicated to wild plants were only trimmed back to avoid excessive seeding but never mowing. A number of local staple crops were grown throughout this period, also in duplicate. Concomitantly, a number of practices were implemented following permaculture principles such as reducing land tilling, applying only natural fertiliser, mulching, monitoring of soil parameters using sensors, no use of herbicides or pesticides, and precision irrigation linked to a desalination system. Numerous environmental parameters were measured at regular intervals so as to quantify any improvements in ecological conditions. Crop output was also measured as kilos of produce per area. The results clearly show that over the two year period, the variety of wild plant species increased, the number of visiting pollinators increased, there were no pest infestations (although an increase in the number of pests was observed), and a slight improvement in overall soil health was also observed. This was obviously limited by the short duration of the testing implementation. Dedicating slightly less than 15% of total land area to wild plants in the form of borders around plots of crops assisted pollination and provided a foraging area for gleaning bats (measured as an increased number of feeding buzzes) whilst not giving rise to any pest infestations and no apparent yield losses or ill effects to the crops. Observed increases in crop yields were not significant. The study concluded that with the right support for the initial establishment of a healthy ecosystem and controlled intervention, the available natural resources on the island can substantially improve the condition of the local agricultural land area, resulting is a more prolonged economical output with greater ecological sustainability. That being said, more comprehensive and long-term monitoring is required in order to fully validate these results and design a sustainable agriculture system that truly achieves the best outcome for the Maltese context.

Keywords: ecological intensification, soil health, sustainable agriculture, permaculture

Procedia PDF Downloads 65

Authors: Ashlie Latiolais

Abstract:

Architecture, as a discipline, is on a trajectory of extension beyond the boundaries of buildings and, more increasingly, is coupled with research that connects to alternative and typically disjointed disciplines. A “both/and” approach and (expanded) definition of architecture, as depicted here, expands the margins that contain the profession. Figuratively, architecture is a series of edges, events, and occurrences that establishes a choreography or stage by which humanity exists. The way in which architecture controls and suggests the movement through these spaces, being within a landscape, city, or building, can be viewed as a datum by which the “dance” of everyday life occurs. This submission views the realm of architecture through the lens of movement and dance as a cross-fertilizer of collaboration, tectonic, and spatial geometry investigations. “Designing on digital programs puts architects at a distance from the spaces they imagine. While this has obvious advantages, it also means that they lose the lived, embodied experience of feeling what is needed in space—meaning that some design ideas that work in theory ultimately fail in practice.” By studying the body in motion through real-time performance, a more holistic understanding of architectural space surfaces and new prospects for theoretical teaching pedagogies emerge. The atypical intersection rethinks how architecture is considered, created, and tested, similar to how “dance artists often do this by thinking through the body, opening pathways and possibilities that might not otherwise be accessible” –this is the essence of this poster submission as explained through unFOLDED, a creative performance work. A new languageismaterialized through unFOLDED, a dynamic occupiable installation by which architecture is investigated through dance, movement, and body analysis. The entry unfolds a collaboration of an architect, dance choreographer, musicians, video artist, and lighting designers to re-create one of the first documented avant-garde performing arts collaborations (Matisse, Satie, Massine, Picasso) from the Ballet Russes in 1917, entitled Parade. Architecturally, this interdisciplinary project orients and suggests motion through structure, tectonic, lightness, darkness, and shadow as it questions the navigation of the dark space (stage) surrounding the installation. Artificial light via theatrical lighting and video graphics brought the blank canvas to life – where the sensitive mix of musicality coordinated with the structure’s movement sequencing was certainly a challenge. The upstage light from the video projections created both flickered contextual imagery and shadowed figures. When the dancers were either upstage or downstage of the structure, both silhouetted figures and revealed bodies are experienced as dancer-controlled installation manipulations occurred throughout the performance. The experimental performance, through structure, prompted moving (dancing) bodies in space, where the architecture served as a key component to the choreography itself. The tectonic of the delicate steel structure allowed for the dancers to interact with the installation, which created a variety of spatial conditions – the contained box of three-dimensional space, to a wall, and various abstracted geometries in between. The development of this research unveils the new role of an Architect as a Choreographer of the built environment.

Keywords: dance, architecture, choreography, installation, architect, choreographer, space

Procedia PDF Downloads 91

Authors: Erica Cochran Hameen, Bobuchi Ken-Opurum, Mounica Guturu

Abstract:

The United Stated (U.S.) populous spends the majority of their time indoors in spaces where building codes and voluntary sustainability standards provide clear Indoor Environmental Quality (IEQ) metrics. The existing sustainable building standards and codes are aimed towards improving IEQ, health of occupants, and reducing the negative impacts of buildings on the environment. While they address the post-occupancy stage of buildings, there are fewer standards on the pre-occupancy stage thereby placing a large labor population in environments much less regulated. Construction personnel are often exposed to a variety of uncomfortable and unhealthy elements while on construction sites, primarily thermal, visual, acoustic, and air quality related. Construction site power generators, equipment, and machinery generate on average 9 decibels (dBA) above the U.S. OSHA regulations, creating uncomfortable noise levels. Research has shown that frequent exposure to high noise levels leads to chronic physiological issues and increases noise induced stress, yet beyond OSHA no other metric focuses directly on the impacts of noise on contractors’ well-being. Research has also associated natural light with higher productivity and attention span, and lower cases of fatigue in construction workers. However, daylight is not always available as construction workers often perform tasks in cramped spaces, dark areas, or at nighttime. In these instances, the use of artificial light is necessary, yet lighting standards for use during lengthy tasks and arduous activities is not specified. Additionally, ambient air, contaminants, and material off-gassing expelled at construction sites are one of the causes of serious health effects in construction workers. Coupled with extreme hot and cold temperatures for different climate zones, health and productivity can be seriously compromised. This research evaluates the impact of existing green building metrics on construction and risk management, by analyzing two codes and nine standards including LEED, WELL, and BREAM. These metrics were chosen based on the relevance to the U.S. construction industry. This research determined that less than 20% of the sustainability context within the standards and codes (texts) are related to the pre-occupancy building sector. The research also investigated the impact of construction personnel’s health and well-being on construction management through two surveys of project managers and on-site contractors’ perception of their work environment on productivity. To fully understand the risks of limited Environmental and Health Quality metrics for contractors (EHQ) this research evaluated the connection between EHQ factors such as inefficient lighting, on construction workers and investigated the correlation between various site coping strategies for comfort and productivity. Outcomes from this research are three-pronged. The first includes fostering a discussion about the existing conditions of EQH elements, i.e. thermal, lighting, ergonomic, acoustic, and air quality on the construction labor force. The second identifies gaps in sustainability standards and codes during the pre-occupancy stage of building construction from ground-breaking to substantial completion. The third identifies opportunities for improvements and mitigation strategies to improve EQH such as increased monitoring of effects on productivity and health of contractors and increased inclusion of the pre-occupancy stage in green building standards.

Keywords: construction contractors, health and well-being, environmental quality, risk management

Procedia PDF Downloads 132

Authors: Rebecca Laycock

Abstract:

Many Urban Agriculture (UA) initiatives, including community/allotment gardens, Community Supported Agriculture, and community/social farms, depend on volunteers. However, initiatives supported or run by volunteers are often faced with a high turnover of labour as a result of the involvement of episodic volunteers (a term describing ad hoc, one-time, and seasonal volunteers), leading to challenges with maintaining project continuity and retaining skills/knowledge within the initiative. This is a notable challenge given that food growing is a knowledge intensive activity where the fruits of labour appear months or sometimes years after investment. Participatory Action Research (PAR) is increasingly advocated for in the field of UA as a solution-oriented approach to research, providing concrete results in addition to advancing theory. PAR is a cyclical methodological approach involving researchers and stakeholders collaboratively 'identifying' and 'theorising' an issue, 'planning' an action to address said issue, 'taking action', and 'reflecting' on the process. Through iterative cycles and prolonged engagement, the theory is developed and actions become better tailored to the issue. The demand for PAR in UA research means that understanding how to use PAR with episodic volunteers is of critical importance. The aim of this paper is to explore (1) the challenges of doing PAR in UA initiatives with episodic volunteers, and (2) how PAR can be harnessed to advance sustainable development of UA through theoretically-informed action. A 2.5 year qualitative PAR study on three English case study student-led food growing initiatives took place between 2014 and 2016. University UA initiatives were chosen as exemplars because most of their volunteers were episodic. Data were collected through 13 interviews, 6 workshops, and a research diary. The results were thematically analysed through eclectic coding using Computer-Assisted Qualitative Data Analysis Software (NVivo). It was found that the challenges of doing PAR with transient participants were (1) a superficial understanding of issues by volunteers because of short term engagement, resulting in difficulties ‘identifying’/‘theorising’ issues to research; (2) difficulties implementing ‘actions’ given those involved in the ‘planning’ phase often left by the ‘action’ phase; (3) a lack of capacity of participants to engage in research given the ongoing challenge of maintaining participation; and (4) that the introduction of the researcher acted as an ‘intervention’. The involvement of a long-term stakeholder (the researcher) changed the group dynamics, prompted critical reflections that had not previously taken place, and improved continuity. This posed challenges for providing a genuine understanding the episodic volunteering PAR initiatives, and also challenged the notion of what constitutes an ‘intervention’ or ‘action’ in PAR. It is recommended that researchers working with episodic volunteers using PAR should (1) adopt a first-person approach by inquiring into the researcher’s own experience to enable depth in theoretical analysis to manage the potentially superficial understandings by short-term participants; and (2) establish safety mechanisms to address the potential for the research to impose artificial project continuity and knowledge retention that will end when the research does. Through these means, we can more effectively use PAR to conduct solution-oriented research about UA.

Keywords: community garden, continuity, first-person research, higher education, knowledge retention, project management, transience, university

Procedia PDF Downloads 248

Authors: Babak Forouraghi

Abstract:

A genetic circuit is a collection of interacting genes and proteins that enable individual cells to implement and perform vital biological functions such as cell division, growth, death, and signaling. In cell engineering, synthetic gene circuits are engineered networks of genes specifically designed to implement functionalities that are not evolved by nature. These engineered networks enable scientists to tackle complex problems such as engineering cells to produce therapeutics within the patient's body, altering T cells to target cancer-related antigens for treatment, improving antibody production using engineered cells, tissue engineering, and production of genetically modified plants and livestock. Construction of computational models to realize genetic circuits is an especially challenging task since it requires the discovery of the flow of genetic information in complex biological systems. Building synthetic biological models is also a time-consuming process with relatively low prediction accuracy for highly complex genetic circuits. The primary goal of this study was to investigate the utility of a pre-trained bidirectional encoder transformer that can accurately predict gene expressions in genetic circuit designs. The main reason behind using transformers is their innate ability (attention mechanism) to take account of the semantic context present in long DNA chains that are heavily dependent on the spatial representation of their constituent genes. Previous approaches to gene circuit design, such as CNN and RNN architectures, are unable to capture semantic dependencies in long contexts, as required in most real-world applications of synthetic biology. For instance, RNN models (LSTM, GRU), although able to learn long-term dependencies, greatly suffer from vanishing gradient and low-efficiency problem when they sequentially process past states and compresses contextual information into a bottleneck with long input sequences. In other words, these architectures are not equipped with the necessary attention mechanisms to follow a long chain of genes with thousands of tokens. To address the above-mentioned limitations, a transformer model was built in this work as a variation to the existing DNA Bidirectional Encoder Representations from Transformers (DNABERT) model. It is shown that the proposed transformer is capable of capturing contextual information from long input sequences with an attention mechanism. In previous works on genetic circuit design, the traditional approaches to classification and regression, such as Random Forrest, Support Vector Machine, and Artificial Neural Networks, were able to achieve reasonably high R2 accuracy levels of 0.95 to 0.97. However, the transformer model utilized in this work, with its attention-based mechanism, was able to achieve a perfect accuracy level of 100%. Further, it is demonstrated that the efficiency of the transformer-based gene expression classifier is not dependent on the presence of large amounts of training examples, which may be difficult to compile in many real-world gene circuit designs.

Keywords: machine learning, classification and regression, gene circuit design, bidirectional transformers

Procedia PDF Downloads 60

Authors: Adriana Haulica

Abstract:

Powered by Machine Learning, Precise medicine is tailored by now to use genetic and molecular profiling, with the aim of optimizing the therapeutic benefits for cohorts of patients. As the majority of Machine Language algorithms come from heuristics, the outputs have contextual validity. This is not very restrictive in the sense that medicine itself is not an exact science. Meanwhile, the progress made in Molecular Biology, Bioinformatics, Computational Biology, and Precise Medicine, correlated with the huge amount of human biology data and the increase in computational power, opens new healthcare challenges. A more accurate diagnosis is needed along with real-time treatments by processing as much as possible from the available information. The purpose of this paper is to present a deeper vision for the future of Artificial Intelligence in Precise medicine. In fact, actual Machine Learning algorithms use standard mathematical knowledge, mostly Euclidian metrics and standard computation rules. The loss of information arising from the classical methods prevents obtaining 100% evidence on the diagnosis process. To overcome these problems, we introduce MEDICOMPILLS, a new architectural concept tool of information processing in Precise medicine that delivers diagnosis and therapy advice. This tool processes poly-field digital resources: global knowledge related to biomedicine in a direct or indirect manner but also technical databases, Natural Language Processing algorithms, and strong class optimization functions. As the name suggests, the heart of this tool is a compiler. The approach is completely new, tailored for omics and clinical data. Firstly, the intrinsic biological intuition is different from the well-known “a needle in a haystack” approach usually used when Machine Learning algorithms have to process differential genomic or molecular data to find biomarkers. Also, even if the input is seized from various types of data, the working engine inside the MEDICOMPILLS does not search for patterns as an integrative tool. This approach deciphers the biological meaning of input data up to the metabolic and physiologic mechanisms, based on a compiler with grammars issued from bio-algebra-inspired mathematics. It translates input data into bio-semantic units with the help of contextual information iteratively until Bio-Logical operations can be performed on the base of the “common denominator “rule. The rigorousness of MEDICOMPILLS comes from the structure of the contextual information on functions, built to be analogous to mathematical “proofs”. The major impact of this architecture is expressed by the high accuracy of the diagnosis. Detected as a multiple conditions diagnostic, constituted by some main diseases along with unhealthy biological states, this format is highly suitable for therapy proposal and disease prevention. The use of MEDICOMPILLS architecture is highly beneficial for the healthcare industry. The expectation is to generate a strategic trend in Precise medicine, making medicine more like an exact science and reducing the considerable risk of errors in diagnostics and therapies. The tool can be used by pharmaceutical laboratories for the discovery of new cures. It will also contribute to better design of clinical trials and speed them up.

Keywords: bio-semantic units, multiple conditions diagnosis, NLP, omics

Procedia PDF Downloads 70