Search results for: adaptive neural controller

Authors: Sajeeha Ansar, Asad Ali Safi, Sheikh Ziauddin, Ahmad R. Shahid, Faraz Ahsan

Abstract:

Gliomas are most challenging and aggressive type of tumors which appear in different sizes, locations, and scattered boundaries. CNN is most efficient deep learning approach with outstanding capability of solving image analysis problems. A fully automatic deep learning based 2D-CNN model for brain tumor segmentation is presented in this paper. We used small convolution filters (3 x 3) to make architecture deeper. We increased convolutional layers for efficient learning of complex features from large dataset. We achieved better results by pushing convolutional layers up to 16 layers for HGG model. We achieved reliable and accurate results through fine-tuning among dataset and hyper-parameters. Pre-processing of this model includes generation of brain pipeline, intensity normalization, bias correction and data augmentation. We used the BRATS-2015, and Dice Similarity Coefficient (DSC) is used as performance measure for the evaluation of the proposed method. Our method achieved DSC score of 0.81 for complete, 0.79 for core, 0.80 for enhanced tumor regions. However, these results are comparable with methods already implemented 2D CNN architecture.

Keywords: brain tumor segmentation, convolutional neural networks, deep learning, HGG

Procedia PDF Downloads 218

Authors: Ailing Yang, Penghan Song, Aihua Cai

Abstract:

The nonlinear maneuvering target tracking problem is mainly a state estimation problem when the target motion model is uncertain. Traditional solutions include Kalman filtering based on Bayesian filtering framework and extended Kalman filtering. However, these methods need prior knowledge such as kinematics model and state system distribution, and their performance is poor in state estimation of nonprior complex dynamic systems. Therefore, in view of the problems existing in traditional algorithms, a convolution LSTM target state estimation (SAConvLSTM-SE) algorithm based on Self-Attention memory (SAM) is proposed to learn the historical motion state of the target and the error distribution information measured at the current time. The measured track point data of airborne radar are processed into data sets. After supervised training, the data-driven deep neural network based on SAConvLSTM can directly obtain the target state at the next moment. Through experiments on two different maneuvering targets, we find that the network has stronger robustness and better tracking accuracy than the existing tracking methods.

Keywords: maneuvering target, state estimation, Kalman filter, LSTM, self-attention

Procedia PDF Downloads 102

Authors: K. C. Ching

Abstract:

There is growing public concern over the increasing prevalence of student suicide in Hong Kong. Some identify the problem with insufficient social support, while some attribute it to the vast fluctuations in emotional experience and the hindrances to emotion-regulation, both typical of adolescence and emerging adulthood. This study is thus designed to explore the respective effect of perceived social support and emotion self-confidence, on positive emotions and negative emotions. Fifty-seven Hong Kong higher-education students (17 males, 40 females) aged between 18 and 25 (M = 21.78) responded to an online questionnaire consisted of self-reported measures of perceived social support, emotional self-confidence, positive emotions, and negative emotions. Hierarchical regression analysis revealed that emotional self-confidence positively associated with positive emotions and negatively with negative emotions, while perceived social support positively associated with positive emotions but was not related to negative emotions. Perceived social support and emotional self-confidence both predicted positive emotions, but did not interact to predict any emotional outcome. It is concluded that students’ positive and negative emotional experiences are closely related to their emotion-regulation process. But for social support, its effect is merely protective, meaning that although perceived social support generally promotes positive emotions, it alone does not suffice to alleviate students’ negative emotions. These conclusions carry profound implications to suicide prevention practices, including that most existing suicide prevention campaigns should advance from merely fostering mutual support to directly promoting adaptive coping of emotional negativity.

Keywords: emerging adulthood, emotional self-confidence, hong kong, perceived social support, suicide prevention

Procedia PDF Downloads 95

Authors: Albert Alexander Stonier

Abstract:

Due to the fossil fuel exhaustion and environmental pollution, renewable energy sources especially solar photovoltaic system plays a predominant role in providing energy to the consumers. It has been estimated that by 2050 the renewable energy sources will satisfy 50% of the total energy requirement of the world. In this context, the faults in the conversion process require a special attention which is considered as a major problem. A fault which remains even for a few seconds will cause undesirable effects to the system. The presentation comprises of the analysis, causes, effects and mitigation methods of various faults occurring in the entire solar photovoltaic energy conversion process. In order to overcome the faults in the system, an intelligent based artificial neural networks and fuzzy logic are proposed which can significantly mitigate the faults. Hence the presentation intends to find the problem in renewable energy and provides the possible solution to overcome it with simulation and experimental results. The work performed in a 3kWp solar photovoltaic plant whose results cites the improvement in reliability, availability, power quality and fault tolerant ability.

Keywords: solar photovoltaic, power electronics, power quality, PWM

Procedia PDF Downloads 247

Authors: Liang Wang, Beste F. Yuksel, David Guy Brizan

Abstract:

A computational model of affect that can distinguish between spontaneous and posed smiles with no errors on a large, popular data set using deep learning techniques is presented in this paper. A Long Short-Term Memory (LSTM) classifier, a type of Recurrent Neural Network, is utilized and compared to human classification. Results showed that while human classification (mean of 0.7133) was above chance, the LSTM model was more accurate than human classification and other comparable state-of-the-art systems. Additionally, a high accuracy rate was maintained with small amounts of training videos (70 instances). The derivation of important features to further understand the success of our computational model were analyzed, and it was inferred that thousands of pairs of points within the eyes and mouth are important throughout all time segments in a smile. This suggests that distinguishing between a posed and spontaneous smile is a complex task, one which may account for the difficulty and lower accuracy of human classification compared to machine learning models.

Keywords: affective computing, affect detection, computer vision, deep learning, human-computer interaction, machine learning, posed smile detection, spontaneous smile detection

Procedia PDF Downloads 99

Authors: Deyu Zhou, Xiao Xue, Lizhen Cui

Abstract:

With the increasing collaboration among various services and the growing complexity of user demands, there are more and more factors affecting the stable development of the cloud manufacturing service economic system (CMSE). This poses new challenges to the evolution analysis of the CMSE. Many researchers have modeled and analyzed the evolution process of CMSE from the perspectives of individual learning and internal factors influencing the system, but without considering other important characteristics of the system's individuals (such as heterogeneity, bounded rationality, etc.) and the impact of external environmental factors. Therefore, this paper proposes an integrated artificial social model for the cloud manufacturing service economic system, which considers both the characteristics of the system's individuals and the internal and external influencing factors of the system. The model consists of three parts: the Agent model, environment model, and rules model (Agent-Environment-Rules, AER): (1) the Agent model considers important features of the individuals, such as heterogeneity and bounded rationality, based on the adaptive behavior mechanisms of perception, action, and decision-making; (2) the environment model describes the activity space of the individuals (real or virtual environment); (3) the rules model, as the driving force of system evolution, describes the mechanism of the entire system's operation and evolution. Finally, this paper verifies the effectiveness of the AER model through computational and experimental results.

Keywords: cloud manufacturing service economic system (CMSE), AER model, artificial social modeling, integrated framework, computing experiment, agent-based modeling, social networks

Procedia PDF Downloads 47

Authors: Nathenal Thomas Lambamo

Abstract:

Based on the fact that more than 85% of the labor force and 90% of the export earnings are taken by agriculture in Ethiopia and it can be said that it is the backbone of the overall socio-economic activities in the country. Among the cereal crops that the agriculture sector provides for the country, wheat is the third-ranking one preceding teff and maize. In the present day, wheat is in higher demand related to the expansion of industries that use them as the main ingredient for their products. The local supply of wheat for these companies covers only 35 to 40% and the rest 60 to 65% percent is imported on behalf of potential customers that exhaust the country’s foreign currency reserves. The above facts show that the need for this crop in the country is too high and in reverse, the productivity of the crop is very less because of these reasons. Wheat disease is the most devastating disease that contributes a lot to this unbalance in the demand and supply status of the crop. It reduces both the yield and quality of the crop by 27% on average and up to 37% when it is severe. This study aims to detect the most frequent and degrading wheat diseases, Septoria and Leaf rust, using the most efficiently used subset of machine learning technology, deep learning. As a state of the art, a deep learning class classification technique called Convolutional Neural Network (CNN) has been used to detect diseases and has an accuracy of 99.01% is achieved.

Keywords: septoria, leaf rust, deep learning, CNN

Procedia PDF Downloads 46

Authors: Joseph Wolff, Jeffrey Eilbott

Abstract:

Autism Spectrum Disorders (ASD) are a spectrum of social disorders characterized by deficits in social communication, verbal ability, and interaction that can vary in severity. In recent years, researchers have used magnetic resonance imaging (MRI) to help detect how neural patterns in individuals with ASD differ from those of neurotypical (NT) controls for classification purposes. This study analyzed the classification of ASD within males and females using functional MRI data. Functional connectivity (FC) correlations among brain regions were used as feature inputs for machine learning algorithms. Analysis was performed on 558 cases from the Autism Brain Imaging Data Exchange (ABIDE) I dataset. When trained specifically on females, the algorithm underperformed in classifying the ASD subset of our testing population. Although the subject size was relatively smaller in the female group, the manual matching of both male and female training groups helps explain the algorithm’s bias, indicating the altered sex abnormalities in functional brain networks compared to typically developing peers. These results highlight the importance of taking sex into account when considering how generalizations of findings on males with ASD apply to females.

Keywords: autism spectrum disorder, machine learning, neuroimaging, sex differences

Procedia PDF Downloads 180

Authors: Thais Cavalheri Santos, Pedro Jose Gabriel Ferreira, Alexandre Daliberto Frugoli, Lucio Leonardo, Pedro Americo Frugoli

Abstract:

This paper presents a low cost education proposal to be used in engineering courses. The engineering education in universities of a developing country that is in need of an increasing number of engineers carried out with quality and affordably, pose a difficult problem to solve. In Brazil, the political and economic scenario requires academic managers able to reduce costs without compromising the quality of education. Within this context, the elaboration of a physics principles teaching method with the construction of an electronic balance is proposed. First, a method to develop and construct a load cell through which the students can understand the physical principle of strain gauges and bridge circuit will be proposed. The load cell structure was made with aluminum 6351T6, in dimensions of 80 mm x 13 mm x 13 mm and for its instrumentation, a complete Wheatstone Bridge was assembled with strain gauges of 350 ohms. Additionally, the process involves the use of a software tool to document the prototypes (design circuits), the conditioning of the signal, a microcontroller, C language programming as well as the development of the prototype. The project also intends to use an open-source I/O board (Arduino Microcontroller). To design the circuit, the Fritizing software will be used and, to program the controller, an open-source software named IDE®. A load cell was chosen because strain gauges have accuracy and their use has several applications in the industry. A prototype was developed for this study, and it confirmed the affordability of this educational idea. Furthermore, the goal of this proposal is to motivate the students to understand the several possible applications in high technology of the use of load cells and microcontroller.

Keywords: Arduino, load cell, low-cost education, strain gauge

Procedia PDF Downloads 272

Authors: Nasiru Medugu Idris

Abstract:

This research is aimed at assessing the vulnerability of climate change to rural communities (farmers, herdsmen and fishermen) in Nigeria with the view to study the underlying causes and degree of vulnerability to climate change and examine the conflict between farmers and herdsmen as a result of climate change. This research employed the use of quantitative and qualitative means of data gathering techniques as well as physical observations. Six states (Kebbi, Adamawa, Nasarawa, Osun, Ebonyi, and Akwa Ibom) have been selected on the ground that they are key food production areas in the country and are therefore essential to continual food security in the country. So also, they also double as fishing communities in order to aid the comprehensive study of all the effects on climate on farmers and fishermen alike. Community focus group discussions were carried out in the various states for an interactive session and also to have firsthand information on their level of awareness on climate change. Climate data from the Nigerian Meteorological Agency over the past decade were collected for the purpose of analyzing trends in climate. The study observed that the level of vulnerability of rural dwellers most especially farmers, herdsmen and fishermen to climate change is very high due to their socioeconomic, ethnic and historical perspective of their trend. The study, therefore, recommends that urgent step needs to be put in place to help control natural hazards and man-made disasters and serious measures are also needed in order to minimize severe societal, economic and political crises; some of which may either escalate to violent conflicts or could be avoided by efforts of conflict resolution and prevention by the initiation of a process of de-escalation. So this study has recommended the best-fit adaptive and mitigation measures to climate change vulnerability in rural communities of Nigeria.

Keywords: adaptation, farmers, fishermen, herdsmen

Procedia PDF Downloads 164

Authors: Huseyin A. Yavasoglu, Yusuf Ziya Tengiz, Ali Göksenli

Abstract:

In this study, an automatic feeding machine for different type and class of bovine animals is designed. Daily nutrition of a bovine consists of grass, corn, straw, silage, oat, wheat and different vitamins and minerals. The amount and mixture amount of each of the nutrition depends on different parameters of the bovine. These parameters are; age, sex, weight and maternity of the bovine, also outside temperature. The problem in a farm is to constitute the correct mixture and amount of nutrition for each animal. Faulty nutrition will cause an insufficient feeding of the animal concluding in an unhealthy bovine. To solve this problem, a new automatic feeding machine is designed. Travelling of the machine is performed by four tires, which is pulled by a tractor. The carrier consists of eight bins, which each of them carries a nutrition type. Capacity of each unit is 250 kg. At the bottom of each chamber is a sensor measuring the weight of the food inside. A funnel is at the bottom of each chamber by which open/close function is controlled by a valve. Each animal will carry a RFID tag including ID on its ear. A receiver on the feeding machine will read this ID and by given previous information by the operator (veterinarian), the system will detect the amount of each nutrition unit which will be given to the selected animal for feeding. In the system, each bin will open its exit gate by the help of the valve under the control of PLC (Programmable Logic Controller). The amount of each nutrition type will be controlled by measuring the open/close time. The exit canals of the bins are collected in a reservoir. To achieve a homogenous nitration, the collected feed will be mixed by a worm gear. Further the mixture will be transported by a help of a funnel to the feeding unit of the animal. The feeding process can be performed in 100 seconds. After feeding of the animal, the tractor pulls the travelling machine to the next animal. By the help of this system animals can be feeded by right amount and mixture of nutrition

Keywords: bovine, feeding, nutrition, transportation, automatic

Procedia PDF Downloads 311

Authors: Skiker Kaoutar, Mounir Maouene

Abstract:

Neuroimaging studies have shown that animal and tool concepts rely on distinct networks of brain areas. Animal concepts depend predominantly on temporal areas while tool concepts rely on fronto-temporo-parietal areas. However, the origin of this neurofunctional distinction for processing animal and tool concepts remains still unclear. Here, we address this question from a network perspective suggesting that the neural distinction between animals and tools might reflect the differences in their structural semantic networks. We build semantic networks for animal and tool concepts derived from Mc Rae and colleagues’s behavioral study conducted on a large number of participants. These two networks are thus analyzed through a large number of graph theoretical measures for small-worldness: centrality, clustering coefficient, average shortest path length, as well as resistance to random and targeted attacks. The results indicate that both animal and tool networks have small-world properties. More importantly, the animal network is more vulnerable to targeted attacks compared to the tool network a result that correlates with brain lesions studies.

Keywords: animals, tools, network, semantics, small-world, resilience to damage

Procedia PDF Downloads 507

Authors: Juan Pablo Botero Torres, Alba Avila, Luis Felipe Giraldo

Abstract:

Independent navigation and mobility through unknown spaces pose a challenge for the autonomy of visually impaired people (VIP), who have relied on the use of traditional assistive tools like the white cane and trained dogs. However, emerging visually assistive technologies (VAT) have proposed several human-machine interfaces (HMIs) that could improve VIP’s ability for self-guidance. Hereby, we introduce the design and implementation of a visually assistive device, Anajaa – Visual Substitution System (AVSS). This system integrates ultrasonic sensors with custom electronics, and computer vision models (convolutional neural networks), in order to achieve a robust system that acquires information of the surrounding space and transmits it to the user in an intuitive and efficient manner. AVSS consists of two modules: the sensing and the actuation module, which are fitted to a chest mount and belt that communicate via Bluetooth. The sensing module was designed for the acquisition and processing of proximity signals provided by an array of ultrasonic sensors. The distribution of these within the chest mount allows an accurate representation of the surrounding space, discretized in three different levels of proximity, ranging from 0 to 6 meters. Additionally, this module is fitted with an RGB-D camera used to detect potentially threatening obstacles, like staircases, using a convolutional neural network specifically trained for this purpose. Posteriorly, the depth data is used to estimate the distance between the stairs and the user. The information gathered from this module is then sent to the actuation module that creates an HMI, by the means of a 3x2 array of vibration motors that make up the tactile display and allow the system to deliver haptic feedback. The actuation module uses vibrational messages (tactones); changing both in amplitude and frequency to deliver different awareness levels according to the proximity of the obstacle. This enables the system to deliver an intuitive interface. Both modules were tested under lab conditions, and the HMI was additionally tested with a focal group of VIP. The lab testing was conducted in order to establish the processing speed of the computer vision algorithms. This experimentation determined that the model can process 0.59 frames per second (FPS); this is considered as an adequate processing speed taking into account that the walking speed of VIP is 1.439 m/s. In order to test the HMI, we conducted a focal group composed of two females and two males between the ages of 35-65 years. The subject selection was aided by the Colombian Cooperative of Work and Services for the Sightless (COOTRASIN). We analyzed the learning process of the haptic messages throughout five experimentation sessions using two metrics: message discrimination and localization success. These correspond to the ability of the subjects to recognize different tactones and locate them within the tactile display. Both were calculated as the mean across all subjects. Results show that the focal group achieved message discrimination of 70% and a localization success of 80%, demonstrating how the proposed HMI leads to the appropriation and understanding of the feedback messages, enabling the user’s awareness of its surrounding space.

Keywords: computer vision on embedded systems, electronic trave aids, human-machine interface, haptic feedback, visual assistive technologies, vision substitution systems

Procedia PDF Downloads 46

Authors: J. Alvarez, J. Pimienta, R. Sarmiento

Abstract:

Diode lasers working in free mode present different shifting and broadening determined by external factors such as temperature, current or mechanical vibrations, and they are not more useful in applications such as spectroscopy, metrology, and cooling of atoms, among others. Different configurations can reduce the spectral width of a laser; one of the most effective is to extend the optical resonator of the laser diode and use optical feedback either with the help of a partially reflective mirror or with a diffraction grating; this latter configuration is not only allowed to reduce the spectral width of the laser line but also to coarsely adjust its working wavelength, within a wide range typically ~ 10nm by slightly varying the angle of the diffraction grating. Two settings are commonly used for this purpose, the Littrow configuration and the Littmann Metcalf. In this paper, we present the design, construction, and characterization of a compact extended laser cavity in Littrow configuration. The designed cavity is compact and was machined on an aluminum block using computer numerical control (CNC); it has a mass of only 380 g. The design was tested on laser diodes with different wavelengths, 650nm, 780nm, and 795 nm, but can be equally efficient at other wavelengths. This report details the results obtained from the extended cavity working at a wavelength of 780 nm, with an output power of around 35mW and a line width of less than 1Mhz. The cavity was used to observe the spectrum of the corresponding Rubidium D2 line. By modulating the current and with the help of phase detection techniques, a dispersion signal with an excellent signal-to-noise ratio was generated that allowed the stabilization of the laser to a transition of the hyperfine structure of Rubidium with an integral proportional controller (PI) circuit made with precision operational amplifiers.

Keywords: Littrow, Littman-Metcalf, line width, laser stabilization, hyperfine structure

Procedia PDF Downloads 184

Authors: Navid Mohammadzadeh, Huy Truong-Ba, Michael Cholette

Abstract:

The solar tower (ST) plant is a promising technology to exploit large-scale solar irradiation. With thermal energy storage, ST plant has the potential to shift generation to high electricity price periods. However, the size of storage limits the dispatchability of the plant, particularly when it should compete with uncertainty in forecasts of solar irradiation and electricity prices. The purpose of this study is to explore the size of storage when Rolling Horizon Control (RHC) is employed for dispatch scheduling. To this end, RHC is benchmarked against perfect knowledge (PK) forecast and two day-ahead dispatching policies. With optimisation of dispatch planning using PK policy, the optimal achievable profit for a specific size of the storage is determined. A sensitivity analysis using Monte-Carlo simulation is conducted, and the size of storage for RHC and day-ahead policies is determined with the objective of reaching the profit obtained from the PK policy. A case study is conducted for a hypothetical ST plant with thermal storage located in South Australia and intends to dispatch under two market scenarios: 1) fixed price and 2) wholesale spot price. The impact of each individual source of uncertainty on storage size is examined for January and August. The exploration of results shows that dispatching with RH controller reaches optimal achievable profit with ~15% smaller storage compared to that in day-ahead policies. The results of this study may be applied to the CSP plant design procedure.

Keywords: solar tower plant, spot market, thermal storage system, optimized dispatch planning, sensitivity analysis, Monte Carlo simulation

Procedia PDF Downloads 91

Authors: Avinash Kumar Ranjan, Akash Anand

Abstract:

The Land Surface Temperature (LST) is an essential factor accompanying to rise urban heat and climate warming within a city in micro level. It is also playing crucial role in global change study as well as radiation budgets measuring in heat balance studies. The information of LST is very substantial to recognize the urban climatology, ecological changes, anthropological and environmental interactions etc. The Chief motivation of present study focus on time series of ANN model that taken a sequence of LST values of 2000, 2008 and 2016, realize the pattern of variation within the data set and predict the LST values for 2024 and 2032. The novelty of this study centers on evaluation of LST using series of multi-temporal MODIS (MOD 11A2) satellite data by Maximum Value Composite (MVC) techniques. The results derived from this study endorse the proficiency of Geoinformatics Technology with integration of ANN to gain knowledge, understanding and building of precise forecast from the complex physical world database. This study will also focus on influence of Land Use/ Land Cover (LU/LC) variation on Land Surface Temperature.

Keywords: LST, geoinformatics technology, ANN, MODIS satellite imagery, MVC

Procedia PDF Downloads 211

Authors: Weiwei Cui, Dejian Lin, Leigang Zhang, Yao Wang, Zheng Sun, Lianfeng Li

Abstract:

Built-in test (BIT) is an important technology in testability field, and it is widely used in state monitoring and fault diagnosis. With the improvement of modern equipment performance and complexity, the scope of BIT becomes larger, and it leads to the emergence of false alarm problem. The false alarm makes the health assessment unstable, and it reduces the effectiveness of BIT. The conventional false alarm suppression methods such as repeated test and majority voting cannot meet the requirement for a complicated system, and the intelligence algorithms such as artificial neural networks (ANN) are widely studied and used. However, false alarm has a very low frequency and small sample, yet a method based on ANN requires a large size of training sample. To recognize the false alarm, we propose a method based on multi-classification support vector machine (SVM) in this paper. Firstly, we divide the state of a system into three states: healthy, false-alarm, and faulty. Then we use multi-classification with '1 vs 1' policy to train and recognize the state of a system. Finally, an example of fault injection system is taken to verify the effectiveness of the proposed method by comparing ANN. The result shows that the method is reasonable and effective.

Keywords: false alarm, fault diagnosis, SVM, k-means, BIT

Procedia PDF Downloads 114

Authors: Xiaodong Li, Peng Gao, Chao-Jung Huang, Shiying Hao, Xuefeng B. Ling, Yongxia Han, Yaqi Zhang, Le Zheng, Chengyin Ye, Modi Liu, Minjie Xia, Changlin Fu, Bo Jin, Karl G. Sylvester, Eric Widen

Abstract:

Predicting the risk of Pancreatic Adenocarcinoma (PA) in advance can benefit the quality of care and potentially reduce population mortality and morbidity. The aim of this study was to develop and prospectively validate a risk prediction model to identify patients at risk of new incident PA as early as 3 months before the onset of PA in a statewide, general population in Maine. The PA prediction model was developed using Deep Neural Networks, a deep learning algorithm, with a 2-year electronic-health-record (EHR) cohort. Prospective results showed that our model identified 54.35% of all inpatient episodes of PA, and 91.20% of all PA that required subsequent chemoradiotherapy, with a lead-time of up to 3 months and a true alert of 67.62%. The risk assessment tool has attained an improved discriminative ability. It can be immediately deployed to the health system to provide automatic early warnings to adults at risk of PA. It has potential to identify personalized risk factors to facilitate customized PA interventions.

Keywords: cancer prediction, deep learning, electronic health records, pancreatic adenocarcinoma

Procedia PDF Downloads 120

Authors: Hyohun Kim, Dongwha Shin, Yeonseok Kim, Ji-Su Ahn, Kensuke Nakamura, Dongeun Choi, Byung-Woo Hong

Abstract:

Artefacts are commonly encountered in the imaging process of clinical computed tomography (CT) where the artefact refers to any systematic discrepancy between the reconstructed observation and the true attenuation coefficient of the object. It is known that CT images are inherently more prone to artefacts due to its image formation process where a large number of independent detectors are involved, and they are assumed to yield consistent measurements. There are a number of different artefact types including noise, beam hardening, scatter, pseudo-enhancement, motion, helical, ring, and metal artefacts, which cause serious difficulties in reading images. Thus, it is desired to remove nuisance factors from the degraded image leaving the fundamental intrinsic information that can provide better interpretation of the anatomical and pathological characteristics. However, it is considered as a difficult task due to the high dimensionality and variability of data to be recovered, which naturally motivates the use of machine learning techniques. We propose an image restoration algorithm based on the deep neural network framework where the denoising auto-encoders are stacked building multiple layers. The denoising auto-encoder is a variant of a classical auto-encoder that takes an input data and maps it to a hidden representation through a deterministic mapping using a non-linear activation function. The latent representation is then mapped back into a reconstruction the size of which is the same as the size of the input data. The reconstruction error can be measured by the traditional squared error assuming the residual follows a normal distribution. In addition to the designed loss function, an effective regularization scheme using residual-driven dropout determined based on the gradient at each layer. The optimal weights are computed by the classical stochastic gradient descent algorithm combined with the back-propagation algorithm. In our algorithm, we initially decompose an input image into its intrinsic representation and the nuisance factors including artefacts based on the classical Total Variation problem that can be efficiently optimized by the convex optimization algorithm such as primal-dual method. The intrinsic forms of the input images are provided to the deep denosing auto-encoders with their original forms in the training phase. In the testing phase, a given image is first decomposed into the intrinsic form and then provided to the trained network to obtain its reconstruction. We apply our algorithm to the restoration of the corrupted CT images by the artefacts. It is shown that our algorithm improves the readability and enhances the anatomical and pathological properties of the object. The quantitative evaluation is performed in terms of the PSNR, and the qualitative evaluation provides significant improvement in reading images despite degrading artefacts. The experimental results indicate the potential of our algorithm as a prior solution to the image interpretation tasks in a variety of medical imaging applications. This work was supported by the MISP(Ministry of Science and ICT), Korea, under the National Program for Excellence in SW (20170001000011001) supervised by the IITP(Institute for Information and Communications Technology Promotion).

Keywords: auto-encoder neural network, CT image artefact, deep learning, intrinsic image representation, noise reduction, total variation

Procedia PDF Downloads 165

Authors: Ranojoy Dutta, Adam Barker

Abstract:

Multifamily residential buildings are increasingly being built with large glazed areas to provide tenants with greater daylight and outdoor views. However, traditional double-glazed window assemblies can lead to significant thermal discomfort from high radiant temperatures as well as increased cooling energy use to address solar gains. Dynamic glazing provides an effective solution by actively controlling solar transmission to maintain indoor thermal comfort, without compromising the visual connection to outdoors. This study uses thermal simulations across three Canadian cities (Toronto, Vancouver and Montreal) to verify if dynamic glazing along with operable windows and ceiling fans can maintain the indoor operative temperature of a prototype southwest facing high-rise apartment unit within the ASHRAE 55 adaptive comfort range for a majority of the year, without any mechanical cooling. Since this study proposes the use of natural ventilation for cooling and the typical building life cycle is 30-40 years, the typical weather files have been modified based on accepted global warming projections for increased air temperatures by 2050. Results for the prototype apartment confirm that thermal discomfort with dynamic glazing occurs only for less than 0.7% of the year. However, in the baseline scenario with low-E glass there are up to 7% annual hours of discomfort despite natural ventilation with operable windows and improved air movement with ceiling fans.

Keywords: electrochromic glazing, multi-family housing, passive cooling, thermal comfort, natural ventilation

Procedia PDF Downloads 74

Authors: Chaudhary Muhammad Aqdus Ilyas, Matthias Rehm, Kamal Nasrollahi, Thomas B. Moeslund

Abstract:

This paper presents the investigation of data collection procedures, associated with robots when placed with traumatic brain injured (TBI) patients for rehabilitation purposes through facial expression and mood analysis. Rehabilitation after TBI is very crucial due to nature of injury and variation in recovery time. It is advantageous to analyze these emotional signals in a contactless manner, due to the non-supportive behavior of patients, limited muscle movements and increase in negative emotional expressions. This work aims at the development of framework where robots can recognize TBI emotions through facial expressions to perform rehabilitation tasks by physical, cognitive or interactive activities. The result of these studies shows that with customized data collection strategies, proposed framework identify facial and emotional expressions more accurately that can be utilized in enhancing recovery treatment and social interaction in robotic context.

Keywords: computer vision, convolution neural network- long short term memory network (CNN-LSTM), facial expression and mood recognition, multimodal (RGB-thermal) analysis, rehabilitation, robots, traumatic brain injured patients

Procedia PDF Downloads 115

Authors: Teja Koduru

Abstract:

Untreated Lyme disease can lead to neurological, cardiac, and dermatological complications. Rapid diagnosis of the erythema migrans (EM) rash, a characteristic symptom of Lyme disease is therefore crucial to early diagnosis and treatment. In this study, we aim to utilize deep learning frameworks including Tensorflow and Keras to create deep convolutional neural networks (DCNN) to detect images of acute Lyme Disease from images of erythema migrans. This study uses a custom database of erythema migrans images of varying quality to train a DCNN capable of classifying images of EM rashes vs. non-EM rashes. Images from publicly available sources were mined to create an initial database. Machine-based removal of duplicate images was then performed, followed by a thorough examination of all images by a clinician. The resulting database was combined with images of confounding rashes and regular skin, resulting in a total of 683 images. This database was then used to create a DCNN with an accuracy of 93% when classifying images of rashes as EM vs. non EM. Finally, this model was converted into a web and mobile application to allow for rapid diagnosis of EM rashes by both patients and clinicians. This tool could be used for patient prescreening prior to treatment and lead to a lower mortality rate from Lyme disease.

Keywords: Lyme, untreated Lyme, erythema migrans rash, EM rash

Procedia PDF Downloads 201

Authors: Alae El Fahsi

Abstract:

This study explores the ‘Tajawob’ app's role in urban development, demonstrating how mobile applications can empower citizens and facilitate urban planning. Tajawob serves as a digital platform for community feedback, engagement, and participatory governance, addressing urban challenges through innovative tech solutions. This research synthesizes data from a variety of sources, including user feedback, engagement metrics, and interviews with city officials, to assess the app’s impact on citizen participation in urban development in Morocco. By integrating advanced data analytics and user experience design, Tajawob has bridged the communication gap between citizens and government officials, fostering a more collaborative and transparent urban planning process. The findings reveal a significant increase in civic engagement, with users actively contributing to urban management decisions, thereby enhancing the responsiveness and inclusivity of urban governance. Challenges such as digital literacy, infrastructure limitations, and privacy concerns are also discussed, providing a comprehensive overview of the obstacles and opportunities presented by mobile app-based citizen engagement platforms. The study concludes with strategic recommendations for scaling the Tajawob model to other contexts, emphasizing the importance of adaptive technology solutions in meeting the evolving needs of urban populations. This research contributes to the burgeoning field of smart city innovations, offering key insights into the role of digital tools in facilitating more democratic and participatory urban environments.

Keywords: smart cities, digital governance, urban planning, strategic design

Procedia PDF Downloads 17

Authors: M. Taheri Tehrani, H. Ajorloo

Abstract:

In this paper, an intelligent multi-agent framework is developed for each router in which agents have two vital functionalities, traffic shaping and buffer allocation and are positioned in the ports of the routers. With traffic shaping functionality agents shape the traffic forward by dynamic and real time allocation of the rate of generation of tokens in a Token Bucket algorithm and with buffer allocation functionality agents share their buffer capacity between each other based on their need and the conditions of the network. This dynamic and intelligent framework gives this opportunity to some ports to work better under burst and more busy conditions. These agents work intelligently based on Reinforcement Learning (RL) algorithm and will consider effective parameters in their decision process. As RL have limitation considering much parameter in its decision process due to the volume of calculations, we utilize our novel method which invokes Principle Component Analysis (PCA) on the RL and gives a high dimensional ability to this algorithm to consider as much as needed parameters in its decision process. This implementation when is compared to our previous work where traffic shaping was done without any sharing and dynamic allocation of buffer size for each port, the lower packet drop in the whole network specifically in the source routers can be seen. These methods are implemented in our previous proposed intelligent simulation environment to be able to compare better the performance metrics. The results obtained from this simulation environment show an efficient and dynamic utilization of resources in terms of bandwidth and buffer capacities pre allocated to each port.

Keywords: principal component analysis, reinforcement learning, buffer allocation, multi- agent systems

Procedia PDF Downloads 476

Authors: Amartya Hatua, Trung Nguyen, Andrew Sung

Abstract:

In this paper, we study the information diffusion process on Twitter as a multivariate time series problem. Our model concerns three measures (volume, network influence, and sentiment of tweets) based on 10 features, and we collected 27 million tweets to build our information diffusion time series dataset for analysis. Then, different time series clustering techniques with Dynamic Time Warping (DTW) distance were used to identify different patterns of information diffusion. Finally, we built the information diffusion prediction models for new hashtags which comprise two phrases: The first phrase is recognizing the pattern using k-NN with DTW distance; the second phrase is building the forecasting model using the traditional Autoregressive Integrated Moving Average (ARIMA) model and the non-linear recurrent neural network of Long Short-Term Memory (LSTM). Preliminary results of performance evaluation between different forecasting models show that LSTM with clustering information notably outperforms other models. Therefore, our approach can be applied in real-world applications to analyze and predict the information diffusion characteristics of selected topics or memes (hashtags) in Twitter.

Keywords: ARIMA, DTW, information diffusion, LSTM, RNN, time series clustering, time series forecasting, Twitter

Procedia PDF Downloads 359

Authors: Ali Reza Sattarzadeh, Ronny J. Kutadinata, Pubudu N. Pathirana, Van Thanh Huynh

Abstract:

Traffic congestion has become a severe worldwide problem, affecting everyday life, fuel consumption, time, and air pollution. The primary causes of these issues are inadequate transportation infrastructure, poor traffic signal management, and rising population. Traffic flow forecasting is one of the essential and effective methods in urban congestion and traffic management, which has attracted the attention of researchers. With the development of technology, undeniable progress has been achieved in existing methods. However, there is a possibility of improvement in the extraction of temporal and spatial features to determine the importance of traffic flow sequences and extraction features. In the proposed model, we implement the convolutional neural network (CNN) and long short-term memory (LSTM) deep learning models for mining nonlinear correlations and their effectiveness in increasing the accuracy of traffic flow prediction in the real dataset. According to the experiments, the results indicate that implementing Conv-LSTM networks increases the productivity and accuracy of deep learning models for traffic flow prediction.

Keywords: deep learning algorithms, intelligent transportation systems, spatiotemporal features, traffic flow prediction

Procedia PDF Downloads 123

Authors: Ae Ra Han, Seoung Eun Huh, Ji Yeon Kim, Joanne Kwak-Kim, Sung Ki Lee

Abstract:

Objective: Prevalence of osteoporosis, cardiovascular disorders, and Alzheimer's disease rapidly increase after menopause. Immune activation and inflammation are suggested as an important pathogenesis of these serious diseases. Several pro-inflammatory cytokines are increased in women with surgical or natural menopause. However, the little is known about IL-17 producing T cells and Foxp3+ regulatory T (Treg) cells in post-menopause. Methods: A total of 34 postmenopausal women, who had no active cardiovascular, endocrine and infectious disorders were recruited as study group and healthy premenopausal women participated as controls. Peripheral blood mononuclear cells were isolated. Immuno-morphologic (CD3, CD4, CD8, CD19, CD56/CD16), intracellular cytokine (TNF-alpha, IFN-gamma, IL-10, IL-17), and Treg cell (Foxp3) studies were carried out using flow cytometry. The proportion of peripheral lymphocytes, including IL-17 producing and Foxp3+ Treg cells immune cell in each group were statistically analyzed. Results: The proportion of NK cells was significantly increased in menopausal women as compared to that of controls (P=.005). The ratios of TNF-alpha/IL-10 producing CD3+CD4+ T cells were increased in postmenopausal women. CD3+IL-17+ T cell level was higher in postmenopausal women and CD4+ Foxp3+ Treg cells was lower than that of controls. The ratios of CD3+IL-17+ T cell to CD3+Foxp3+ and to CD4+Foxp3+ Treg cells were significantly increased in postmenopausal women (P=.001). Conclusions: We found enhanced innate immunity and Th1- and Th17-mediated adaptive immunity in postmenopausal women. This may explain increasing prevalence of chronic inflammatory diseases after menopause. Further studies are needed to elucidate what factors contribute to this inflammatory shift in the postmenopause.

Keywords: inflammation, immune cell, menopause, Th17, regulatory T cell

Procedia PDF Downloads 292

Authors: V. Santamarina Campos, M. de Miguel Molina, B. de Miguel Molina, M. Á. Carabal Montagud

Abstract:

With this contribution, we want to show a successful example of the application of the Design Thinking methodology, in the European project 'Technology transfer of Remotely Piloted Aircraft Systems (RPAS) for the creative industry'. The use of this methodology has allowed us to design and build a drone, based on the real needs of prospective users. It has demonstrated that this is a powerful tool for generating innovative ideas in the field of robotics, by focusing its effectiveness on understanding and solving real user needs. In this way, with the support of an interdisciplinary team, comprised of creatives, engineers and economists, together with the collaboration of prospective users from three European countries, a non-linear work dynamic has been created. This teamwork has generated a sense of appreciation towards the creative industries, through continuously adaptive, inventive, and playful collaboration and communication, which has facilitated the development of prototypes. These have been designed to enable filming and photography in interior spaces, within 13 sectors of European creative industries: Advertising, Architecture, Fashion, Film, Antiques and Museums, Music, Photography, Televison, Performing Arts, Publishing, Arts and Crafts, Design and Software. Furthermore, it has married the real needs of the creative industries, with what is technologically and commercially viable. As a result, a product of great value has been obtained, which offers new business opportunities for small companies across this sector.

Keywords: design thinking, design for effectiveness, methodology, active toolkit, storyboards, PAR, focus group, innovation, RPAS, indoor drone, aerial film, creative industry, end users, stakeholder

Procedia PDF Downloads 176

Authors: Masoud Sheidai

Abstract:

Biologists with a special field of population genetics and phylogeny have different research tasks such as populations’ genetic variability and divergence, species relatedness, the evolution of genetic and morphological characters, and identification of DNA SNPs with adaptive potential. To tackle these problems and reach a concise conclusion, they must use the proper and efficient statistical and bioinformatic methods as well as suitable genetic and morphological characteristics. In recent years application of different bioinformatic and statistical methods, which are based on various well-documented assumptions, are the proper analytical tools in the hands of researchers. The species delineation is usually carried out with the use of different clustering methods like K-means clustering based on proper distance measures according to the studied features of organisms. A well-defined species are assumed to be separated from the other taxa by molecular barcodes. The species relationships are studied by using molecular markers, which are analyzed by different analytical methods like multidimensional scaling (MDS) and principal coordinate analysis (PCoA). The species population structuring and genetic divergence are usually investigated by PCoA and PCA methods and a network diagram. These are based on bootstrapping of data. The Association of different genes and DNA sequences to ecological and geographical variables is determined by LFMM (Latent factor mixed model) and redundancy analysis (RDA), which are based on Bayesian and distance methods. Molecular and morphological differentiating characters in the studied species may be identified by linear discriminant analysis (DA) and discriminant analysis of principal components (DAPC). We shall illustrate these methods and related conclusions by giving examples from different edible and medicinal plant species.

Keywords: GWAS analysis, K-Means clustering, LFMM, multidimensional scaling, redundancy analysis

Procedia PDF Downloads 87

Authors: Aminul Haque, Aminul Islam, Prabal Kumar Chowdhury

Abstract:

One of the world’s most pressing issues right now is the lack of a competent waste management system, particularly in emerging and underdeveloped countries. Recycling solid waste, which comprises numerous dangerous non-biodegradable sub-stances like glass, metals, plastics, etc, is the most essential step in reducing waste-related issues in the environment. Typically, collected waste includes all types of waste that must be thoroughly sorted to be recycled efficiently. Most countries use manual waste sorting techniques, which are efficient. Nevertheless, the waste sorting process by human beings is not safe as there is always a risk of exposing themselves to toxic wastes, which could be serious for their health. Our thesis presents a Deep Learning technique based on computer vision for automatically identifying waste. To construct the model, we used Convolutional Neural Networks, real-time object detection systems, such as YOLOv5 and YOLOv7, as well as several transfers learning-based architectures, including VGG16, MobileNet, Inception-Resnet-v2. The model is trained on numerous images for each type of waste to ensure no overfitting and greater accuracy. The highest accuracy we achieved for our waste detection model YOLOv5x, is 93.7%.

Keywords: deep learning, object detection, YOLOv7, image processing, computer vision

Procedia PDF Downloads 13