Search results for: deep vein imaging

Authors: S. N. Derrar, M. Sekkal-Rahal

Abstract:

The Nonlinear optics (NLO) technique is widely used in the field of biological imaging. In fact, grafting biological entities with a high NLO response on tissues and cells enhances the NLO responses of these latter, and ameliorates, consequently, their biological imaging quality. In this optics, we carried out a theoretical study, in the aim of analyzing the peptide bonds created between alanine amino acid and both unnatural amino acids: L-Dopa and Azatryptophan, respectively. Ramachandran plots have been performed for these systems, and their structural parameters have been analyzed. The NLO responses of these peptides have been reported by calculating the first hyperpolarizability values of all the minima found on the plots. The use of such unnatural amino acids as endogenous probing molecules has been investigated through this study. The Density Functional Theory (DFT) has been used for structural properties, while the Second-order Møller-Plesset Perturbation Theory (MP2) has been employed for the NLO calculations.

Keywords: biological imaging, hyperpolarizability, nonlinear optics, probing molecule

Procedia PDF Downloads 359

Authors: Jun Ming Moey, Zhiyaun Chen, David Nicholson

Abstract:

Bad actors are often hard to detect in data that imprints their behaviour patterns because they are comparatively rare events embedded in non-bad actor data. An unsupervised machine learning framework is applied here to detect bad actors in financial crime datasets that record millions of transactions undertaken by hundreds of actors (<0.01% bad). Specifically, the framework combines ‘shallow’ (PCA, Isolation Forest) and ‘deep’ (Autoencoder) methods to detect outlier patterns. Detection performance analysis for both the individual methods and their combination is reported.

Keywords: detection, machine learning, deep learning, unsupervised, outlier analysis, data science, fraud, financial crime

Procedia PDF Downloads 73

Authors: Haozhe Xiang

Abstract:

With the rise of the Internet of Things era, intelligent products are gradually integrating into people's lives. Pedestrian trajectory prediction has become a key issue, which is crucial for the motion path planning of intelligent agents such as autonomous vehicles, robots, and drones. In the current technological context, deep learning technology is becoming increasingly sophisticated and gradually replacing traditional models. The pedestrian trajectory prediction algorithm combining neural networks and attention mechanisms has significantly improved prediction accuracy. Based on in-depth research on deep learning and pedestrian trajectory prediction algorithms, this article focuses on physical environment modeling and learning of historical trajectory time dependence. At the same time, social interaction between pedestrians and scene interaction between pedestrians and the environment were handled. An improved pedestrian trajectory prediction algorithm is proposed by analyzing the existing model architecture. With the help of these improvements, acceptable predicted trajectories were successfully obtained. Experiments on public datasets have demonstrated the algorithm's effectiveness and achieved acceptable results.

Keywords: deep learning, graph convolutional network, attention mechanism, LSTM

Procedia PDF Downloads 46

Authors: Fei Ye, Åsa Barrefelt, Manuchehr Abedi-Valugerdi, Khalid M. Abu-Salah, Salman A. Alrokayan, Mamoun Muhammed, Moustapha Hassan

Abstract:

With appropriate encapsulation in functional nanoparticles drugs are more stable in physiological environment and the kinetics of the drug can be more carefully controlled and monitored. Furthermore, targeted drug delivery can be developed to improve chemotherapy in cancer treatment, not only by enhancing intracellular uptake by target cells but also by reducing the adverse effects in non-target organs. Inorganic imaging agents, delivered together with anti-cancer drugs, enhance the local imaging contrast and provide precise diagnosis as well as evaluation of therapy efficacy. We have developed biodegradable polymeric vesicles as a nanocarrier system for multimodal bio-imaging and anticancer drug delivery. The poly (lactic-co-glycolic acid) PLGA) vesicles were fabricated by encapsulating inorganic imaging agents of superparamagnetic iron oxide nanoparticles (SPION), manganese-doped zinc sulfide (MN:ZnS) quantum dots (QDs) and the anticancer drug busulfan into PLGA nanoparticles via an emulsion-evaporation method. T2-weighted magnetic resonance imaging (MRI) of PLGA-SPION-Mn:ZnS phantoms exhibited enhanced negative contrast with r2 relaxivity of approximately 523 s-1 mM-1 Fe. Murine macrophage (J774A) cellular uptake of PLGA vesicles started fluorescence imaging at 2 h and reached maximum intensity at 24 h incubation. The drug delivery ability PLGA vesicles was demonstrated in vitro by release of busulfan. PLGA vesicles degradation was studied in vitro, showing that approximately 32% was degraded into lactic and glycolic acid over a period of 5 weeks. The biodistribution of PLGA vesicles was investigated in vivo by MRI in a rat model. Change of contrast in the liver could be visualized by MRI after 7 min and maximal signal loss detected after 4 h post-injection of PLGA vesicles. Histological studies showed that the presence of PLGA vesicles in organs was shifted from the lungs to the liver and spleen over time.

Keywords: biodegradable polymers, multifunctional nanoparticles, quantum dots, anticancer drugs

Procedia PDF Downloads 454

Authors: Bang Young Kim, Sang Hoon Park, Chul Gyu Song

Abstract:

A conventional optical coherence tomography (OCT) system has limited imaging depth, which is 1-2 mm, and suffers unwanted noise such as speckle noise. The motorized-stage-based OCT system, using a common-path Fourier-domain optical coherence tomography (CP-FD-OCT) configuration, provides enhanced imaging depth and less noise so that we can overcome these limitations. Using this OCT systems, OCT images were obtained from an onion, and their subsurface structure was observed. As a result, the images obtained using the developed motorized-stage-based system showed enhanced imaging depth than the conventional system, since it is real-time accurate depth tracking. Consequently, the developed CP-FD-OCT systems and algorithms have good potential for the further development of endoscopic OCT for microsurgery.

Keywords: common-path OCT, FD-OCT, OCT, tracking algorithm

Procedia PDF Downloads 364

Authors: Krzysztof Malczewski

Abstract:

Magnetic Resonance Imaging (MRI) reconstruction algorithm using compressed sensing is presented in this paper. It is exhibited that the offered approach improves MR images spatial resolution in circumstances when highly undersampled k-space trajectories are applied. Compressed Sensing (CS) aims at signal and images reconstructing from significantly fewer measurements than were conventionally assumed necessary. Magnetic Resonance Imaging (MRI) is a fundamental medical imaging method struggles with an inherently slow data acquisition process. The use of CS to MRI has the potential for significant scan time reductions, with visible benefits for patients and health care economics. In this study the objective is to combine super-resolution image enhancement algorithm with CS framework benefits to achieve high resolution MR output image. Both methods emphasize on maximizing image sparsity on known sparse transform domain and minimizing fidelity. The presented algorithm considers the cardiac and respiratory movements.

Keywords: super-resolution, MRI, compressed sensing, sparse-sense, image enhancement

Procedia PDF Downloads 407

Authors: Gaokai Liu

Abstract:

Automatic defect detection for nanomaterial images is widely required in industrial scenarios. Deep learning approaches are considered as the most effective solutions for the great majority of image-based tasks. In this paper, an edge guidance network for defect segmentation is proposed. First, the encoder path with multiple convolution and downsampling operations is applied to the acquisition of shared features. Then two decoder paths both are connected to the last convolution layer of the encoder and supervised by the edge and segmentation labels, respectively, to guide the whole training process. Meanwhile, the edge and encoder outputs from the same stage are concatenated to the segmentation corresponding part to further tune the segmentation result. Finally, the effectiveness of the proposed method is verified via the experiments on open nanofibrous datasets.

Keywords: deep learning, defect detection, image segmentation, nanomaterials

Procedia PDF Downloads 123

Authors: A. S. Lunev, A. A. Larenkov, O. E. Klementyeva, G. E. Kodina

Abstract:

Background and aims: 68Ga-citrate is one of prospective radiopharmaceutical for PET-imaging of inflammation and infection. 68Ga-citrate is 67Ga-citrate analogue using since 1970s for SPECT-imaging. There's known rebinding reaction occurs past Ga-citrate injection and gallium (similar iron Fe3+) binds with blood transferrin. Then radiolabeled protein complex is delivered to pathological foci (inflammation/infection sites). But excessive gallium bindings with transferrin are cause of slow blood clearance, long accumulation time in foci (24-72 h) and exception of application possibility of the short-lived gallium-68 (T½ = 68 min). Injection of additional chemical agents (e.g. Fe3+ compounds) competing with radioactive gallium to the blood transferrin joining (blocking of its metal binding capacity) is one of the ways to solve formulated problem. This phenomenon can be used for correction of 68Ga-citrate pharmacokinetics for increasing of the blood clearance and accumulation in foci. The aim of real studying is research of effect of stable Fe-citrate on 68Ga-citrate tissue distribution. Materials and methods: 68Ga-citrate without/with extra injection of stable Fe-citrate (III) was injected nonlinear mice with inflammation models (aseptic soft tissue inflammation, lung infection, osteomyelitis). PET/X-RAY Genisys4 (Sofie Bioscience, USA) was used for non-invasive PET imaging (for 30, 60, 120 min past injection 68Ga-citrate) with subsequent reconstruction of imaging and their analysis (value of clearance, distribution volume). Scanning time is 10 min. Results and conclusions: I. v. injection of stable Fe-citrate blocks the metal-binding capability of transferrin serum and allows decreasing gallium-68 radioactivity in blood significantly and increasing accumulation in inflammation (3-5 time). It allows receiving more informative PET-images of inflammation early (for 30-60 min after injection). Pharmacokinetic parameters prove it. Noted there is no statistically significant difference between 68Ga-citrate accumulation for different inflammation model because PET imaging is indication of pathological processes and is not their identification.

Keywords: 68Ga-citrate, Fe-citrate, PET imaging, mice, inflammation, infection

Procedia PDF Downloads 464

Authors: Li Qingjian, Li Ke, He Chun, Huang Yong

Abstract:

In this paper, the method of combining the Pohl Seidman's deep belief network with the self-organizing neural network is proposed to classify the target. This method is mainly aimed at the high nonlinearity of the hyperspectral image, the high sample dimension and the difficulty in designing the classifier. The main feature of original data is extracted by deep belief network. In the process of extracting features, adding known labels samples to fine tune the network, enriching the main characteristics. Then, the extracted feature vectors are classified into the self-organizing neural network. This method can effectively reduce the dimensions of data in the spectrum dimension in the preservation of large amounts of raw data information, to solve the traditional clustering and the long training time when labeled samples less deep learning algorithm for training problems, improve the classification accuracy and robustness. Through the data simulation, the results show that the proposed network structure can get a higher classification precision in the case of a small number of known label samples.

Keywords: DBN, SOM, pattern classification, hyperspectral, data compression

Procedia PDF Downloads 323

Authors: Shubham Gogri, Lucia Florescu

Abstract:

Medical diagnosis and radiotherapy treatment planning using Computed Tomography (CT) rely on the quantitative accuracy and quality of the CT images. At the same time, requirements for CT imaging include reducing the radiation dose exposure to patients and minimizing scanning time. A solution to this is the sparse-view CT technique, based on a reduced number of projection views. This, however, introduces a new problem— the incomplete projection data results in lower quality of the reconstructed images. To tackle this issue, deep learning methods have been applied to enhance the quality of the sparse-view CT images. A first approach involved employing Mir-Net, a dedicated deep neural network designed for image enhancement. This showed promise, utilizing an intricate architecture comprising encoder and decoder networks, along with the incorporation of the Charbonnier Loss. However, this approach was computationally demanding. Subsequently, a specialized Generative Adversarial Network (GAN) architecture, rooted in the Pix2Pix framework, was implemented. This GAN framework involves a U-Net-based Generator and a Discriminator based on Convolutional Neural Networks. To bolster the GAN's performance, both Charbonnier and Wasserstein loss functions were introduced, collectively focusing on capturing minute details while ensuring training stability. The integration of the perceptual loss, calculated based on feature vectors extracted from the VGG16 network pretrained on the ImageNet dataset, further enhanced the network's ability to synthesize relevant images. A series of comprehensive experiments with clinical CT data were conducted, exploring various GAN loss functions, including Wasserstein, Charbonnier, and perceptual loss. The outcomes demonstrated significant image quality improvements, confirmed through pertinent metrics such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) between the corrected images and the ground truth. Furthermore, learning curves and qualitative comparisons added evidence of the enhanced image quality and the network's increased stability, while preserving pixel value intensity. The experiments underscored the potential of deep learning frameworks in enhancing the visual interpretation of CT scans, achieving outcomes with SSIM values close to one and PSNR values reaching up to 76.

Keywords: generative adversarial networks, sparse view computed tomography, CT image correction, Mir-Net

Procedia PDF Downloads 130

Authors: Agoubi Belgacem, Adel Kharroubi

Abstract:

Geothermal groundwater is the main water source to supply various sectors in El Hamma city, southeastern Tunisia. This region was long the destination of thousands of people from Tunisia and neighboring countries for care and bathing. The main objective of this study is to understand the groundwater mineralization origins and factors that control. The second goal is the appraisal of radon in geothermal groundwater in the study area. For this aim, geothermal groundwater was sampled and collected from different locations (thermal baths and deep wells). Physical parameters were measured and major ions were analyzed. Results reveal three water types. The water first type has Na-Mg-Ca-SO4-Cl facies and T>55°C. The second water type dominated by Na-Ca-Cl-SO4 facies with a temperature < 45 °C. However the third water type is dominated by Ca-SO4-Na-Cl-Mg. The three water types may be controlled by depth and geology. The first represent groundwater from deep aquifer (lower cretaceous), the second type was the shallow aquifer and the first is mixed water from deep and shallow water with a temperature ranging from 45 to 55°C. Measured Radon shows that shallow aquifer has a higher 222Rn concentration (677 to 2903 Bq.m-3) than deep water (203 to 1100 Bq.m-3). R-222 in El Hamma thermal aquifer was controlled by structures, porosity and permeability of aquifers. Geostatistical analyses of hydrogeological data and radon activities confirm the vertical flow and communication between deep and shallow aquifers through vertical faults system.

Keywords: Radon-222, geothermal, water, environment, Tunisia

Procedia PDF Downloads 344

Authors: Jae Won Lee, Kyung Hyun Min, Hong Jae Lee, Sang Cheon Lee

Abstract:

To date, imaging techniques have attracted much attention in medicine because the detection of diseases at an early stage provides greater opportunities for successful treatment. Consequently, over the past few decades, diverse imaging modalities including magnetic resonance (MR), positron emission tomography, computed tomography, and ultrasound (US) have been developed and applied widely in the field of clinical diagnosis. However, each of the above-mentioned imaging modalities possesses unique strengths and intrinsic weaknesses, which limit their abilities to provide accurate information. Therefore, multimodal imaging systems may be a solution that can provide improved diagnostic performance. Among the current medical imaging modalities, US is a widely available real-time imaging modality. It has many advantages including safety, low cost and easy access for patients. However, its low spatial resolution precludes accurate discrimination of diseased region such as cancer sites. In contrast, MR has no tissue-penetrating limit and can provide images possessing exquisite soft tissue contrast and high spatial resolution. However, it cannot offer real-time images and needs a comparatively long imaging time. The characteristics of these imaging modalities may be considered complementary, and the modalities have been frequently combined for the clinical diagnostic process. Biominerals such as calcium carbonate (CaCO3) and calcium phosphate (CaP) exhibit pH-dependent dissolution behavior. They demonstrate pH-controlled drug release due to the dissolution of minerals in acidic pH conditions. In particular, the application of this mineralization technique to a US contrast agent has been reported recently. The CaCO3 mineral reacts with acids and decomposes to generate calcium dioxide (CO2) gas in an acidic environment. These gas-generating mineralized nanoparticles generated CO2 bubbles in the acidic environment of the tumor, thereby allowing for strong echogenic US imaging of tumor tissues. On the basis of this previous work, it was hypothesized that the loading of MR contrast agents into the CaCO3 mineralized nanoparticles may be a novel strategy in designing a contrast agent for dual imaging. Herein, CaCO3 mineralized nanoparticles that were capable of generating CO2 bubbles to trigger the release of entrapped MR contrast agents in response to tumoral acidic pH were developed for the purposes of US and MR dual-modality imaging of tumors. Gd2O3 nanoparticles were selected as an MR contrast agent. A key strategy employed in this study was to prepare Gd2O3 nanoparticle-loaded mineralized nanoparticles (Gd2O3-MNPs) using block copolymer-templated CaCO3 mineralization in the presence of calcium cations (Ca2+), carbonate anions (CO32-) and positively charged Gd2O3 nanoparticles. The CaCO3 core was considered suitable because it may effectively shield Gd2O3 nanoparticles from water molecules in the blood (pH 7.4) before decomposing to generate CO2 gas, triggering the release of Gd2O3 nanoparticles in tumor tissues (pH 6.4~7.4). The kinetics of CaCO3 dissolution and CO2 generation from the Gd2O3-MNPs were examined as a function of pH and pH-dependent in vitro magnetic relaxation; additionally, the echogenic properties were estimated to demonstrate the potential of the particles for the tumor-specific US and MR imaging.

Keywords: calcium carbonate, mineralization, ultrasound imaging, magnetic resonance imaging

Procedia PDF Downloads 218

Authors: Mohamad Ahrari, Kayvan Sadri, Mahmoud Reza Jafari

Abstract:

PEGylated liposomes have a smaller volume of distribution and decreased clearance, consequently, due to their more prolonged presence in bloodstream and maintaining their stability during this period, these liposomes can be applied for imaging tumoral sites. The purpose of this study is to develop an appropriate radiopharmaceutical agent in long-term imaging for improved diagnosis and evaluation of tumors. In this study, liposomal formulations encapsulating albumin is synthesized by solvent evaporation method along with homogenization, and their characteristics were assessed. Then these liposomes labeled by Philips method and the rate of stability of labeled liposomes in serum, and ultimately the rate of biodistribution and gamma scintigraphy in C26-colon carcinoma tumor-bearing mice, were studied. The result of the study of liposomal characteristics displayed that capable of accumulating in tumor sites based of EPR phenomenon. these liposomes also have high stability for maintaining encapsulated albumin in a long time. In the study of biodistribution of these liposomes in mice, they accumulated more in the kidney, liver, spleen, and tumor sites, which, even after clearing formulations in the bloodstream, they existed in high levels in these organs up to 96 hours. In gamma scintigraphy also, organs with high activity accumulation from early hours up to 96 hours were visible in the form of hot spots. concluded that PEGylated liposomal formulation encapsulating albumin can be labeled with In-Oxine, and obtained stabilized formulation for long-term imaging, that have more favorable conditions for the evaluation of tumors and it will cause early diagnosis of tumors.

Keywords: nano liposome, 111In-oxine, imaging, biodistribution, tumor

Procedia PDF Downloads 89

Authors: Shunzuo Wu, Xudong Luo, Yuanxiu Liao

Abstract:

Recently, the rapid development of deep learning makes artificial intelligence (AI) penetrate into many fields, replacing manual work there. In particular, AI systems also become a research focus in the field of automatic office. To meet real needs in automatic officiating, in this paper we develop an automatic form filling system. Specifically, it uses two classical neural network models and several word embedding models to classify various relevant information elicited from the Internet. When training the neural network models, we use less noisy and balanced data for training. We conduct a series of experiments to test my systems and the results show that our system can achieve better classification results.

Keywords: artificial intelligence and office, NLP, deep learning, text classification

Procedia PDF Downloads 167

Authors: T. Zaki, Fathi S. Soliman

Abstract:

Desulfurization process is required by most, if not all refineries, to achieve ultra-low sulfur fuel, that contains less than 10 ppm sulfur. A lot of research works and many effective technologies have been studied to achieve deep desulfurization process in moderate reaction environment, such as adsorption desulfurization (ADS), oxidative desulfurization (ODS), biodesulfurization and extraction desulfurization (EDS). Extraction desulfurization using deep eutectic solvents (DESs) is considered as simple, cheap, highly efficient and environmentally friend process. In this work, four DESs were designed and synthesized. Choline chloride (ChCl) was selected as typical hydrogen bond acceptors (HBA), and ethylene glycol (EG), glycerol (Gl), urea (Ur) and thiourea (Tu) were selected as hydrogen bond donors (HBD), from which a series of deep eutectic solvents were synthesized. The experimental data showed that the synthesized DESs showed desulfurization affinities towards the thiophene species in cyclohexane solvent. Ethylene glycol molecules showed more affinity to create hydrogen bond with thiophene instead of choline chloride. Accordingly, ethylene glycol choline chloride DES has the highest extraction efficiency.

Keywords: DES, desulfurization, green solvent, extraction

Procedia PDF Downloads 257

Authors: Gizem Eser Erdek

Abstract:

This study is research on predicting the remaining life of industrial cutting tools used in the industrial production process with deep learning methods. When the life of cutting tools decreases, they cause destruction to the raw material they are processing. This study it is aimed to predict the remaining life of the cutting tool based on the damage caused by the cutting tools to the raw material. For this, hole photos were collected from the hole-drilling machine for 8 months. Photos were labeled in 5 classes according to hole quality. In this way, the problem was transformed into a classification problem. Using the prepared data set, a model was created with convolutional neural networks, which is a deep learning method. In addition, VGGNet and ResNet architectures, which have been successful in the literature, have been tested on the data set. A hybrid model using convolutional neural networks and support vector machines is also used for comparison. When all models are compared, it has been determined that the model in which convolutional neural networks are used gives successful results of a %74 accuracy rate. In the preliminary studies, the data set was arranged to include only the best and worst classes, and the study gave ~93% accuracy when the binary classification model was applied. The results of this study showed that the remaining life of the cutting tools could be predicted by deep learning methods based on the damage to the raw material. Experiments have proven that deep learning methods can be used as an alternative for cutting tool life estimation.

Keywords: classification, convolutional neural network, deep learning, remaining life of industrial cutting tools, ResNet, support vector machine, VggNet

Procedia PDF Downloads 56

Authors: M. Zehtabian, Z. Molaiemanesh, Z. Shafahi, M. Papie, M. Zahraie Moghaddam, M. Mehralizadeh, M. R. Vahidi, S. Sina

Abstract:

The introduction of CT scans has been a great improvement in diagnosis of different diseases. However, this imaging modality can expose the patients to cumulative radiation doses which may increase the risks of some health problems like cancer. In this study, the dose delivered to different organs in lumbar-abdominal imaging was measured by putting the TLD-100, and TLD-100H chips inside the Alderson Rando phantom. The lumbar-abdominal image of the phantom was obtained, while TLD chips were inside the holes of the phantom. According to the results obtained in this study using TLD-100 chips, the average dose received by liver, bladder, rectum, kidneys, and uterus were found to be 12.9 mSv, 8.9 mSv, 10.1 mSv, 11.0 mSv, 11.2 mSv, and 10.5 mSv respectively, while the measurements performed by TLD-100H show that the average dose to liver, bladder, rectum, kidneys, and uterus were found to be 12.4 mSv, 9.2 mSv, 9.5 mSv, 10.5 mSv, 10.7 mSv, and 9.9 mSv respectively. The results of this study indicates that the dose measured by the TLD-100H chips are in close agreement with those obtained by TLD-100.

Keywords: CT scan, dose, TLD-100, diagnosis

Procedia PDF Downloads 609

Authors: Chia-Yu Lin, Yu-Hsiang Shen, Cing-Ciao Ke, Chia-Hao Chang, Fan-Pin Tseng, Yu-Ching Ni, Sheng-Pin Tseng

Abstract:

Tomosynthesis is an imaging system that generates a 3D image by scanning in a limited angular range. It could provide more depth information than traditional 2D X-ray single projection. Radiation dose in tomosynthesis is less than computed tomography (CT). Because of limited angular range scanning, there are many properties depending on scanning direction. Therefore, non-coplanar imaging technique was developed to improve image quality in traditional tomosynthesis. The purpose of this study was to establish the non-coplanar imaging technique of tomosynthesis system and evaluate this technique by the reconstructed image. INER prototype tomosynthesis system contains an X-ray tube, a flat panel detector, and a motion machine. This system could move X-ray tube in multiple directions during the acquisition. In this study, we investigated three different imaging techniques that were 2D X-ray single projection, traditional tomosynthesis, and non-coplanar tomosynthesis. An anthropopathic chest phantom was used to evaluate the image quality. It contained three different size lesions (3 mm, 5 mm and, 8 mm diameter). The traditional tomosynthesis acquired 61 projections over a 30 degrees angular range in one scanning direction. The non-coplanar tomosynthesis acquired 62 projections over 30 degrees angular range in two scanning directions. A 3D image was reconstructed by iterative image reconstruction algorithm (ML-EM). Our qualitative method was to evaluate artifacts in tomosynthesis reconstructed image. The quantitative method was used to calculate a peak-to-valley ratio (PVR) that means the intensity ratio of the lesion to the background. We used PVRs to evaluate the contrast of lesions. The qualitative results showed that in the reconstructed image of non-coplanar scanning, anatomic structures of chest and lesions could be identified clearly and no significant artifacts of scanning direction dependent could be discovered. In 2D X-ray single projection, anatomic structures overlapped and lesions could not be discovered. In traditional tomosynthesis image, anatomic structures and lesions could be identified clearly, but there were many artifacts of scanning direction dependent. The quantitative results of PVRs show that there were no significant differences between non-coplanar tomosynthesis and traditional tomosynthesis. The PVRs of the non-coplanar technique were slightly higher than traditional technique in 5 mm and 8 mm lesions. In non-coplanar tomosynthesis, artifacts of scanning direction dependent could be reduced and PVRs of lesions were not decreased. The reconstructed image was more isotropic uniformity in non-coplanar tomosynthesis than in traditional tomosynthesis. In the future, scan strategy and scan time will be the challenges of non-coplanar imaging technique.

Keywords: image reconstruction, non-coplanar imaging technique, tomosynthesis, X-ray imaging

Procedia PDF Downloads 346

Authors: Auwal Abubakar, Ahmed Ahidjo, Shazril Imran Shaukat, Noor Khairiah A. Karim, Gokula Kumar Appalanaido, Hafiz Mohd Zin

Abstract:

Background: The use of optical surface imaging systems (OSISs) is increasingly becoming popular in radiotherapy practice, especially during breast cancer treatment. This study reviews the accuracy of the available commercial OSISs for breast radiotherapy. Method: A literature search was conducted and identified the available commercial OSISs from different manufacturers that are integrated into radiotherapy practice for setup verification during breast radiotherapy. Studies that evaluated the accuracy of the OSISs during breast radiotherapy using cone beam computed tomography (CBCT) as a reference were retrieved and analyzed. The physics and working principles of the systems from each manufacturer were discussed together with their respective strength and limitations. Results: A total of five (5) different commercially available OSISs from four (4) manufacturers were identified, each with a different working principle. Six (6) studies were found to evaluate the accuracy of the systems during breast radiotherapy in conjunction with CBCT as a goal standard. The studies revealed that the accuracy of the system in terms of mean difference ranges from 0.1 to 2.1 mm. The correlation between CBCT and OSIS ranges between 0.4 and 0.9. The limit of agreements obtained using bland Altman analysis in the studies was also within an acceptable range. Conclusion: The OSISs have an acceptable level of accuracy and could be used safely during breast radiotherapy. The systems are non-invasive, ionizing radiation-free, and provide real-time imaging of the target surface at no extra concomitant imaging dose. However, the system should only be used to complement rather than replace x-ray-based image guidance techniques such as CBCT.

Keywords: optical surface imaging system, Cone beam computed tomography (CBCT), surface guided radiotherapy, Breast radiotherapy

Procedia PDF Downloads 38

Authors: M. G. R. S. Perera, B. S. Weerakoon, L. P. G. Sherminie, M. L. Jayatilake, R. D. Jayasinghe, W. Huang

Abstract:

The aim of this study was to measure and evaluate the Longitudinal Relaxation Times (T1) in bone marrow of an Acute Myeloid Leukaemia (AML) patient in order to explore the potential for a prognostic biomarker using Magnetic Resonance Imaging (MRI) which will be a non-invasive prognostic approach to AML. MR image data were collected in the DICOM format and MATLAB Simulink software was used in the image processing and data analysis. For quantitative MRI data analysis, Region of Interests (ROI) on multiple image slices were drawn encompassing vertebral bodies of L3, L4, and L5. T1 was evaluated using the T1 maps obtained. The estimated bone marrow mean value of T1 was 790.1 (ms) at 3T. However, the reported T1 value of healthy subjects is significantly (946.0 ms) higher than the present finding. This suggests that the T1 for bone marrow can be considered as a potential prognostic biomarker for AML patients.

Keywords: acute myeloid leukaemia, longitudinal relaxation time, magnetic resonance imaging, prognostic biomarker.

Procedia PDF Downloads 507

Authors: Sameer Masood, Lucas Chartier

Abstract:

Head injuries are a commonly encountered presentation in emergency departments (ED) and the Choosing Wisely Canada (CWC) campaign was released in June 2015 in an attempt to decrease imaging utilization for patients with minor head injuries. The impact of the CWC campaign on imaging utilization for head injuries has not been explored in the ED setting. In our study, we describe the characteristics of patients with head injuries presenting to a tertiary care academic ED and the impact of the CWC campaign on CT head utilization. This retrospective cohort study used linked databases from the province of Ontario, Canada to assess emergency department visits with a primary diagnosis of head injury made between June 1, 2014 and Aug 31, 2016 at the University Health Network in Toronto, Canada. We examined the number of visits during the study period, the proportion of patients that had a CT head performed before and after the release of the CWC campaign, as well as mode of arrival, and disposition. There were 4,322 qualifying visits at our site during the study period. The median presenting age was 44.12 years (IQR 27.83,67.45), the median GCS was 15 (IQR 15,15) and the majority of patients presenting had intermediate acuity (CTAS 3). Overall, 43.17% of patients arrived via ambulance, 49.24 % of patients received a CT head and 10.46% of patients were admitted. Compared to patients presenting before the CWC campaign release, there was no significant difference in the rate of CT heads after the CWC (50.41% vs 47.68%, P = 0.07). There were also no significant differences between the two groups in mode of arrival (ambulance vs ambulatory) (42.94% vs 43.48%, P = 0.72) or admission rates (9.85% vs 11.26%, P = 0.15). However, more patients belonged to the high acuity groups (CTAS 1 or 2) in the post CWC campaign release group (12.98% vs 8.11% P <0.001). Visits for head injuries make up a significant proportion of total ED visits and approximately half of these patients receive CT imaging in the ED. The CWC campaign did not seem to impact imaging utilization for head injuries in the 14 months following its launch. Further efforts, including local quality improvement initiatives, are likely needed to increase adherence to its recommendation and reduce imaging utilization for head injuries.

Keywords: choosing wisely, emergency department, head injury, quality improvement

Procedia PDF Downloads 204

Authors: Sung-Hyeon Park, Jin-Wook Jang, Hak-Sung Kim

Abstract:

In recent years, Terahertz (THz) time-domain spectroscopy (TDS) imaging method has been received considerable interest as a promising non-destructive technique for detection of internal defects. In comparison to other non-destructive techniques such as x-ray inspection method, scanning acoustic tomograph (SAT) and microwave inspection method, THz-TDS imaging method has many advantages: First, it can measure the exact thickness and location of defects. Second, it doesn’t require the liquid couplant while it is very crucial to deliver that power of ultrasonic wave in SAT method. Third, it didn’t damage to materials and be harmful to human bodies while x-ray inspection method does. Finally, it exhibits better spatial resolution than microwave inspection method. However, this technology couldn’t be applied to IC package because THz radiation can penetrate through a wide variety of materials including polymers and ceramics except of metals. Therefore, it is difficult to detect the defects in IC package which are composed of not only epoxy and semiconductor materials but also various metals such as copper, aluminum and gold. In this work, we proposed a special method for detecting the void in the IC package using THz-TDS imaging system. The IC package specimens for this study are prepared by Packaging Engineering Team in Samsung Electronics. Our THz-TDS imaging system has a special reflection mode called pitch-catch mode which can change the incidence angle in the reflection mode from 10 o to 70 o while the others have transmission and the normal reflection mode or the reflection mode fixed at certain angle. Therefore, to find the voids in the IC package, we investigated the appropriate angle as changing the incidence angle of THz wave emitter and detector. As the results, the voids in the IC packages were successfully detected using our THz-TDS imaging system.

Keywords: terahertz, non-destructive technique, void, IC package

Procedia PDF Downloads 459

Authors: Mahdi Farajzadeh Ajirlou

Abstract:

Ordinary fetal brain development can be considered by in vivo attractive reverberation imaging (MRI) from the 18th gestational week (GW) to term and depends fundamentally on T2-weighted and diffusion-weighted (DW) arrangements. The foremost commonly suspected brain pathologies alluded to fetal MRI for assist assessment are ventriculomegaly, lost corpus callosum, and anomalies of the posterior fossa. Brain division could be a crucial to begin with step in neuroimage examination. Within the case of fetal MRI it is especially challenging and critical due to the subjective introduction of the hatchling, organs that encompass the fetal head, and irregular fetal movement. A few promising strategies have been proposed but are constrained in their execution in challenging cases and in realtime division. Fetal MRI is routinely performed on a 1.5-Tesla scanner without maternal or fetal sedation. The mother lies recumbent amid the course of the examination, the length of which is ordinarily 45 to 60 minutes. The accessibility and continuous approval of standardizing fetal brain development directions will give critical devices for early discovery of impeded fetal brain development upon which to oversee high-risk pregnancies.

Keywords: brain, fetal, MRI, imaging

Procedia PDF Downloads 55

Authors: Dingyang Hu, Dan Liu

Abstract:

DNN (Deep Neural Network) deep learning models are widely used in classification, prediction, and other task scenarios. To address the difficulties of generic adversarial perturbation generation for deep learning models under black-box conditions, a generic adversarial ingestion generation method based on a saliency map (CJsp) is proposed to obtain salient image regions by counting the factors that influence the input features of an image on the output results. This method can be understood as a saliency map attack algorithm to obtain false classification results by reducing the weights of salient feature points. Experiments also demonstrate that this method can obtain a high success rate of migration attacks and is a batch adversarial sample generation method.

Keywords: adversarial sample, gradient, probability, black box

Procedia PDF Downloads 72

Authors: Ons Slimene, Aroua Taamallah, Maha Khemaja

Abstract:

In recent years, human activity recognition (HAR) has been a key area of research due to its diverse applications. It has garnered increasing attention in the field of computer vision. HAR plays an important role in people’s daily lives as it has the ability to learn advanced knowledge about human activities from data. In HAR, activities are usually represented by exploiting different types of sensors, such as embedded sensors or visual sensors. However, these sensors have limitations, such as local obstacles, image-related obstacles, sensor unreliability, and consumer concerns. Recently, several deep learning-based approaches have been proposed for HAR and these approaches are classified into two categories based on the type of data used: vision-based approaches and sensor-based approaches. This research paper highlights the importance of multimodal data fusion from skeleton data obtained from videos and data generated by embedded sensors using deep neural networks for achieving HAR. We propose a deep multimodal fusion network based on a twostream architecture. These two streams use the Convolutional Neural Network combined with the Bidirectional LSTM (CNN BILSTM) to process skeleton data and data generated by embedded sensors and the fusion at the feature level is considered. The proposed model was evaluated on a public OPPORTUNITY++ dataset and produced a accuracy of 96.77%.

Keywords: human activity recognition, action recognition, sensors, vision, human-centric sensing, deep learning, context-awareness

Procedia PDF Downloads 78

Authors: Kozo Tanigawa, Tappei Yotsumoto, Kenichi Takahashi, Takao Kawamura, Kazunori Sugahara

Abstract:

An automatic human tracking system using mobile agent technology is realized because a mobile agent moves in accordance with a migration of a target person. In this paper, we propose a method for determining the neighbor node in consideration of the imaging range of cameras.

Keywords: human tracking, mobile agent, Pan/Tilt/Zoom, neighbor relation

Procedia PDF Downloads 498

Authors: Yuanqing Li, Changyi Lei, Zhaopeng Xue, Zhuo Zheng, Yanbo Long

Abstract:

In agricultural production and breeding, implementing automatic picking robot in orchard farming to reduce human labour and error is challenging. The core function of it is automatic identification based on machine vision. This paper focuses on apple detection and counting in orchards and implements several deep learning methods. Extensive datasets are used and a semi-automatic annotation method is proposed. The proposed deep learning models are in state-of-the-art YOLO family. In view of the essence of the models with various backbones, a multi-dimensional comparison in details is made in terms of counting accuracy, mAP and model memory, laying the foundation for realising automatic precision agriculture.

Keywords: agricultural object detection, deep learning, machine vision, YOLO family

Procedia PDF Downloads 175

Authors: Florian Kleber, Martin Kampel

Abstract:

The increasing demand of gallium, indium and rare-earth elements for the production of electronics, e.g. solid state-lighting, photovoltaics, integrated circuits, and liquid crystal displays, will exceed the world-wide supply according to current forecasts. Recycling systems to reclaim these materials are not yet in place, which challenges the sustainability of these technologies. This paper proposes a multispectral imaging system as a basis for a vision based recognition system for valuable components of electronics waste. Multispectral images intend to enhance the contrast of images of printed circuit boards (single components, as well as labels) for further analysis, such as optical character recognition and entire printed circuit board recognition. The results show that a higher contrast is achieved in the near infrared compared to ultraviolet and visible light.

Keywords: electronics waste, multispectral imaging, printed circuit boards, rare-earth elements

Procedia PDF Downloads 402

Authors: Tayeb Aissaoui, Inas M. AlNashef

Abstract:

In recent years, green solvents named deep eutectic solvents (DESs) have been found to possess significant properties and to be applicable in several technologies. Choline chloride (ChCl) mixed with urea at a ratio of 1:2 and 80 °C was the first discovered DES. In this article, chemical structure and combination mechanism of ChCl: urea based DES were investigated. Moreover, the implementation of this DES in water removal from natural gas was reported. Dehydration of natural gas by ChCl:urea shows significant absorption efficiency compared to triethylene glycol. All above operations were retrieved from COSMOthermX software. This article confirms the potential application of DESs in gas industry.

Keywords: COSMO-RS, deep eutectic solvents, dehydration, natural gas, structure, organic salt

Procedia PDF Downloads 270

Authors: Angelis P. Barlampas

Abstract:

The purpose of this study is to inform the reader, about the various applications of artificial intelligence (AI), in cardiac imaging. AI grows fast and its role is crucial in medical specialties, which use large amounts of digital data, that are very difficult or even impossible to be managed by human beings and especially doctors.Artificial intelligence (AI) refers to the ability of computers to mimic human cognitive function, performing tasks such as learning, problem-solving, and autonomous decision making based on digital data. Whereas AI describes the concept of using computers to mimic human cognitive tasks, machine learning (ML) describes the category of algorithms that enable most current applications described as AI. Some of the current applications of AI in cardiac imaging are the follows: Ultrasound: Automated segmentation of cardiac chambers across five common views and consequently quantify chamber volumes/mass, ascertain ejection fraction and determine longitudinal strain through speckle tracking. Determine the severity of mitral regurgitation (accuracy > 99% for every degree of severity). Identify myocardial infarction. Distinguish between Athlete’s heart and hypertrophic cardiomyopathy, as well as restrictive cardiomyopathy and constrictive pericarditis. Predict all-cause mortality. CT Reduce radiation doses. Calculate the calcium score. Diagnose coronary artery disease (CAD). Predict all-cause 5-year mortality. Predict major cardiovascular events in patients with suspected CAD. MRI Segment of cardiac structures and infarct tissue. Calculate cardiac mass and function parameters. Distinguish between patients with myocardial infarction and control subjects. It could potentially reduce costs since it would preclude the need for gadolinium-enhanced CMR. Predict 4-year survival in patients with pulmonary hypertension. Nuclear Imaging Classify normal and abnormal myocardium in CAD. Detect locations with abnormal myocardium. Predict cardiac death. ML was comparable to or better than two experienced readers in predicting the need for revascularization. AI emerge as a helpful tool in cardiac imaging and for the doctors who can not manage the overall increasing demand, in examinations such as ultrasound, computed tomography, MRI, or nuclear imaging studies.

Keywords: artificial intelligence, cardiac imaging, ultrasound, MRI, CT, nuclear medicine

Procedia PDF Downloads 56