Search results for: convolutional recurrent networks
3081 1D Convolutional Networks to Compute Mel-Spectrogram, Chromagram, and Cochleogram for Audio Networks
Authors: Elias Nemer, Greg Vines
Abstract:
Time-frequency transformation and spectral representations of audio signals are commonly used in various machine learning applications. Training networks on frequency features such as the Mel-Spectrogram or Cochleogram have been proven more effective and convenient than training on-time samples. In practical realizations, these features are created on a different processor and/or pre-computed and stored on disk, requiring additional efforts and making it difficult to experiment with different features. In this paper, we provide a PyTorch framework for creating various spectral features as well as time-frequency transformation and time-domain filter-banks using the built-in trainable conv1d() layer. This allows computing these features on the fly as part of a larger network and enabling easier experimentation with various combinations and parameters. Our work extends the work in the literature developed for that end: First, by adding more of these features and also by allowing the possibility of either starting from initialized kernels or training them from random values. The code is written as a template of classes and scripts that users may integrate into their own PyTorch classes or simply use as is and add more layers for various applications.Keywords: neural networks Mel-Spectrogram, chromagram, cochleogram, discrete Fourrier transform, PyTorch conv1d()
Procedia PDF Downloads 2253080 Sign Language Recognition of Static Gestures Using Kinect™ and Convolutional Neural Networks
Authors: Rohit Semwal, Shivam Arora, Saurav, Sangita Roy
Abstract:
This work proposes a supervised framework with deep convolutional neural networks (CNNs) for vision-based sign language recognition of static gestures. Our approach addresses the acquisition and segmentation of correct inputs for the CNN-based classifier. Microsoft Kinect™ sensor, despite complex environmental conditions, can track hands efficiently. Skin Colour based segmentation is applied on cropped images of hands in different poses, used to depict different sign language gestures. The segmented hand images are used as an input for our classifier. The CNN classifier proposed in the paper is able to classify the input images with a high degree of accuracy. The system was trained and tested on 39 static sign language gestures, including 26 letters of the alphabet and 13 commonly used words. This paper includes a problem definition for building the proposed system, which acts as a sign language translator between deaf/mute and the rest of the society. It is then followed by a focus on reviewing existing knowledge in the area and work done by other researchers. It also describes the working principles behind different components of CNNs in brief. The architecture and system design specifications of the proposed system are discussed in the subsequent sections of the paper to give the reader a clear picture of the system in terms of the capability required. The design then gives the top-level details of how the proposed system meets the requirements.Keywords: sign language, CNN, HCI, segmentation
Procedia PDF Downloads 1513079 Validity and Reliability of Lifestyle Measurement of the LSAS among Recurrent Stroke Patients in Selected Hospital, Central Java, Indonesia
Authors: Meida Laely Ramdani, Earmporn Thongkrajai, Dedy Purwito
Abstract:
Lifestyle is one of the most important factors affecting health. Measurement of lifestyle behaviors is necessary for the identification of causal associations between unhealthy lifestyle and health outcomes. There was many instruments have been measured for lifestyle, but not specific for stroke recurrence. This study aimed to develop a new questionnaire of Lifestyle Adjustment Scale (LSAS) among recurrent stroke patients in Indonesia and to measure the reliability and validity of LSAS. The instrument consist of 33 items was developed from the responses of 30 recurrent stroke patients with the maximum age 60 years. Data was collected during October to November 2015. The properties of the instrument were evaluated by validity assessment and reliability measures. The content validity was judged adequate by a panel of five experts, with the result of I-CVI was 0.97. The Cronbach’s alpha analysis was carried out to measure the reliability of LSAS. The result showed that Cronbach’s alpha coefficient was 0.819. LSAS were classified under the domains of dietary habit, smoking habit, physical activity, and stress management. The results of Cronbach’s alpha coefficient for each subscale were 0.60, 0.39, 0.67, 0.65 and 0.76 respectively. LSAS instrument was valid and reliable therefore can be used as research tool among recurrent stroke patients. The development of this questionnaire has been adapted to the socio-cultural context in Indonesia.Keywords: LSAS, recurrent stroke patients, lifestyle, Indonesia
Procedia PDF Downloads 2413078 Prevalence of Autoimmune Thyroid Disease in Recurrent Aphthous Stomatitis
Authors: Arghavan Tonkaboni, Shamsolmolouk Najafi, Mohmmad Taghi Kiani, Mehrzad Gholampour, Touraj Goli
Abstract:
Introduction: Recurrent aphthous stomatitis (RAS) is a multifactorial recurrent oral lesion; which is an autoimmune disease. TH1 cytokines are the most important etiological factors. Autoimmune thyroid disease (ATD) is one of the most common autoimmune diseases and generally coexists with other autoimmune diseases. This study assessed the prevalence of thyroid disease in patients with recurrent aphthous stomatitis. Materials and Methods: This case control study assessed 100 known RAS patients who were diagnosed clinically by oral medicine specialists; venous blood samples were analyzed for thyroid stimulating hormone (TSH), free triiodothyronine (fT3), total thyroxine (fT4), thyroglobulin, anti-thyroid peroxidase antibody (anti-TPO) and anti-thyroglobulin antibody (anti-TG) levels. Results: Fifty patients with RAS aged between 18-42 years (28.5±5.8) and 50 healthy volunteers aged 19-45 years (27.3±5.4) participated. In RAS patients, fT3 and TSH levels were significantly higher (P=0.031, P=0.706); however, fT4 level was lower in the RAS group (P=0.447). Anti TG and anti-TPO levels were significantly higher in the RAS group (P=0.008, P=0.067). Conclusion: Our study showed that ATD prevalence was significantly higher in RAS patients. Based on this study, we recommend assessment of thyroid hormones and antibodies in RAS patients.Keywords: recurrent aphthous stomatitis, thyroid antibodies, thyroid hormone, thyroid autoimmune disease
Procedia PDF Downloads 3383077 Impact of Integrated Signals for Doing Human Activity Recognition Using Deep Learning Models
Authors: Milagros Jaén-Vargas, Javier García Martínez, Karla Miriam Reyes Leiva, María Fernanda Trujillo-Guerrero, Francisco Fernandes, Sérgio Barroso Gonçalves, Miguel Tavares Silva, Daniel Simões Lopes, José Javier Serrano Olmedo
Abstract:
Human Activity Recognition (HAR) is having a growing impact in creating new applications and is responsible for emerging new technologies. Also, the use of wearable sensors is an important key to exploring the human body's behavior when performing activities. Hence, the use of these dispositive is less invasive and the person is more comfortable. In this study, a database that includes three activities is used. The activities were acquired from inertial measurement unit sensors (IMU) and motion capture systems (MOCAP). The main objective is differentiating the performance from four Deep Learning (DL) models: Deep Neural Network (DNN), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN) and hybrid model Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM), when considering acceleration, velocity and position and evaluate if integrating the IMU acceleration to obtain velocity and position represent an increment in performance when it works as input to the DL models. Moreover, compared with the same type of data provided by the MOCAP system. Despite the acceleration data is cleaned when integrating, results show a minimal increase in accuracy for the integrated signals.Keywords: HAR, IMU, MOCAP, acceleration, velocity, position, feature maps
Procedia PDF Downloads 913076 Study of Transformer and Motor Winding under Pulsed Power Application
Authors: Arijit Basuray, Saibal Chatterjee
Abstract:
Pulsed Power in the form of Recurrent Surge Generator (RSG) can be used for testing various parameters of Motor or Transformer windings including inter-turn, interlayer insulation. Windings with solid insulation in motor and transformer have many interfaces and undesirable defects, and these defects can be exposed under this nondestructive testing methodology. Due to rapid development in power electronics variable frequency drives (VFD), Dry Type or cast resin Transformer used with PWM Sine wave inverters for solar power, solid insulation system used nowadays are shifting more and more to a high-frequency application. Authors have used the recurrent surge generator for testing winding integrity as well as Partial Discharge(PD) at fast rising voltage enabling PD measurement at closer situation under which the insulation system is supposed to work. Authors have discussed test results on a different system with recurrent surge voltages of different rise time.Keywords: fast rising voltage, partial discharge, pulsed power, recurrent surge generator, solid insulation
Procedia PDF Downloads 2693075 Convolutional Neural Networks-Optimized Text Recognition with Binary Embeddings for Arabic Expiry Date Recognition
Authors: Mohamed Lotfy, Ghada Soliman
Abstract:
Recognizing Arabic dot-matrix digits is a challenging problem due to the unique characteristics of dot-matrix fonts, such as irregular dot spacing and varying dot sizes. This paper presents an approach for recognizing Arabic digits printed in dot matrix format. The proposed model is based on Convolutional Neural Networks (CNN) that take the dot matrix as input and generate embeddings that are rounded to generate binary representations of the digits. The binary embeddings are then used to perform Optical Character Recognition (OCR) on the digit images. To overcome the challenge of the limited availability of dotted Arabic expiration date images, we developed a True Type Font (TTF) for generating synthetic images of Arabic dot-matrix characters. The model was trained on a synthetic dataset of 3287 images and 658 synthetic images for testing, representing realistic expiration dates from 2019 to 2027 in the format of yyyy/mm/dd. Our model achieved an accuracy of 98.94% on the expiry date recognition with Arabic dot matrix format using fewer parameters and less computational resources than traditional CNN-based models. By investigating and presenting our findings comprehensively, we aim to contribute substantially to the field of OCR and pave the way for advancements in Arabic dot-matrix character recognition. Our proposed approach is not limited to Arabic dot matrix digit recognition but can also be extended to text recognition tasks, such as text classification and sentiment analysis.Keywords: computer vision, pattern recognition, optical character recognition, deep learning
Procedia PDF Downloads 823074 Robust Heart Sounds Segmentation Based on the Variation of the Phonocardiogram Curve Length
Authors: Mecheri Zeid Belmecheri, Maamar Ahfir, Izzet Kale
Abstract:
Automatic cardiac auscultation is still a subject of research in order to establish an objective diagnosis. Recorded heart sounds as Phonocardiogram signals (PCG) can be used for automatic segmentation into components that have clinical meanings. These are the first sound, S1, the second sound, S2, and the systolic and diastolic components, respectively. In this paper, an automatic method is proposed for the robust segmentation of heart sounds. This method is based on calculating an intermediate sawtooth-shaped signal from the length variation of the recorded Phonocardiogram (PCG) signal in the time domain and, using its positive derivative function that is a binary signal in training a Recurrent Neural Network (RNN). Results obtained in the context of a large database of recorded PCGs with their simultaneously recorded ElectroCardioGrams (ECGs) from different patients in clinical settings, including normal and abnormal subjects, show a segmentation testing performance average of 76 % sensitivity and 94 % specificity.Keywords: heart sounds, PCG segmentation, event detection, recurrent neural networks, PCG curve length
Procedia PDF Downloads 1753073 Breast Cancer Metastasis Detection and Localization through Transfer-Learning Convolutional Neural Network Classification Based on Convolutional Denoising Autoencoder Stack
Authors: Varun Agarwal
Abstract:
Introduction: With the advent of personalized medicine, histopathological review of whole slide images (WSIs) for cancer diagnosis presents an exceedingly time-consuming, complex task. Specifically, detecting metastatic regions in WSIs of sentinel lymph node biopsies necessitates a full-scanned, holistic evaluation of the image. Thus, digital pathology, low-level image manipulation algorithms, and machine learning provide significant advancements in improving the efficiency and accuracy of WSI analysis. Using Camelyon16 data, this paper proposes a deep learning pipeline to automate and ameliorate breast cancer metastasis localization and WSI classification. Methodology: The model broadly follows five stages -region of interest detection, WSI partitioning into image tiles, convolutional neural network (CNN) image-segment classifications, probabilistic mapping of tumor localizations, and further processing for whole WSI classification. Transfer learning is applied to the task, with the implementation of Inception-ResNetV2 - an effective CNN classifier that uses residual connections to enhance feature representation, adding convolved outputs in the inception unit to the proceeding input data. Moreover, in order to augment the performance of the transfer learning CNN, a stack of convolutional denoising autoencoders (CDAE) is applied to produce embeddings that enrich image representation. Through a saliency-detection algorithm, visual training segments are generated, which are then processed through a denoising autoencoder -primarily consisting of convolutional, leaky rectified linear unit, and batch normalization layers- and subsequently a contrast-normalization function. A spatial pyramid pooling algorithm extracts the key features from the processed image, creating a viable feature map for the CNN that minimizes spatial resolution and noise. Results and Conclusion: The simplified and effective architecture of the fine-tuned transfer learning Inception-ResNetV2 network enhanced with the CDAE stack yields state of the art performance in WSI classification and tumor localization, achieving AUC scores of 0.947 and 0.753, respectively. The convolutional feature retention and compilation with the residual connections to inception units synergized with the input denoising algorithm enable the pipeline to serve as an effective, efficient tool in the histopathological review of WSIs.Keywords: breast cancer, convolutional neural networks, metastasis mapping, whole slide images
Procedia PDF Downloads 1253072 Analyzing Large Scale Recurrent Event Data with a Divide-And-Conquer Approach
Authors: Jerry Q. Cheng
Abstract:
Currently, in analyzing large-scale recurrent event data, there are many challenges such as memory limitations, unscalable computing time, etc. In this research, a divide-and-conquer method is proposed using parametric frailty models. Specifically, the data is randomly divided into many subsets, and the maximum likelihood estimator from each individual data set is obtained. Then a weighted method is proposed to combine these individual estimators as the final estimator. It is shown that this divide-and-conquer estimator is asymptotically equivalent to the estimator based on the full data. Simulation studies are conducted to demonstrate the performance of this proposed method. This approach is applied to a large real dataset of repeated heart failure hospitalizations.Keywords: big data analytics, divide-and-conquer, recurrent event data, statistical computing
Procedia PDF Downloads 1593071 An Empirical Study on Switching Activation Functions in Shallow and Deep Neural Networks
Authors: Apoorva Vinod, Archana Mathur, Snehanshu Saha
Abstract:
Though there exists a plethora of Activation Functions (AFs) used in single and multiple hidden layer Neural Networks (NN), their behavior always raised curiosity, whether used in combination or singly. The popular AFs –Sigmoid, ReLU, and Tanh–have performed prominently well for shallow and deep architectures. Most of the time, AFs are used singly in multi-layered NN, and, to the best of our knowledge, their performance is never studied and analyzed deeply when used in combination. In this manuscript, we experiment with multi-layered NN architecture (both on shallow and deep architectures; Convolutional NN and VGG16) and investigate how well the network responds to using two different AFs (Sigmoid-Tanh, Tanh-ReLU, ReLU-Sigmoid) used alternately against a traditional, single (Sigmoid-Sigmoid, Tanh-Tanh, ReLUReLU) combination. Our results show that using two different AFs, the network achieves better accuracy, substantially lower loss, and faster convergence on 4 computer vision (CV) and 15 Non-CV (NCV) datasets. When using different AFs, not only was the accuracy greater by 6-7%, but we also accomplished convergence twice as fast. We present a case study to investigate the probability of networks suffering vanishing and exploding gradients when using two different AFs. Additionally, we theoretically showed that a composition of two or more AFs satisfies Universal Approximation Theorem (UAT).Keywords: activation function, universal approximation function, neural networks, convergence
Procedia PDF Downloads 1493070 Studying Relationship between Local Geometry of Decision Boundary with Network Complexity for Robustness Analysis with Adversarial Perturbations
Authors: Tushar K. Routh
Abstract:
If inputs are engineered in certain manners, they can influence deep neural networks’ (DNN) performances by facilitating misclassifications, a phenomenon well-known as adversarial attacks that question networks’ vulnerability. Recent studies have unfolded the relationship between vulnerability of such networks with their complexity. In this paper, the distinctive influence of additional convolutional layers at the decision boundaries of several DNN architectures was investigated. Here, to engineer inputs from widely known image datasets like MNIST, Fashion MNIST, and Cifar 10, we have exercised One Step Spectral Attack (OSSA) and Fast Gradient Method (FGM) techniques. The aftermaths of adding layers to the robustness of the architectures have been analyzed. For reasoning, separation width from linear class partitions and local geometry (curvature) near the decision boundary have been examined. The result reveals that model complexity has significant roles in adjusting relative distances from margins, as well as the local features of decision boundaries, which impact robustness.Keywords: DNN robustness, decision boundary, local curvature, network complexity
Procedia PDF Downloads 693069 Slice Bispectrogram Analysis-Based Classification of Environmental Sounds Using Convolutional Neural Network
Authors: Katsumi Hirata
Abstract:
Certain systems can function well only if they recognize the sound environment as humans do. In this research, we focus on sound classification by adopting a convolutional neural network and aim to develop a method that automatically classifies various environmental sounds. Although the neural network is a powerful technique, the performance depends on the type of input data. Therefore, we propose an approach via a slice bispectrogram, which is a third-order spectrogram and is a slice version of the amplitude for the short-time bispectrum. This paper explains the slice bispectrogram and discusses the effectiveness of the derived method by evaluating the experimental results using the ESC‑50 sound dataset. As a result, the proposed scheme gives high accuracy and stability. Furthermore, some relationship between the accuracy and non-Gaussianity of sound signals was confirmed.Keywords: environmental sound, bispectrum, spectrogram, slice bispectrogram, convolutional neural network
Procedia PDF Downloads 1213068 Transformation of Positron Emission Tomography Raw Data into Images for Classification Using Convolutional Neural Network
Authors: Paweł Konieczka, Lech Raczyński, Wojciech Wiślicki, Oleksandr Fedoruk, Konrad Klimaszewski, Przemysław Kopka, Wojciech Krzemień, Roman Shopa, Jakub Baran, Aurélien Coussat, Neha Chug, Catalina Curceanu, Eryk Czerwiński, Meysam Dadgar, Kamil Dulski, Aleksander Gajos, Beatrix C. Hiesmayr, Krzysztof Kacprzak, łukasz Kapłon, Grzegorz Korcyl, Tomasz Kozik, Deepak Kumar, Szymon Niedźwiecki, Dominik Panek, Szymon Parzych, Elena Pérez Del Río, Sushil Sharma, Shivani Shivani, Magdalena Skurzok, Ewa łucja Stępień, Faranak Tayefi, Paweł Moskal
Abstract:
This paper develops the transformation of non-image data into 2-dimensional matrices, as a preparation stage for classification based on convolutional neural networks (CNNs). In positron emission tomography (PET) studies, CNN may be applied directly to the reconstructed distribution of radioactive tracers injected into the patient's body, as a pattern recognition tool. Nonetheless, much PET data still exists in non-image format and this fact opens a question on whether they can be used for training CNN. In this contribution, the main focus of this paper is the problem of processing vectors with a small number of features in comparison to the number of pixels in the output images. The proposed methodology was applied to the classification of PET coincidence events.Keywords: convolutional neural network, kernel principal component analysis, medical imaging, positron emission tomography
Procedia PDF Downloads 1373067 Non-intrusive Hand Control of Drone Using an Inexpensive and Streamlined Convolutional Neural Network Approach
Authors: Evan Lowhorn, Rocio Alba-Flores
Abstract:
The purpose of this work is to develop a method for classifying hand signals and using the output in a drone control algorithm. To achieve this, methods based on Convolutional Neural Networks (CNN) were applied. CNN's are a subset of deep learning, which allows grid-like inputs to be processed and passed through a neural network to be trained for classification. This type of neural network allows for classification via imaging, which is less intrusive than previous methods using biosensors, such as EMG sensors. Classification CNN's operate purely from the pixel values in an image; therefore they can be used without additional exteroceptive sensors. A development bench was constructed using a desktop computer connected to a high-definition webcam mounted on a scissor arm. This allowed the camera to be pointed downwards at the desk to provide a constant solid background for the dataset and a clear detection area for the user. A MATLAB script was created to automate dataset image capture at the development bench and save the images to the desktop. This allowed the user to create their own dataset of 12,000 images within three hours. These images were evenly distributed among seven classes. The defined classes include forward, backward, left, right, idle, and land. The drone has a popular flip function which was also included as an additional class. To simplify control, the corresponding hand signals chosen were the numerical hand signs for one through five for movements, a fist for land, and the universal “ok” sign for the flip command. Transfer learning with PyTorch (Python) was performed using a pre-trained 18-layer residual learning network (ResNet-18) to retrain the network for custom classification. An algorithm was created to interpret the classification and send encoded messages to a Ryze Tello drone over its 2.4 GHz Wi-Fi connection. The drone’s movements were performed in half-meter distance increments at a constant speed. When combined with the drone control algorithm, the classification performed as desired with negligible latency when compared to the delay in the drone’s movement commands.Keywords: classification, computer vision, convolutional neural networks, drone control
Procedia PDF Downloads 2063066 Recurrent Neural Networks with Deep Hierarchical Mixed Structures for Chinese Document Classification
Authors: Zhaoxin Luo, Michael Zhu
Abstract:
In natural languages, there are always complex semantic hierarchies. Obtaining the feature representation based on these complex semantic hierarchies becomes the key to the success of the model. Several RNN models have recently been proposed to use latent indicators to obtain the hierarchical structure of documents. However, the model that only uses a single-layer latent indicator cannot achieve the true hierarchical structure of the language, especially a complex language like Chinese. In this paper, we propose a deep layered model that stacks arbitrarily many RNN layers equipped with latent indicators. After using EM and training it hierarchically, our model solves the computational problem of stacking RNN layers and makes it possible to stack arbitrarily many RNN layers. Our deep hierarchical model not only achieves comparable results to large pre-trained models on the Chinese short text classification problem but also achieves state of art results on the Chinese long text classification problem.Keywords: nature language processing, recurrent neural network, hierarchical structure, document classification, Chinese
Procedia PDF Downloads 593065 Malignancy Assessment of Brain Tumors Using Convolutional Neural Network
Authors: Chung-Ming Lo, Kevin Li-Chun Hsieh
Abstract:
The central nervous system in the World Health Organization defines grade 2, 3, 4 gliomas according to the aggressiveness. For brain tumors, using image examination would have a lower risk than biopsy. Besides, it is a challenge to extract relevant tissues from biopsy operation. Observing the whole tumor structure and composition can provide a more objective assessment. This study further proposed a computer-aided diagnosis (CAD) system based on a convolutional neural network to quantitatively evaluate a tumor's malignancy from brain magnetic resonance imaging. A total of 30 grade 2, 43 grade 3, and 57 grade 4 gliomas were collected in the experiment. Transferred parameters from AlexNet were fine-tuned to classify the target brain tumors and achieved an accuracy of 98% and an area under the receiver operating characteristics curve (Az) of 0.99. Without pre-trained features, only 61% of accuracy was obtained. The proposed convolutional neural network can accurately and efficiently classify grade 2, 3, and 4 gliomas. The promising accuracy can provide diagnostic suggestions to radiologists in the clinic.Keywords: convolutional neural network, computer-aided diagnosis, glioblastoma, magnetic resonance imaging
Procedia PDF Downloads 1423064 Training a Neural Network to Segment, Detect and Recognize Numbers
Authors: Abhisek Dash
Abstract:
This study had three neural networks, one for number segmentation, one for number detection and one for number recognition all of which are coupled to one another. All networks were trained on the MNIST dataset and were convolutional. It was assumed that the images had lighter background and darker foreground. The segmentation network took 28x28 images as input and had sixteen outputs. Segmentation training starts when a dark pixel is encountered. Taking a window(7x7) over that pixel as focus, the eight neighborhood of the focus was checked for further dark pixels. The segmentation network was then trained to move in those directions which had dark pixels. To this end the segmentation network had 16 outputs. They were arranged as “go east”, ”don’t go east ”, “go south east”, “don’t go south east”, “go south”, “don’t go south” and so on w.r.t focus window. The focus window was resized into a 28x28 image and the network was trained to consider those neighborhoods which had dark pixels. The neighborhoods which had dark pixels were pushed into a queue in a particular order. The neighborhoods were then popped one at a time stitched to the existing partial image of the number one at a time and trained on which neighborhoods to consider when the new partial image was presented. The above process was repeated until the image was fully covered by the 7x7 neighborhoods and there were no more uncovered black pixels. During testing the network scans and looks for the first dark pixel. From here on the network predicts which neighborhoods to consider and segments the image. After this step the group of neighborhoods are passed into the detection network. The detection network took 28x28 images as input and had two outputs denoting whether a number was detected or not. Since the ground truth of the bounds of a number was known during training the detection network outputted in favor of number not found until the bounds were not met and vice versa. The recognition network was a standard CNN that also took 28x28 images and had 10 outputs for recognition of numbers from 0 to 9. This network was activated only when the detection network votes in favor of number detected. The above methodology could segment connected and overlapping numbers. Additionally the recognition unit was only invoked when a number was detected which minimized false positives. It also eliminated the need for rules of thumb as segmentation is learned. The strategy can also be extended to other characters as well.Keywords: convolutional neural networks, OCR, text detection, text segmentation
Procedia PDF Downloads 1513063 A Study on the Application of Machine Learning and Deep Learning Techniques for Skin Cancer Detection
Authors: Hritwik Ghosh, Irfan Sadiq Rahat, Sachi Nandan Mohanty, J. V. R. Ravindra
Abstract:
In the rapidly evolving landscape of medical diagnostics, the early detection and accurate classification of skin cancer remain paramount for effective treatment outcomes. This research delves into the transformative potential of Artificial Intelligence (AI), specifically Deep Learning (DL), as a tool for discerning and categorizing various skin conditions. Utilizing a diverse dataset of 3,000 images representing nine distinct skin conditions, we confront the inherent challenge of class imbalance. This imbalance, where conditions like melanomas are over-represented, is addressed by incorporating class weights during the model training phase, ensuring an equitable representation of all conditions in the learning process. Our pioneering approach introduces a hybrid model, amalgamating the strengths of two renowned Convolutional Neural Networks (CNNs), VGG16 and ResNet50. These networks, pre-trained on the ImageNet dataset, are adept at extracting intricate features from images. By synergizing these models, our research aims to capture a holistic set of features, thereby bolstering classification performance. Preliminary findings underscore the hybrid model's superiority over individual models, showcasing its prowess in feature extraction and classification. Moreover, the research emphasizes the significance of rigorous data pre-processing, including image resizing, color normalization, and segmentation, in ensuring data quality and model reliability. In essence, this study illuminates the promising role of AI and DL in revolutionizing skin cancer diagnostics, offering insights into its potential applications in broader medical domains.Keywords: artificial intelligence, machine learning, deep learning, skin cancer, dermatology, convolutional neural networks, image classification, computer vision, healthcare technology, cancer detection, medical imaging
Procedia PDF Downloads 783062 Drug-Drug Interaction Prediction in Diabetes Mellitus
Authors: Rashini Maduka, C. R. Wijesinghe, A. R. Weerasinghe
Abstract:
Drug-drug interactions (DDIs) can happen when two or more drugs are taken together. Today DDIs have become a serious health issue due to adverse drug effects. In vivo and in vitro methods for identifying DDIs are time-consuming and costly. Therefore, in-silico-based approaches are preferred in DDI identification. Most machine learning models for DDI prediction are used chemical and biological drug properties as features. However, some drug features are not available and costly to extract. Therefore, it is better to make automatic feature engineering. Furthermore, people who have diabetes already suffer from other diseases and take more than one medicine together. Then adverse drug effects may happen to diabetic patients and cause unpleasant reactions in the body. In this study, we present a model with a graph convolutional autoencoder and a graph decoder using a dataset from DrugBank version 5.1.3. The main objective of the model is to identify unknown interactions between antidiabetic drugs and the drugs taken by diabetic patients for other diseases. We considered automatic feature engineering and used Known DDIs only as the input for the model. Our model has achieved 0.86 in AUC and 0.86 in AP.Keywords: drug-drug interaction prediction, graph embedding, graph convolutional networks, adverse drug effects
Procedia PDF Downloads 953061 Artificial Neural Networks and Hidden Markov Model in Landslides Prediction
Authors: C. S. Subhashini, H. L. Premaratne
Abstract:
Landslides are the most recurrent and prominent disaster in Sri Lanka. Sri Lanka has been subjected to a number of extreme landslide disasters that resulted in a significant loss of life, material damage, and distress. It is required to explore a solution towards preparedness and mitigation to reduce recurrent losses associated with landslides. Artificial Neural Networks (ANNs) and Hidden Markov Model (HMMs) are now widely used in many computer applications spanning multiple domains. This research examines the effectiveness of using Artificial Neural Networks and Hidden Markov Model in landslides predictions and the possibility of applying the modern technology to predict landslides in a prominent geographical area in Sri Lanka. A thorough survey was conducted with the participation of resource persons from several national universities in Sri Lanka to identify and rank the influencing factors for landslides. A landslide database was created using existing topographic; soil, drainage, land cover maps and historical data. The landslide related factors which include external factors (Rainfall and Number of Previous Occurrences) and internal factors (Soil Material, Geology, Land Use, Curvature, Soil Texture, Slope, Aspect, Soil Drainage, and Soil Effective Thickness) are extracted from the landslide database. These factors are used to recognize the possibility to occur landslides by using an ANN and HMM. The model acquires the relationship between the factors of landslide and its hazard index during the training session. These models with landslide related factors as the inputs will be trained to predict three classes namely, ‘landslide occurs’, ‘landslide does not occur’ and ‘landslide likely to occur’. Once trained, the models will be able to predict the most likely class for the prevailing data. Finally compared two models with regards to prediction accuracy, False Acceptance Rates and False Rejection rates and This research indicates that the Artificial Neural Network could be used as a strong decision support system to predict landslides efficiently and effectively than Hidden Markov Model.Keywords: landslides, influencing factors, neural network model, hidden markov model
Procedia PDF Downloads 3793060 Using AI to Advance Factory Planning: A Case Study to Identify Success Factors of Implementing an AI-Based Demand Planning Solution
Authors: Ulrike Dowie, Ralph Grothmann
Abstract:
Rational planning decisions are based upon forecasts. Precise forecasting has, therefore, a central role in business. The prediction of customer demand is a prime example. This paper introduces recurrent neural networks to model customer demand and combines the forecast with uncertainty measures to derive decision support of the demand planning department. It identifies and describes the keys to the successful implementation of an AI-based solution: bringing together data with business knowledge, AI methods, and user experience, and applying agile software development practices.Keywords: agile software development, AI project success factors, deep learning, demand forecasting, forecast uncertainty, neural networks, supply chain management
Procedia PDF Downloads 1803059 Person Re-Identification using Siamese Convolutional Neural Network
Authors: Sello Mokwena, Monyepao Thabang
Abstract:
In this study, we propose a comprehensive approach to address the challenges in person re-identification models. By combining a centroid tracking algorithm with a Siamese convolutional neural network model, our method excels in detecting, tracking, and capturing robust person features across non-overlapping camera views. The algorithm efficiently identifies individuals in the camera network, while the neural network extracts fine-grained global features for precise cross-image comparisons. The approach's effectiveness is further accentuated by leveraging the camera network topology for guidance. Our empirical analysis on benchmark datasets highlights its competitive performance, particularly evident when background subtraction techniques are selectively applied, underscoring its potential in advancing person re-identification techniques.Keywords: camera network, convolutional neural network topology, person tracking, person re-identification, siamese
Procedia PDF Downloads 673058 LGG Architecture for Brain Tumor Segmentation Using Convolutional Neural Network
Authors: Sajeeha Ansar, Asad Ali Safi, Sheikh Ziauddin, Ahmad R. Shahid, Faraz Ahsan
Abstract:
The most aggressive form of brain tumor is called glioma. Glioma is kind of tumor that arises from glial tissue of the brain and occurs quite often. A fully automatic 2D-CNN model for brain tumor segmentation is presented in this paper. We performed pre-processing steps to remove noise and intensity variances using N4ITK and standard intensity correction, respectively. We used Keras open-source library with Theano as backend for fast implementation of CNN model. In addition, we used BRATS 2015 MRI dataset to evaluate our proposed model. Furthermore, we have used SimpleITK open-source library in our proposed model to analyze images. Moreover, we have extracted random 2D patches for proposed 2D-CNN model for efficient brain segmentation. Extracting 2D patched instead of 3D due to less dimensional information present in 2D which helps us in reducing computational time. Dice Similarity Coefficient (DSC) is used as performance measure for the evaluation of the proposed method. Our method achieved DSC score of 0.77 for complete, 0.76 for core, 0.77 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, LGG
Procedia PDF Downloads 1763057 A Custom Convolutional Neural Network with Hue, Saturation, Value Color for Malaria Classification
Authors: Ghazala Hcini, Imen Jdey, Hela Ltifi
Abstract:
Malaria disease should be considered and handled as a potential restorative catastrophe. One of the most challenging tasks in the field of microscopy image processing is due to differences in test design and vulnerability of cell classifications. In this article, we focused on applying deep learning to classify patients by identifying images of infected and uninfected cells. We performed multiple forms, counting a classification approach using the Hue, Saturation, Value (HSV) color space. HSV is used since of its superior ability to speak to image brightness; at long last, for classification, a convolutional neural network (CNN) architecture is created. Clusters of focus were used to deliver the classification. The highlights got to be forbidden, and a few more clamor sorts are included in the information. The suggested method has a precision of 99.79%, a recall value of 99.55%, and provides 99.96% accuracy.Keywords: deep learning, convolutional neural network, image classification, color transformation, HSV color, malaria diagnosis, malaria cells images
Procedia PDF Downloads 853056 Thrombophilic Mutations in Tunisian Patients with Recurrent Pregnancy Loss
Authors: Frikha Rim, Abdelmoula Bouayed Nouha, Rebai Tarek
Abstract:
Pregnancy is a hypercoagulable state which causing a defective maternal haemostatic response and leading to thrombosis of the uteroplacental vasculature, that might cause pregnancy complications as recurrent pregnancy loss (RPL). Since heritable Thrombophilic defects are associated with increased thrombosis, their prevalence was evaluated in patients with special emphasis on combinations of the above pathologies. Especially, Factor V Leiden (FVL) G1691A, methylene tetra hydro folate reductase (MTHFR) C677T, and factor II (FII) G20210A mutations are three important causes of thrombophilia, which might be related to recurrent pregnancy loss (RPL). In this study we evaluated the presence of these three mutations [factor V Leiden (FVL), prothrombin G20210A (PTG) and methylenetetrahydrofolate reductase (MTHFR) C677T] amongst 35 Tunisian women with more than 2 miscarriages, referred to our genetic counseling. DNA was extracted from peripheral blood samples and PCR-RFLP was performed for the molecular diagnosis of each mutation. Factor V Leiden and Prothrombin mutation were detected respectively in 5.7% and 2.9% of women with particular history of early fetal loss and thrombotic events. Despites the luck of strength of this study, we insist that testing for the most inherited thrombophilia (FVL and FII mutation) should be performed in women with RPL in the context of thrombotic events. Multi-centre collaboration is necessary to clarify the real impact of thrombotic molecular defects on the pregnancy outcome, to ascertain the effect of thrombophilia on recurrent pregnancy loss and then to evaluate the appropriate therapeutic approach.Keywords: thrombophilia, recurrent pregnancy loss, factor V Leiden, prothrombin G20210A, methylene tetra hydro folate reductase
Procedia PDF Downloads 4503055 Attention Based Fully Convolutional Neural Network for Simultaneous Detection and Segmentation of Optic Disc in Retinal Fundus Images
Authors: Sandip Sadhukhan, Arpita Sarkar, Debprasad Sinha, Goutam Kumar Ghorai, Gautam Sarkar, Ashis K. Dhara
Abstract:
Accurate segmentation of the optic disc is very important for computer-aided diagnosis of several ocular diseases such as glaucoma, diabetic retinopathy, and hypertensive retinopathy. The paper presents an accurate and fast optic disc detection and segmentation method using an attention based fully convolutional network. The network is trained from scratch using the fundus images of extended MESSIDOR database and the trained model is used for segmentation of optic disc. The false positives are removed based on morphological operation and shape features. The result is evaluated using three-fold cross-validation on six public fundus image databases such as DIARETDB0, DIARETDB1, DRIVE, AV-INSPIRE, CHASE DB1 and MESSIDOR. The attention based fully convolutional network is robust and effective for detection and segmentation of optic disc in the images affected by diabetic retinopathy and it outperforms existing techniques.Keywords: attention-based fully convolutional network, optic disc detection and segmentation, retinal fundus image, screening of ocular diseases
Procedia PDF Downloads 1393054 INRAM-3DCNN: Multi-Scale Convolutional Neural Network Based on Residual and Attention Module Combined with Multilayer Perceptron for Hyperspectral Image Classification
Authors: Jianhong Xiang, Rui Sun, Linyu Wang
Abstract:
In recent years, due to the continuous improvement of deep learning theory, Convolutional Neural Network (CNN) has played a great superior performance in the research of Hyperspectral Image (HSI) classification. Since HSI has rich spatial-spectral information, only utilizing a single dimensional or single size convolutional kernel will limit the detailed feature information received by CNN, which limits the classification accuracy of HSI. In this paper, we design a multi-scale CNN with MLP based on residual and attention modules (INRAM-3DCNN) for the HSI classification task. We propose to use multiple 3D convolutional kernels to extract the packet feature information and fully learn the spatial-spectral features of HSI while designing residual 3D convolutional branches to avoid the decline of classification accuracy due to network degradation. Secondly, we also design the 2D Inception module with a joint channel attention mechanism to quickly extract key spatial feature information at different scales of HSI and reduce the complexity of the 3D model. Due to the high parallel processing capability and nonlinear global action of the Multilayer Perceptron (MLP), we use it in combination with the previous CNN structure for the final classification process. The experimental results on two HSI datasets show that the proposed INRAM-3DCNN method has superior classification performance and can perform the classification task excellently.Keywords: INRAM-3DCNN, residual, channel attention, hyperspectral image classification
Procedia PDF Downloads 733053 Multimodal Direct Neural Network Positron Emission Tomography Reconstruction
Authors: William Whiteley, Jens Gregor
Abstract:
In recent developments of direct neural network based positron emission tomography (PET) reconstruction, two prominent architectures have emerged for converting measurement data into images: 1) networks that contain fully-connected layers; and 2) networks that primarily use a convolutional encoder-decoder architecture. In this paper, we present a multi-modal direct PET reconstruction method called MDPET, which is a hybrid approach that combines the advantages of both types of networks. MDPET processes raw data in the form of sinograms and histo-images in concert with attenuation maps to produce high quality multi-slice PET images (e.g., 8x440x440). MDPET is trained on a large whole-body patient data set and evaluated both quantitatively and qualitatively against target images reconstructed with the standard PET reconstruction benchmark of iterative ordered subsets expectation maximization. The results show that MDPET outperforms the best previously published direct neural network methods in measures of bias, signal-to-noise ratio, mean absolute error, and structural similarity.Keywords: deep learning, image reconstruction, machine learning, neural network, positron emission tomography
Procedia PDF Downloads 1063052 Online Authenticity Verification of a Biometric Signature Using Dynamic Time Warping Method and Neural Networks
Authors: Gałka Aleksandra, Jelińska Justyna, Masiak Albert, Walentukiewicz Krzysztof
Abstract:
An offline signature is well-known however not the safest way to verify identity. Nowadays, to ensure proper authentication, i.e. in banking systems, multimodal verification is more widely used. In this paper the online signature analysis based on dynamic time warping (DTW) coupled with machine learning approaches has been presented. In our research signatures made with biometric pens were gathered. Signature features as well as their forgeries have been described. For verification of authenticity various methods were used including convolutional neural networks using DTW matrix and multilayer perceptron using sums of DTW matrix paths. System efficiency has been evaluated on signatures and signature forgeries collected on the same day. Results are presented and discussed in this paper.Keywords: dynamic time warping, handwritten signature verification, feature-based recognition, online signature
Procedia PDF Downloads 165