Search

In most regions of the world, tuberculosis (TB) is classified as a malignant infectious disease that can be fatal. Using advanced tools and technology, automatic analysis and classification of chest X-rays (CXRs) into TB and non-TB can be a reliable alternative to the subjective assessment performed by healthcare professionals. Thus, in the study, we propose an automatic TB detection system using advanced deep learning (DL) models. A significant portion of a CXR image is dark, providing no information for diagnosis and potentially confusing DL models. Therefore, in the proposed system, we use sophisticated segmentation networks to extract the region of interest from multimedia CXRs. Then, segmented images are fed into the DL models. For the subjective assessment, we use explainable artificial intelligence to visualize TB-infected parts of the lung. We use different convolutional neural network (CNN) models in our experiments and compare their classification performance using three publicly available CXR datasets. EfficientNetB3, one of the CNN models, achieves the highest accuracy of 99.1%, with a receiver operating characteristic of 99.9%, and an average accuracy of 98.7%. Experiment results confirm that using segmented lung CXR images produces better performance than does using raw lung CXR images. [ABSTRACT FROM AUTHOR]

Early illness detection enables medical professionals to deliver the best care and increases the likelihood of a full recovery. In this work, we show that computer-aided design (CAD) systems are capable of using chest X-ray (CXR) medical imaging modalities for the identification of respiratory system disorders. At present, the COVID-19 pandemic is the most well-known illness. We propose a system based on explainable artificial intelligence to detect COVID-19 from CXR images by using several cutting-edge convolutional neural network (CNN) models, as well as the Vision of Transformer (ViT) models. The proposed system also visualizes the infected areas of the CXR images. This gives doctors and other medical professionals a second option for supporting their decision. The proposed system uses some preprocessing of the images, which includes the segmentation of the region of interest using a UNet model and rotation augmentation. CNN employs pixel arrays, while ViT divides the image into visual tokens; therefore, one of the objectives is to compare their performance in COVID-19 detection. In the experiments, a publicly available dataset (COVID-QU-Ex) is used. The experimental results show that the performances of the CNN-based models and the ViT-based models are comparable. The best accuracy was 99.82%, obtained by the EfficientNetB7 (CNN-based) model, followed by the SegFormer (ViT-based). In addition, the segmentation and augmentation enhanced the performance. [ABSTRACT FROM AUTHOR]

Human falls, especially for elderly people, can cause serious injuries that might lead to permanent disability. Approximately 20–30% of the aged people in the United States who experienced fall accidents suffer from head trauma, injuries, or bruises. Fall detection is becoming an important public healthcare problem. Timely and accurate fall incident detection could enable the instant delivery of medical services to the injured. New advances in vision-based technologies, including deep learning, have shown significant results in action recognition, where some focus on the detection of fall actions. In this paper, we propose an automatic human fall detection system using multi-stream convolutional neural networks with fusion. The system is based on a multi-level image-fusion approach of every 16 frames of an input video to highlight movement differences within this range. This results of four consecutive preprocessed images are fed to a new proposed and efficient lightweight multi-stream CNN model that is based on a four-branch architecture (4S-3DCNN) that classifies whether there is an incident of a human fall. The evaluation included the use of more than 6392 generated sequences from the Le2i fall detection dataset, which is a publicly available fall video dataset. The proposed method, using three-fold cross-validation to validate generalization and susceptibility to overfitting, achieved a 99.03%, 99.00%, 99.68%, and 99.00% accuracy, sensitivity, specificity, and precision, respectively. The experimental results prove that the proposed model outperforms state-of-the-art models, including GoogleNet, SqueezeNet, ResNet18, and DarkNet19, for fall incident detection. [ABSTRACT FROM AUTHOR]

Brain signals can be captured via electroencephalogram (EEG) and be used in various brain–computer interface (BCI) applications. Classifying motor imagery (MI) using EEG signals is one of the important applications that can help a stroke patient to rehabilitate or perform certain tasks. Dealing with EEG-MI signals is challenging because the signals are weak, may contain artefacts, are dependent on the patient's mood and posture, and have low signal-to-noise ratio. This paper proposes a multi-branch convolutional neural network model called the Multi-Branch EEGNet with Convolutional Block Attention Module (MBEEGCBAM) using attention mechanism and fusion techniques to classify EEG-MI signals. The attention mechanism is applied both channel-wise and spatial-wise. The proposed model is a lightweight model that has fewer parameters and higher accuracy compared to other state-of-the-art models. The accuracy of the proposed model is 82.85% and 95.45% using the BCI-IV2a motor imagery dataset and the high gamma dataset, respectively. Additionally, when using the fusion approach (FMBEEGCBAM), it achieves 83.68% and 95.74% accuracy, respectively. [ABSTRACT FROM AUTHOR]

Electroencephalography-based motor imagery (EEG-MI) classification is a critical component of the brain-computer interface (BCI), which enables people with physical limitations to communicate with the outside world via assistive technology. Regrettably, EEG decoding is challenging because of the complexity, dynamic nature, and low signal-to-noise ratio of the EEG signal. Developing an end-to-end architecture capable of correctly extracting EEG data's high-level features remains a difficulty. This study introduces a new model for decoding MI known as a Multi-Branch EEGNet with squeeze-and-excitation blocks (MBEEGSE). By clearly specifying channel interdependencies, a multi-branch CNN model with attention blocks is employed to adaptively change channel-wise feature responses. When compared to existing state-of-the-art EEG motor imagery classification models, the suggested model achieves good accuracy (82.87%) with reduced parameters in the BCI-IV2a motor imagery dataset and (96.15%) in the high gamma dataset. [ABSTRACT FROM AUTHOR]

Stroke, particularly ischemic stroke, is a major cause of disability and one of the leading causes of adult mortality worldwide. Early and prompt management of stroke patients can reduce the severity of the disease. Doctors usually determine the severity of a stroke by focusing on the region of interest (ROI) in the MRI or CT scan images. An accurate and effective automatic image segmentation system can assist medical professionals as well as automatic detection and classification systems. Deep learning is the current advanced approach for dealing with machine learning and artificial intelligence. However, conventional deep learning requires a large amount of data for training, and the amount of labeled data in the medical field is limited. In this paper, we propose a few-shot learning strategy and integrate it with a base convolutional neural network model, which utilizes a self-attention mechanism to segment MRI for ischemic stroke. By combining the base model with self-attention, we can focus more on the ROI and disregard less important features. Additionally, the proposed system only selects slices with lesions and ignores unlesioned slices. This helps to improve efficiency and reduce the computational load by eliminating the need to tune unnecessary parameters. To achieve even better results, the system also combines two weighted images, FLAIR and DWI, as an early fusion process. Experiments have shown that this approach leads to higher performance compared to using the same system without fusion. The proposed system is evaluated using a publicly available dataset, ISLES 2015 SSIS, and compared with other state-of-the-art (SOTA) systems. It achieves a dice coefficient score of 0.68, which is significantly better than that of other SOTA systems. • An automatic ischemic stroke segmentation system using few-shot learning is proposed. • The proposed system selects only slices with lesions and excludes unlesioned slices. • We adopted prototypical networks with metric learning to achieve few-shot segmentation. • Two weighted MRIs, FLAIR, and DWI are fused in the few-shot learning paradigm. [ABSTRACT FROM AUTHOR]

Fake news is frequently disseminated through social media, which significantly impacts public perception and individual decision-making. Accurate identification of fake news on social media is usually time-consuming, laborious, and difficult. Although the leveraging of machine learning technologies can facilitate automated authenticity checks, the time-sensitive and voluminous nature of the data brings considerable challenge for fake news detection. To address this issue, this paper proposes a quantum multimodal fusion-based model for fake news detection (QMFND). QMFND integrates the extracted images and textual features, and passes them through a proposed quantum convolutional neural network (QCNN) to obtain discriminative results. By testing QMFND on two social media datasets, Gossip and Politifact, it is proved that its detection performance is equal to or even surpasses that of classical models. The effects of various parameters are further investigated. The QCNN not only has good expressibility and entangling capability but also has good robustness against quantum noise. The code is available at • Quantum multimodal fusion-based fake news detection model is proposed for social media. • Amplitude encoding is adopted to encode multimodal features into a VQC. • QCNN is proposed for efficient training of encoded multimodal data. [ABSTRACT FROM AUTHOR]

Automating polyps' detection and segmentation is very helpful in achieving early detection and potentially effective treatment of colon cancer. Because of the constant rise in processing power at a reasonable cost and the availability of training data, deep learning more than ever is becoming very powerful and increasingly widespread in many domains, including computer vision and medical image processing and segmentation. In this paper, we propose a novel method for polyp image segmentation using convolutional neural networks (CNNs) and utilizing multi-parallel U-Net encoder architecture. The proposed method uses multiple encoders that utilize pre-trained CNNs to enrich the extracted features. Skip connections from these parallel encoders are concatenated and propagated to the decoder in partially and fully meshed fashions allowing for data from shallow to deeper layers in the network to be exchanged and vice versa. We trained and tested the method on five publicly available datasets: Kvasir, CVC-Clinic DB, CVC-Colon DB, CVC-T, and ETIS-Larib, which are well-known benchmark datasets for polyp image segmentation. We tested the performance of the method on these five datasets using multiple testing scenarios including the effect of using attention, vision transformers, and multiple decoders. Experimental results showed that the proposed method outperforms other state-of-the-art methods on two of the used benchmark datasets and ranked second on a third. • A novel CNN-based method using parallel encoders for polyp segmentation is proposed. • Combinations of two and three parallel encoders, and decoders with different pre-trained networks are investigated. • The effect of attention and vision transformers is examined. • Five publicly available polyp image databases were used in the experiments. • Dual encoder with attention block and partial mesh skip connections achieved the best result. [ABSTRACT FROM AUTHOR]

Human Activity Recognition (HAR) has become a crucial element for smart healthcare applications due to the fast adoption of wearable sensors and mobile technologies. Most of the existing human activity recognition frameworks deal with a single modality of data that degrades the reliability and recognition accuracy of the system for heterogeneous data sources. In this article, we propose a multi-level feature fusion technique for multimodal human activity recognition using multi-head Convolutional Neural Network (CNN) with Convolution Block Attention Module (CBAM) to process the visual data and Convolutional Long Short Term Memory (ConvLSTM) for dealing with the time-sensitive multi-source sensor information. The architecture is developed to be able to analyze and retrieve channel and spatial dimension features through the use of three branches of CNN along with CBAM for visual information. The ConvLSTM network is designed to capture temporal features from the multiple sensors' time-series data for efficient activity recognition. An open-access multimodal HAR dataset named UP-Fall detection dataset is utilized in experiments and evaluations to measure the performance of the developed fusion architecture. Finally, we deployed an Internet of Things (IoT) system to test the proposed fusion network in real-world smart healthcare application scenarios. The findings from the experimental results reveal that the developed multimodal HAR framework surpasses the existing state-of-the-art methods in terms of multiple performance metrics. • Multimodal data fusion for HAR through multi-head architecture, and ConvLSTM. • Retrieving the channel and spatial dimension features via CBAM. • Model deployment in IoT networks for gauging performance in IoHT. • Tests on the UP-Fall detection dataset demonstrate state-of-the-art results. [ABSTRACT FROM AUTHOR]

The demand for automatic detection of Novel Coronavirus or COVID-19 is increasing across the globe. The exponential rise in cases burdens healthcare facilities, and a vast amount of multimedia healthcare data is being explored to find a solution. This study presents a practical solution to detect COVID-19 from chest X-rays while distinguishing those from normal and impacted by Viral Pneumonia via Deep Convolution Neural Networks (CNN). In this study, three pre-trained CNN models (EfficientNetB0, VGG16, and InceptionV3) are evaluated through transfer learning. The rationale for selecting these specific models is their balance of accuracy and efficiency with fewer parameters suitable for mobile applications. The dataset used for the study is publicly available and compiled from different sources. This study uses deep learning techniques and performance metrics (accuracy, recall, specificity, precision, and F1 scores). The results show that the proposed approach produced a high-quality model, with an overall accuracy of 92.93%, COVID-19, a sensitivity of 94.79%. The work indicates a definite possibility to implement computer vision design to enable effective detection and screening measures. [ABSTRACT FROM AUTHOR]

COVID-19 is an infectious disease caused by the deadly virus SARS-CoV-2 that affects the lung of the patient. Different symptoms, including fever, muscle pain and respiratory syndrome, can be identified in COVID-19-affected patients. The disease needs to be diagnosed in a timely manner, otherwise the lung infection can turn into a severe form and the patient's life may be in danger. In this work, an ensemble deep learning-based technique is proposed for COVID-19 detection that can classify the disease with high accuracy, efficiency, and reliability. A weighted average ensemble (WAE) prediction was performed by combining three CNN models, namely Xception, VGG19 and ResNet50V2, where 97.25% and 94.10% accuracy was achieved for binary and multiclass classification, respectively. To accurately detect the disease, different test methods have been proposed and developed, some of which are even being used in real-time situations. RT-PCR is one of the most successful COVID-19 detection methods, and is being used worldwide with high accuracy and sensitivity. However, complexity and time-consuming manual processes are limitations of this method. To make the detection process automated, researchers across the world have started to use deep learning to detect COVID-19 applied on medical imaging. Although most of the existing systems offer high accuracy, different limitations, including high variance, overfitting and generalization errors, can be found that can degrade the system performance. Some of the reasons behind those limitations are a lack of reliable data resources, missing preprocessing techniques, a lack of proper model selection, etc., which eventually create reliability issues. Reliability is an important factor for any healthcare system. Here, transfer learning with better preprocessing techniques applied on two benchmark datasets makes the work more reliable. The weighted average ensemble technique with hyperparameter tuning ensures better accuracy than using a randomly selected single CNN model. [ABSTRACT FROM AUTHOR]

Sign language is the main channel for hearing-impaired people to communicate with others. It is a visual language that conveys highly structured components of manual and non-manual parameters such that it needs a lot of effort to master by hearing people. Sign language recognition aims to facilitate this mastering difficulty and bridge the communication gap between hearing-impaired people and others. This study presents an efficient architecture for sign language recognition based on a convolutional graph neural network (GCN). The presented architecture consists of a few separable 3DGCN layers, which are enhanced by a spatial attention mechanism. The limited number of layers in the proposed architecture enables it to avoid the common over-smoothing problem in deep graph neural networks. Furthermore, the attention mechanism enhances the spatial context representation of the gestures. The proposed architecture is evaluated on different datasets and shows outstanding results. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) remains a challenging yet crucial problem to address in computer vision. HAR is primarily intended to be used with other technologies, such as the Internet of Things, to assist in healthcare and eldercare. With the development of deep learning, automatic high-level feature extraction has become a possibility and has been used to optimize HAR performance. Furthermore, deep-learning techniques have been applied in various fields for sensor-based HAR. This study introduces a new methodology using convolution neural networks (CNN) with varying kernel dimensions along with bi-directional long short-term memory (BiLSTM) to capture features at various resolutions. The novelty of this research lies in the effective selection of the optimal video representation and in the effective extraction of spatial and temporal features from sensor data using traditional CNN and BiLSTM. Wireless sensor data mining (WISDM) and UCI datasets are used for this proposed methodology in which data are collected through diverse methods, including accelerometers, sensors, and gyroscopes. The results indicate that the proposed scheme is efficient in improving HAR. It was thus found that unlike other available methods, the proposed method improved accuracy, attaining a higher score in the WISDM dataset compared to the UCI dataset (98.53% vs. 97.05%). [ABSTRACT FROM AUTHOR]

Human Activity Recognition (HAR) is the process of automatic recognition of Activities of Daily Living (ADL) from human motion data captured in various data modalities by wearable and ambient sensors. Advances in Deep Learning, especially Convolutional Neural Networks (CNNs) and sequential models have revolutionalized HAR from video data sources. Although these models are incredibly accurate and utilize both spatial and temporal information, they require huge computation and memory resources — making them unsuitable for edge or wearable applications. Tiny Machine Learning (TinyML) aims to optimize these models in terms of compute and memory requirements – aiming to make them suitable for always-on resource constrained devices – leading to a reduction in communication latency and network traffic for HAR frameworks. In this paper, we propose a two-stream multi-resolution fusion architecture for HAR from video data modality. The context stream takes a resized image as input and the fovea stream takes the cropped center portion of the resized image as input, reducing the overall dimensionality. We tested two quantization methods: Post-Training Quantization (PTQ) and Quantization Aware Training (QAT) to optimize these models for deployment in edge devices and tested the performance in two challenging video datasets: KTH and UCF11. We performed ablation studies to validate the two-stream model performance. We deployed the proposed architecture in commercial resource constrained devices and monitored their performance in terms of inference latency and power consumption. The results indicate that the proposed architecture clearly outperforms other relevant single-stream models tested in this work in terms of accuracy, precision, recall, and F1-Score while also reducing the overall model size. • A novel two-stream multi-resolution fusion architecture for HAR from video data. • The context stream inputs the lower-resolution images. • The fovea stream takes center-cropped portions of the images as inputs. • Quantizing the architecture to reduce the size to fit into low-power devices. • Deployment of the proposed approach in commodity tiny edge devices. [ABSTRACT FROM AUTHOR]

• Bio-Inspired novel attention-based inception architecture is proposed that can adapt to different types of spatial contexts using convolution filters of different sizes. The characteristics of each dataset are unique, hence the attention mechanism helps focus on those features to improve classification performance. • The shallow inception model is designed with a two-layer attention mechanism with fewer layers but with a large number of convolution filters that can address the overfitting problem caused by small dataset sizes. • LSTM-based recurrent neural network (RNN) module is proposed to extract temporal features after the inception module is applied. • The proposed model is lightweight with fewer parameters and has less processing time. • The proposed model achieves good performance for both dynamic and static signs and gestures. Bio-inspired deep learning models have revolutionized sign language classification, achieving extraordinary accuracy and human-like video understanding. Recognition and classification of sign language videos in real-time are challenging because the duration and speed of each sign vary for different subjects, the background of videos is dynamic in most cases, and the classification result should be produced in real-time. This study proposes a model based on a convolution neural network (CNN) Inception model with an attention mechanism for extracting spatial features and Bi-LSTM (long short-term memory) for temporal feature extraction. The proposed model is tested on datasets with highly variable characteristics such as different clothing, variable lighting, and variable distance from the camera. Real-time classification achieves significant early detections while achieving performance comparable to the offline operation. The proposed model has fewer parameters, fewer deep learning layers, and requires significantly less processing time than state-of-the-art models. The Inception model with an attention mechanism with two attention blocks [Display omitted] [ABSTRACT FROM AUTHOR]

• We proposed a new self-attention based recurrent convolutional neural network for prediction of cerebral infarction disease. Experiment results exhibit that the proposed model based on clinical intelligence approach perform appropriate diagnosis for prediction of cerebral infarction disease. •.Clinical notes are extracted from real-life hospital dataset to use in experiment. After that, put forward a model which automatically learns high-level semantic features from unstructured clinical text using sliding window convolution approach with the intra-layer recurrent connection. • Self-attention mechanism is used to get the focus of the model on essential words which contribute to meaning in the clinical text by using softmax probability distribution at the interior of the model. Background and Objective: Nowadays computer-aided disease diagnosis from medical data through deep learning methods has become a wide area of research. Existing works of analyzing clinical text data in the medical domain, which substantiate useful information related to patients with disease in large quantity, benefits early-stage disease diagnosis. However, benefits of analysis not achieved well when the traditional rule-based and classical machine learning methods used; which are unable to handle the unstructured clinical text and only a single method is not able to handle all challenges related to the analysis of the unstructured text, Moreover, the contribution of all words in clinical text is not the same in the prediction of disease. Therefore, there is a need to develop a neural model which solve the above clinical application problems, is an interesting topic which needs to be explored. Methods: Thus considering the above problems, first, this paper present self-attention based recurrent convolutional neural network (RCNN) model using real-life clinical text data collected from a hospital in Wuhan, China. This model automatically learns high-level semantic features from clinical text by using bi-direction recurrent connection within convolution. Second, to deal with other clinical text challenges, we combine the ability of RCNN with the self-attention mechanism. Thus, self-attention gets the focus of the model on essential convolve features which have effective meaning in the clinical text by calculating the probability of each convolve feature through softmax. Results: The proposed model is evaluated on real-life hospital dataset and used measurement metrics as Accuracy and recall. Experiment results exhibit that the proposed model reaches up to accuracy 95.71%, which is better than many existing methods for cerebral infarction disease. Conclusions: This article presented the self-attention based RCNN model by combining the RCNN with self-attention mechanism for prediction of cerebral infarction disease. The obtained results show that the presented model better predict the cerebral infarction disease risk compared to many existing methods. The same model can also be used for the prediction of other disease risks. [ABSTRACT FROM AUTHOR]