Search

Recently, huge concerns have been raised in diagnosing chest diseases, especially after the COVID-19 pandemic. Regular diagnosis processes of chest diseases sometimes fail to distinguish between Corona and Viral Pneumonia diseases through Polymerase Chain Reaction (PCR) tests which are a time-engrossing process that needs convoluted manual procedures. Artificial Intelligence (AI) techniques have achieved high performance in aiding medical diagnostic processes. The innovation of this work lies in using a new diagnostic technique to distinguish between COVID-19 and Viral Pneumonia diseases using advanced AI technologies. This is done by extracting novel features from chest X-ray images based on Wavelet analysis, Scale Invariant Feature Transformation (SIFT), and the Mel Frequency Cepstral Coefficient (MFCC). Support vector machines (SVM) and artificial neural networks (ANN) were utilized to build classification algorithms using 1200 chest X-ray mages for each case. Using Wavelet features, the results of evaluating the SVM and ANN models were 97% accurate, and with SIFT features, they were closer to 99%. The proposed models were very effective at identifying COVID-19 and Viral Pneumonitis, so physicians can determine the best treatment course for patients with the support of this high accuracy. Moreover, this model can be used in hospitals and emergency rooms when a massive number of patients are waiting, as it is faster and more accurate than the regular diagnosis processes as each step takes few seconds on average to complete. [ABSTRACT FROM AUTHOR]

Biomedical signal processing is crucial in many sectors that save lives. Artificial intelligence improvement in signal collection and conditioning boosted this application's adaptability to varied bodily circumstances. In this study, a novel method is put forth for predicting the type of stroke in the human brain based on the observation of the Electroencephalography (EEG) signal. The signal is the first condition for removing undesirable frequencies by passing through a lowpass filter. To accurately extract the signal features, the signal is first transformed into a 1-second frame format and then normalised. Certain statistical and frequency domain aspects are highlighted to increase taxonomic accuracy. Under the wavelet packet transform, the empirical mode decomposition approach is utilised to recover the most information feasible from the signal. After training on extracted characteristics, the extreme learning machine is regarded to conduct classification. These work achieves 94.95 of Sensitivity, 84.95 of Specificity, 93.74 of Precision, 96.96 of Accuracy, 96.12 of F1 Score. Compared to the standard procedures, the proposed techniques have a greater accuracy rate of about 98%. [ABSTRACT FROM AUTHOR]

Retinal image segmentation process deals with problems like spurious vascularisation and thin vessel detection. In this paper, a three-step methodology has been proposed for retinal vessel segmentation. In first step, RGB to YIQ conversion is performed. In second step, Y component enhancement is performed. A novel Chaotic weighted Elephant Herding Optimization (CWEHO) has been proposed to optimize the clip limit and block size values of Contrast Limited Adaptive Histogram Equalization (CLAHE). CWEHO-based CLAHE along with morphological operations, non-local means filter, and median filter is applied to enhance retinal images. In third step, thin and thick vessel segmentation is performed. Top hat transformation, otsu thresholding algorithm, and vessel point selection are applied for thick vessel extraction. The first-order Gaussian derivative in conjunction with the match filter is used to extract thin vessels. DRIVE and HRF datasets are used to assess the effectiveness of proposed methodology. The average values of segmentation accuracy, specificity, sensitivity, and Mathew Correlation Coefficient (MCC) are observed to be 0.9650, 0.9895, 0.7007, 0.7650, respectively, for observer1 and 0.9696, 0.9912, 0.7390, 0.7901 for observer2 using DRIVE dataset. Similarly, 0.9592, 0.9839, 0.6850, and 0.7116, respectively, metrics for HRF dataset. Compared to state-of-the-art methods, the proposed segmentation methodology provides better results. [ABSTRACT FROM AUTHOR]

Image muscle segmentation is useful to quantitatively assess musculoskeletal diseases by extracting biomarkers such as shape, texture and water diffusivity metrics. Although volumetric manual segmentation is time consuming and a bottleneck in practice, fully automatic approaches are still in progress to reach an acceptable accuracy. In this paper, we provide a robust semi-automated tool to segment two musculoskeletal systems, i.e. thigh and shoulder in MRI and CT modalities, respectively. The tool only needs a few manually labelled cross-sections to build a directed graph-structure of corresponding points between the successive spaced slices. The boundaries of each muscle are obtained by performing a spline interpolation based on the directed graph-structure. Each muscle label and its corresponding 3D mesh are deduced using post-processing techniques. We evaluated the tool on 26 MRI thighs and 16 CT shoulders. Three metrics along with inter-muscle overlapping were employed to evaluate the tool by comparison to an expert manual segmentation and a publicly available tools (ITK-SNAP, 3D Slicer). The results showed a mean Dice $0.988 \pm 0.003$ 0.988 ± 0.003 , and Hausdorff Distance $4.86 \pm 1.67$ 4.86 ± 1.67 mm in comparison to the manual reference for thigh muscle segmentation, and a mean Dice $0.961 \pm 0.005$ 0.961 ± 0.005 and Hausdorff Distance $2.42 \pm 0.79$ 2.42 ± 0.79 mm for shoulder muscle segmentation, outperformed the other methods. The tool is proposed as slicer module available at . [ABSTRACT FROM AUTHOR]

One of the common eye conditions affecting patients with diabetes is diabetic retinopathy (DR). It is characterised by the progressive impairment to the blood vessels with the increase of glucose level in the blood. The grading efficiency still finds challenging because of the existence of intra-class variations and imbalanced data distributions on the retinal images. Traditional machine learning techniques utilise hand-engineered features for classification of the affected retinal images. As convolutional neural network produces better image classification accuracy in many medical images, this work utilises the CNN-based feature extraction method. This feature has been used to build Gaussian mixture model (GMM) for each class that maps the CNN features to log-likelihood dimensional vector spaces. Since the Gaussian mixture model can be realised as a mixture of both parametric and nonparametric density models and has their flexibility in capturing different data distributions, probabilistic outputs, interpretability, efficient parameter estimation, and robustness to outliers, the proposed model aimed to obtain and provide a smooth approximation of the underlying distribution of features for training the model. Then these vector spaces are trained by the SVM classifier. Experimental results illustrate the efficacy of the proposed model with accuracy 86.3% and 89.1%, respectively. [ABSTRACT FROM AUTHOR]

Canine Hip Dysplasia (CHD) is a congenital disease with a polygenic hereditary component, characterised by abnormal development of the coxo-femoral joint which results in poor coaptation of the femoral head in the acetabulum; the disease rapidly progresses to osteoarthritis of the hip. While dysplasia has been recognised in practically all canine breeds, it is much more common and of concern in medium and large dog breeds with rapid development. Dysplasia in predisposed breeds, particularly the German Shepherd, is the object of screening based on systematic radiological control in some countries. Our collected dataset comprises 507 X-ray images of dogs affected by hip dysplasia (HD). These images were meticulously evaluated using six Deep Convolutional Neural Network (CNN) models. Following an extensive analysis of the top-performing models, VGG16 emerged as the leader, achieving remarkable accuracy, recall, and precision scores of 98.32%, 98.35%, and 98.44%, respectively. Leveraging deep learning (DL) techniques, this approach excels in diagnosing CHD from hip X-rays with a high degree of accuracy. [ABSTRACT FROM AUTHOR]

Markerless motion capture has been made available by the development of pose estimation algorithms that provide both an estimate of the location of body keypoints in two-dimensional images and its associated confidence. It seems relevant to use this additional information for three-dimensional (3D) point reconstruction. Yet, it has been little described, nor has its influence on 3D point reconstruction. Eight participants performed a manual material handling task, which was recorded by 10 video cameras. Each video was processed using OpenPose. Different 3D point reconstruction methods were compared: direct linear transform (DLT) and weighted DLT (wDLT), with 10 cameras and with a subset of 4 cameras. For each keypoint, confidence and position deviation from the 3D reconstructed point with 10 cameras projected in the image reference frame were assessed. Results suggest that using confidence information reduces both average and maximum 3D distance between 3D points reconstructed with 4 and 10 cameras. [ABSTRACT FROM AUTHOR]

Setting up a complex CNN requires a powerful platform, several hours of run-time, and a lot of data for training. Here, we propose a generalised lightweight solution that exploits super-resolution and scalable vector graphics and uses a small-scale UNet as the baseline framework to segment different organs in MR and CT data. We selected the UNet since many researchers use it as the baseline, modify it in their proposal, and perform an ablation study to show the effectiveness of the proposed modification. First, we downsample the input 2D CT slices by bicubic interpolation. Using the architecture of the conventional UNet, we reduce the size of the network's input, and the number of layers and filters to construct a lightweight UNet. The network segments the low-resolution images and prepares the mask of an organ. Then, we upscale the boundary of the output mask by the Scalable Vector Graphics technique to obtain the final border. This design reduces the number of parameters and the run-time by a factor of two. We segmented several tissues to prove the stability of our method to the type of organ. The experiments proved the feasibility of setting up complex deep neural networks with conventional platforms. [ABSTRACT FROM AUTHOR]

Facial Emotion Recognition is gaining interest from researchers in various fields due to its numerous applications. The Vision Transformer (ViT) outperforms convolutional neural network (CNN)-based systems significantly in terms of performance when compared to contemporary image categorisation algorithms. In this context, the primary objective of this paper is to present an innovative and efficient deep learning model for facial emotion recognition. This hybrid intelligent model combines the Vision Transformer with Particle Swarm Optimization (PSO). The proposed model accurately identified the majority of emotions in images, achieving an accuracy of 95.93% for the Fer2013plus dataset and 100% for the CK+ dataset, as demonstrated through experiments conducted on both datasets. According to the results, the proposed facial expression recognition approach has either outperformed or closely matched the performance of state-of-the-art methods. [ABSTRACT FROM AUTHOR]

This paper introduces a Computer-Aided Detection (CAD) system for categorizing breast masses in mammogram images from the DDSM database as Benign, Malignant, or Normal. The CAD process involves Pre-processing, Segmentation, Feature Extraction, Feature Selection, and Classification. Three feature selection methods, namely the Genetic Algorithm (GA), t-test, and Particle Swarm Optimization (PSO) are used. In the classification phase, three machine learning algorithms (kNN, multiSVM, and Naive Bayes) are explored. Evaluation metrics like accuracy, AUC, precision, recall, F1-score, MCC, Dice coefficient, and Jaccard coefficient are used for performance assessment. Training and testing accuracy are assessed for the three classes. The system is evaluated using nine algorithm combinations, producing the following AUC values: GA+kNN (0.93), GA+multiSVM (0.88), GA+NB (0.91), t-test+kNN (0.91), t-test+multiSVM (0.86), t-test+NB (0.89), PSO+kNN (0.89), PSO+multiSVM (0.85), and PSO+NB (0.86). The study shows that the GA and kNN combination outperforms others. [ABSTRACT FROM AUTHOR]