Search

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]

Medical imaging research has experienced significant growth over the past decade, particularly in the fields of computer vision and pattern recognition. Computational approaches have been proposed to address the challenges in breast cancer detection, classification and segmentation. However, recent advancements in computational technology such as machine learning (ML) methods have dramatically changed the landscape of breast cancer imaging research. This survey aims to provide a compilation of information for future breast cancer imaging researchers by 1) comprehensively examining how various ML techniques are being used to address the main challenges in breast cancer imaging; 2) providing an in-depth discussion and review of publicly available datasets for the development and evaluation of novel breast cancer detection, classification and segmentation approaches; and 3) outlining current evaluation metrics used by breast cancer imaging researchers. The insights and findings presented in this survey will serve as a valuable resource for researchers and clinicians interested in breast cancer imaging. By providing an overview of the current state-of-the-art techniques and highlighting areas for future research, we hope to facilitate the development of more accurate and effective breast cancer imaging ML techniques, and contribute to advancing our understanding and improving the diagnosis and treatment of breast cancer. [ABSTRACT FROM AUTHOR]

Augmented reality (AR) is a dynamic field intersecting various disciplines, with applications spanning medicine and diagnostics. Our focus on advancing medical imaging has led to groundbreaking progress in brain tumor diagnosis, fostering optimism in treatment. In this pursuit, we introduce HoloBrain, an accessible AR application designed to transform medical education. Utilizing smartphones, HoloBrain enables automatic segmentation, AR visualization, and 3D interaction without the need for head-mounted devices. Emphasizing practicality over hyper-realism, our innovation allows medical students to seamlessly integrate a virtual brain tumor model into their surroundings. This immersive approach facilitates exploration of the 3D representation of an afflicted brain, enhancing educational experiences. Tests in a controlled environment using a phantom head yielded promising results, with our solution demonstrating average errors of 2 mm to 4.5 mm on smartphone displays. These outcomes highlight the precision and effectiveness of our contribution, underlining its potential in transforming medical education and training. HoloBrain represents a step forward in making AR accessible and impactful for aspiring medical professionals, offering a valuable tool for enhancing their understanding of complex medical scenarios. [ABSTRACT FROM AUTHOR]

In recent years, with the rapid advancements in deep learning, image processing has progressively attracted the attention of medical researchers to achieve accurate diagnoses of numerous human diseases in a non-invasive manner. Image processing systems are among the most helpful and crucial disease identification technologies. Liver cancer is the world's fifth most prevalent and leading cause of cancer deaths. Diagnoses and treatment are essential because of the high mortality rate and recurrence after treatment. Medical image analysis is extremely beneficial in the healthcare industry. In the last several years, creating and developing computer-assisted imaging systems has been increasingly vital to aid physicians or doctors in strengthening their diagnoses. This research presents a deep model to identify cancerous regions in the liver aims using Computerized Tomography (CT) scans from the Kaggle dataset. Basic preprocessing is applied to CT images from the dataset acquired from the Kaggle repository. The performance of the proposed model is evaluated using different performance metrics, and an accuracy score of 94.3% was achieved in tumour identification. [ABSTRACT FROM AUTHOR]

This study aims to tackle the implementation of a single deep learning (DL) technique to the brain tumour recognition, segmentation and classification problem based on MR images. We propose utilising the Mask R-CNN approach along with transfer learning models on images. This was done by assigning bounding boxes and successfully constructing a border on each image for the tumour volume in order to distinguish it from neighbouring tissues and structures. By processing the images in this way, a single DL model could accurately identify, classify and segment three different categories of brain tumours (meningioma, glioma or pituitary). ResNet-50 and ResNet-101 network architectures were trained for 1000 epochs, and their classification and segmentation performances were assessed. A higher classification success was achieved for the ResNet-101 backbone pretrained on COCO images (%75 accuracy) in the testing phase. ResNet-50, on the other hand, achieved a higher classification accuracy (%87) when used with the ImageNet dataset. The classification and segmentation successes of meningioma and pituitary tumours were comparable, while glioma tumours could be segmented at relatively lower success rates (41–73% for all models), despite similar classification performances (73–95%) with other tumour types. [ABSTRACT FROM AUTHOR]

A brain tumour is fatal; thus, early discovery is crucial to preserving life. Brain magnetic resonance image (MRI) segmentation can be used to find the tumour area. Radiologic assessments can pinpoint a suspected brain tumour's exact location and size. For future diagnostic and treatment planning, the assessment report is crucial. For the goal of diagnosis, the tumour detection process must be quick and precise. This paper suggests an automated system to distinguish between malignant brain MRIs and those not. Each image is assigned a set of characteristics utilising statistical, edge. Then, to compare the precision of the suggested framework, the Extreme Learning Machine (ELM) classifier is used with various cross-validations on the feature set. The BRATS 2020 dataset is used to validate the suggested technique. The approach has an average accuracy, sensitivity, and specificity of 99.1%, 99.9%, and 99.3%. It can detect tumours more precisely and with less processing time than current techniques. This study suggests a powerful method for identifying brain cancers and predicting the kind of tumour. The suggested method first uses the adaptive histogram equalisation strategy to boost the resolution of BRATS-2020 MRI images. The goal is to improve segmentation performance as a result. After that, the WFCM technique is used to segment the MRI picture. [ABSTRACT FROM AUTHOR]

This work uses Fuzzy Lattices and Neural Reinforcement Learning techniques for seizure classification. EEG database of Bonn University and CHB-MIT has been used for the evaluation of the proposed method. Here, features are represented in the form of fuzzy lattices, and feature vectors are created in the form of Kinetic Energy (K.E.) using Schrödinger equation. Then, the highest K.E.-based seven fuzzy lattices have been used for classification using Neural Reinforcement Learning classifier. Neural Reinforcement Learning classifier is a self-learner method that classifies different seizure sub-classes (healthy eyes open, healthy eye closed, Epileptogenic Zone, hippocampal formation of opposite hemisphere, epileptic seizure). The effectiveness of the proposed method has been tested on two different public datasets. The average classification accuracy achieved is 97.6% and 97.5% for Bonn and CHB-MIT datasets, respectively. Results are compared with existing techniques to show the precedence of proposed approach. Also, computation speed of proposed classifier is more than 1.5 times compared to Fuzzy Q Learning classifier. The objective is to develop a hybrid model integrating fuzzy lattices and neural reinforcement learning (adaptive methods) for accurate classification, aiming to enhance and speeding up seizure detection and diagnosis to improve patient care through advanced computational techniques. [ABSTRACT FROM AUTHOR]

This paper intends to introduce a novel sugarcane plant disease prediction. There are three stages to the projected method, namely: (i) Preprocessing, (ii) Feature extraction, and (iii) Disease detection phase. Initially, the input image is given to the pre-processing stage in which the grey transformation and bilinear interpolation are carried out. The grey transformation is undergone to improve the clarity of the picture and the bilinear interpolation is used as a re-sampling technique. The feature extraction step follows the pre-processed picture, where both texture and colour features are retrieved. More specifically, the texture features like GLCM and suggested Distance-based LBP (D-LBP) features will be extracted. Subsequently, the extracted features are subjected to the disease detection phase process, where a hybrid classifier is introduced. This hybrid classifier blends the concepts of 'Neural Network (NN) and Support Vector Machine (SVM)' for better prediction results. In order to verify the effectiveness of the work that is suggested, a comparison analysis is conducted comparing the suggested and current approaches. The proposed method achieves highest accuracy in 80% of learning which is 13.97%, 16.12%, 11.82%, and 12.9% better than the other methods such as SVM, DT, RF, and NN, respectively. [ABSTRACT FROM AUTHOR]

Attention is a cognitive process that is essential for human performance. Human development frequently depends on attention; however, this topic still requires additional research. Recently, EEG brain waves have been utilised to identify a person's attention states. This work aims to review numerous machine learning algorithms to analyse the Electroencephalographic (EEG) data to recognise human attention. Various machine learning approaches for the analysis of emotional states with EEG data were reviewed. Moreover, the analysis includes the performance achieved in various works; their benefits and disadvantages are reviewed. An EEG data processing pipeline and a review of human attention recognition are developed to evaluate the performance of various machine learning techniques. According to the analysis, the neural network framework achieves 99.81% accuracy. The results serve as a design framework for future systems using EEG data on brain activity to monitor people's health. [ABSTRACT FROM AUTHOR]

This paper has aimed to develop the Breast Cancer Detection (BCD) technique along with the mammography reports of the patients. For this proposed framework, the required images are collected from publicly available datasets. Further, the collected images are given into the phase of feature extraction, where the deep features based on texture features, ResNet, morphological features and density features are extracted and fused together. In addition, the weighted fused feature selection is performed with the fused features, in which the weight is optimised with the Modified Hybrid based Leader Optimization (MHLBO) obtained from Hybrid Leader-Based Optimization(HLBO) to elevate the feature performance on classification. Then, breast cancer can be classified with the developed Adaptive Cascaded Deep Network (ACDN) by combing the classification results of, Recurrent Neural Network (RNN), Gate Recurrent Unit (GRU), and Long Short-Term Memory (LSTM). Here, the parameter tuning has occurred in all three classifiers using the same enhanced HLBO. Throughout the experimentation, the designed model shows 97% and 96% regarding accuracy and F1-score, respectively. The tentative outcomes show the efficacy of the designed BCD method is compared with other baseline approaches. [ABSTRACT FROM AUTHOR]

Corneal Ulcers are defined as inflammation or even infection. They are one of the most frequent diseases that affect eye health. The proper measurement of corneal ulcer lesions enables the physician to evaluate the treatment effectiveness and decision making. This paper presents a segmentation method that aims to assist doctors in monitoring the treatment of corneal ulcer lesions. We applied a stepwise fine-tuning in U-Net Convolutional Neural Network architecture to train a model with 358 Point-flaky corneal ulcer images. The result from the model using U-Net architecture is then submitted to post-processing operations. Based on experiments performed with 91 Flaky corneal ulcer images, our approach achieved 0.823 of the Dice Coefficient on average, 88.9% of Recall, 99.4% of Specificity, and True Dice Coefficient of 0.835. The results are promising; we then have evidence that the proposed method could generalise the data from the training phase to segment the test data. [ABSTRACT FROM AUTHOR]

The Multimodal Medical Image Fusion (MMIF) is affected by poor image quality, which leads to the extraction of inefficient features. The main intent of this work is to fuse various planes in the PET and MRI medical images efficiently using the MMIF approach. Initially, the sample images containing the axial plane of PET and MRI images are aggregated from standard datasets. Then, the collected images are employed for the decomposition process, which is accomplished via Optimal Non-Subsampled Contourlet Transform (ONSCT). The parameters in the NSCT are optimized using the Modified Water Strider Algorithm (MWSA. Once the images are decomposed, it is segmented into two sub-bands as high frequency and low-frequency sub-bands. Consequently, the high-frequency sub-bands of both PET and MRI images are fused by using the optimal weighted average fusion, in which the weight factor is obtained optimally by the MWSA algorithm. Similarly, the low-frequency sub-bands of both medical images are combined by sparse fusion technique. Finally, both the resultant fused images are subjected to Inverse Non-Subsampled Contourlet Transform (INSCT) to get desired fused images. The experimental findings suggest that the proposed model has effectively fused the images, and it also enhances the similarity score with axial planes. [ABSTRACT FROM AUTHOR]

Some of the viruses may cause lung parenchyma and airway involvement. Usually, viral pneumonia causes ground-glass opacities, bilateral peripheral distribution, consolidation, vascular thickening, and reticular opacity. These features are common in COVID-19 rather than Non-Covid-19 viral pneumonia. However, in advanced cases, COVID-19 viral pneumonia may cause organising pneumonia and fibrosis of the lung. Atypical findings of Non-Covid-19 pneumonia have included central peripheral distribution, pleural effusion, lymphadenopathy, nodules, tree-in-bud opacities, and pneumothorax. Therefore, differentiating Non-Covid-19 pneumonia from COVID-19 pneumonia at chest computed tomography (CT) is necessary. In that case, CT scans of the thorax are one of the essential tools for early identification and future prognosis of viral pneumonia. We have proposed a Computer-Aided Diagnostic (CAD) system that can detect features of chest CT using a Deep Neural Network (DNN) with Long Short-Term Memory (LSTM). Transfer learning using pre-trained DNN models (ResNet50, VGG19, InceptionV3, Xception, DenseNet121, and VGG16) is applied to retain both high-level and low-level features effectively. The deep features are passed to the LSTM layer. The LSTM is utilised as a classifier and detects long short-term dependencies. The proposed method employs a hybrid DNN-LSTM network for automatic detection to take advantage of the uniqueness of the two models. The proposed models are trained with common and different features present in the chest CT of COVID-19 and Non-Covid-19 viral pneumonia. The 5-fold cross-validation (CV) method validated and tested the proposed model. The proposed DNN model's performance is quite improved with LSTM and CV. As a result, the proposed LM-DNN (VGG16+LSTM+CV) model has achieved the classification test accuracy of 91.58% and specificity of 93.86%, which offers superior performance with state-of-the-art. Also, the DenseNet121+LSTM+CV model has reached the classification test accuracy of 90.1% and sensitivity of 92%. [ABSTRACT FROM AUTHOR]

Upper-limb prostheses are either too expensive for many consumers or have a greatly simplified choice of actions. This research aims to enable an improvement to the quality of life for recipients of these devices by providing an understanding of how we use our hands in modern everyday life. To achieve this a taxonomy of grasps used in activities of daily living has been created. A portable motion capture system was developed to collect data. Thirteen participants' hand movements were recorded (totalling 62 hours and 10 minutes of data). From these data, 38 and 22 grasps were observed from the left and right hands, respectively. The portable system effectively captured natural hand motions, which formed an updated taxonomy of grasps. [ABSTRACT FROM AUTHOR]

Skin melanoma is a potentially fatal form of cancer. If left untreated or allowed to spread, it can lead to death or serious disability. Therefore, early diagnosis is crucial for improving patient prognosis and outcomes. Recent advances in machine learning (ML) and deep learning (DL) have greatly contributed to the categorisation and identification of melanoma. The goal of this survey is to evaluate 60 publications that have been submitted in order to create an overview of the melanoma detection process. It examines several feature extraction techniques.The evaluation concentrates on different melanoma detection methods, including 'deep learning (DL)' and 'machine learning (ML)' models. The analysis includes performance metrics and a review of the results obtained from the PH2 dataset, which achieved a high accuracy of 96.5%. Finally, the survey addresses research gaps to facilitate future investigations into melanoma identification strategies. [ABSTRACT FROM AUTHOR]

A virus known as the Coronavirus 2019 (COVID-19) can have serious effects on a variety of body parts. Preventing this spread of the virus requires early detection and proper treatment. Many methods are followed for early detection of this disease based on symptoms and classification for proper treatment. In this paper, classification of COVID-19 using Computed Tomography (CT) scan images with hybrid optimisation-enabled deep learning is researched. Here, a hybrid optimisation algorithm called War Snake Optimisation Algorithm (WSOA) is proposed to train Deep Q-Network (Deep Q-Net). Moreover, the proposed WSOA is formed by integrating War strategy optimisation (WSO) and Snake Optimiser (SO). In this paper, pre-processing is carried out using a median filter, followed by lung lobe segmentation that is done by Generative Adversarial Network (GAN). Furthermore, image augmentation is done by translation, rotation, flipping and scaling. Then, feature extraction is followed after image augmentation, where needed features get extracted to end up with the classification of COVID-19. This classification brings out the patient's health condition of normal and abnormal situations. Moreover, this research work is analysed with three performance metrics, such as accuracy, sensitivity and specificity with values of 93.7%, 94.8% and 89.5%. [ABSTRACT FROM AUTHOR]