A Hybrid Convolutional Neural Network Model for Diagnosis of COVID-19 Using Chest X-ray Images

COVID-19 declared as a pandemic that has a faster rate of infection and has impacted the lives and the country’s economy due to forced lockdowns. Its detection using RT-PCR is required long time and due to which its infection has grown exponentially. This creates havoc for the shortage of testing kits in many countries. This work has proposed a new image processing-based technique for the health care systems named “C19D-Net”, to detect “COVID-19” infection from “Chest X-Ray” (XR) images, which can help radiologists to improve their accuracy of detection COVID-19. The proposed system extracts deep learning (DL) features by applying the InceptionV4 architecture and Multiclass SVM classifier to classify and detect COVID-19 infection into four different classes. The dataset of 1900 Chest XR images has been collected from two publicly accessible databases. Images are pre-processed with proper scaling and regular feeding to the proposed model for accuracy attainments. Extensive tests are conducted with the proposed model (“C19D-Net”) and it has succeeded to achieve the highest COVID-19 detection accuracy as 96.24% for 4-classes, 95.51% for three-classes, and 98.1% for two-classes. The proposed method has outperformed well in expressions of “precision”, “accuracy”, “F1-score” and “recall” in comparison with most of the recent previously published methods. As a result, for the present situation of COVID-19, the proposed “C19D-Net” can be employed in places where test kits are in short supply, to help the radiologists to improve their accuracy of detection of COVID-19 patients through XR-Images.

Wearable IoT based diagnosis of prostate cancer using GLCM-multiclass SVM and SIFT-multiclass SVM feature extraction strategies

Purpose This study has mainly aimed to compare and contrast two completely different image processing algorithms that are very adaptive for detecting prostate cancer using wearable Internet of Things (IoT) devices. Cancer in these modern times is still considered as one of the most dreaded disease, which is continuously pestering the mankind over a past few decades. According to Indian Council of Medical Research, India alone registers about 11.5 lakh cancer related cases every year and closely up to 8 lakh people die with cancer related issues each year. Earlier the incidence of prostate cancer was commonly seen in men aged above 60 years, but a recent study has revealed that this type of cancer has been on rise even in men between the age groups of 35 and 60 years as well. These findings make it even more necessary to prioritize the research on diagnosing the prostate cancer at an early stage, so that the patients can be cured and can lead a normal life. Design/methodology/approach The research focuses on two types of feature extraction algorithms, namely, scale invariant feature transform (SIFT) and gray level co-occurrence matrix (GLCM) that are commonly used in medical image processing, in an attempt to discover and improve the gap present in the potential detection of prostate cancer in medical IoT. Later the results obtained by these two strategies are classified separately using a machine learning based classification model called multi-class support vector machine (SVM). Owing to the advantage of better tissue discrimination and contrast resolution, magnetic resonance imaging images have been considered for this study. The classification results obtained for both the SIFT as well as GLCM methods are then compared to check, which feature extraction strategy provides the most accurate results for diagnosing the prostate cancer. Findings The potential of both the models has been evaluated in terms of three aspects, namely, accuracy, sensitivity and specificity. Each model’s result was checked against diversified ranges of training and test data set. It was found that the SIFT-multiclass SVM model achieved a highest performance rate of 99.9451% accuracy, 100% sensitivity and 99% specificity at 40:60 ratio of the training and testing data set. Originality/value The SIFT-multi SVM versus GLCM-multi SVM based comparison has been introduced for the first time to perceive the best model to be used for the accurate diagnosis of prostate cancer. The performance of the classification for each of the feature extraction strategies is enumerated in terms of accuracy, sensitivity and specificity.

Leaf Diseases Classification on Peanut Leaves Based on Texture and Colour Features

Peanuts are a food crop commodity that Indonesians widely consume as a vegetable fat and protein source. However, the quality and quantity of peanut productivity may decline, one of which is due to plant diseases. Efforts that can be made to maintain peanut productivity are the application of technology to detect peanut plant diseases early; thus, disease control can be carried out earlier. This study presents a technology development application, particularly digital image processing, to identify disease features of infected peanut leaves based on GLCM texture features and colour features in the HSV colour space and classified using the SVM method. The development of the SVM method that is applied is the Multiclass SVM with the DAGSVM strategy, which can classify more than two classes. Based on the experimental results, it confirms that the combination of HSV colour features and GLCM texture features with an angular orientation of 0 degrees and classified by the Multiclass SVM method with polynomial kernels produces the highest accuracy, i.e. 99.1667% for leaf spot class, 97.5% for leaf rust class, 98.8333% for eyespot class, 100% for normal leaf class and 100% for other leaf class.

Fatigue Detection on Face Image Using FaceNet Algorithm and K-Nearest Neighbor Classifier

Background: The COVID-19 pandemic has made people spend more time on online meetings more than ever. The prolonged time looking at the monitor may cause fatigue, which can subsequently impact the mental and physical health. A fatigue detection system is needed to monitor the Internet users well-being. Previous research related to the fatigue detection system used a fuzzy system, but the accuracy was below 85%. In this research, machine learning is used to improve accuracy.Objective: This research examines the combination of the FaceNet algorithm with either k-nearest neighbor (K-NN) or multiclass support vector machine (SVM) to improve the accuracy.Methods: In this study, we used the UTA-RLDD dataset. The features used for fatigue detection come from the face, so the dataset is segmented using the Haar Cascades method, which is then resized. The feature extraction process uses FaceNet's pre-trained algorithm. The extracted features are classified into three classes—focused, unfocused, and fatigue—using the K-NN or multiclass SVM method.Results: The combination between the FaceNet algorithm and K-NN, with a value of resulted in a better accuracy than the FaceNet algorithm with multiclass SVM with the polynomial kernel (at 94.68% and 89.87% respectively). The processing speed of both combinations of methods has allowed for real-time data processing.Conclusion: This research provides an overview of methods for early fatigue detection while working at the computer so that we can limit staring at the computer screen too long and switch places to maintain the health of our eyes.

Study of Dementia Severity Prognosis in Mr Images Using Grey Wolf Optimization Based Dual Deep Learning Technique

Prognosis of in a dementia disorder is a tedious task in preclinical stage. Ventricle pathology changes in dementia appear to be overlapped for neuro degeneration in brain. Identification of these overlaps among the groups severity helps to understand the pathogenesis of this disorder. In this work impact of changes in ventricle region on severity stages of dementia is observed using dual deep learning techniques (DDLT). Alzheimer's Disease Neuroimaging Initiative (ADNI) database that contains 1169 MR images are used in this study. Segmentation of ventricle region is carried out using multilevel threshold based Grey Wolf Optimization (GWO) technique. The feature vectors obtained from combined AlexNet and ResNet are analysed. The fused feature vectors are given to support vector machine (SVM) to observe the severity changes. Consequently, symmetry analysis of ventricle is carried out to perceive the distinctive changes in progression. The obtained results show that ventricle region is accurately delineated from other region with optimized thresholds. The segmented ventricle shows better correlation for all considered classes (> 0.9). It is observed that DDLT with multiclass SVM provides an improved accuracy of about 79.87% compared to individual transfer learning such as AlexNet (74%) and ResNet (76.53%). Further, symmetry analysis shows that left side ventricle with DDLT features shows an improved performance than right side for onset stages. Further, clinical correlation of left ventricle seems to be statically significant (p<0.0001) which prominently differentiate dementia severity variations. This framework is more prominent and clinically useful to identify the distinct ventricle region variation in dementia.