Applications of deep learning in disease diagnosis of chest radiographs: A survey on materials and methods

Recent advances in deep learning have given rise to high performance in image analysis operations in healthcare. Lung diseases are of particular interest, as most can be identified using non-invasive image modalities. Deep learning techniques such as convolutional neural networks, convolution autoencoders, and graph convolutional networks have been implemented in several pulmonary disease identification applications, e.g., lung nodule classification, Covid-19, and pneumonia detection. Various sources of medical images such as X-rays, computed tomography scans, magnetic resonance imaging, and positron emission tomography scans make deep learning techniques favorable to identify lung diseases with great accuracy. This paper discusses state-of-the-art methods that use deep learning on various medical imaging modalities to detect and classify diseases in the lungs. A description of a few publicly available databases is included in this study, along with some distinct deep learning techniques developed in recent times. Furthermore, several challenges and open research areas for pulmonary disease diagnosis using deep learning are discussed. The objective of this work is to direct researchers in the field of diagnosis of lung diseases.

Clinico-epidemiological Characteristics of Asymptomatic and Symptomatic COVID-19-Positive Patients in Bangladesh.

As of August 15, 2020, Bangladesh lost 3591 lives since the first Coronavirus disease 2019 (COVID-19) case announced on March 8. The objective of the study was to report the clinical manifestation of both symptomatic and asymptomatic COVID-19-positive patients. An online-based cross-sectional survey was conducted for initial recruitment of participants with subsequent telephone interview by the three trained physicians in 237 adults with confirmed COVID-19 infection in Bangladesh. The study period was 27 April to 26th May 2020. Consent was ensured before commencing the interview. Collected data were entered in a pre-designed case record form and subsequently analyzed by SPSS 20.0. The mean±SD age at presentation was 41.59±13.73 years and most of the cases were male (73.0%). A total of 90.29% of patients reside in urban areas. Among the positive cases, 13.1% (n=31) were asymptomatic. Asymptomatic cases were significantly more common in households with 2 to 4 members (p=0.008). Both symptomatic and asymptomatic patients shared similar ages of presentation (p=0.23), gender differences (p=0.30) and co-morbidities (p=0.11). Only 5.3% of patients received ICU care during their treatment. The most frequent presentation was fever (88.3%), followed by cough (69.9%), chest pain (34.5%), body ache (31.1%), and sore throat (30.1%). Thirty-nine percent (n=92) of the patients had co-morbidities, with diabetes and hypertension being the most frequently observed. There has been an upsurge in COVID-19 cases in Bangladesh. Patients were mostly middle-aged and male. Typical presentations were fever and cough. Maintenance of social distancing and increased testing are required to meet the current public health challenge.

Towards Automated Multiclass Severity Prediction Approach for COVID-19 Infections Based on Combinations of Clinical Data

The recent dramatic expansion of the COVID-19 outbreak is placing enormous strain on human society as a whole. Numerous biomarkers are being investigated in an effort to track the condition of the patient. This could interfere with signs of many other illnesses, making it more difficult for a specialist to diagnose or predict the severity level of the case. As a result, the focus of this research was on the development of a multiclass prediction system capable of dealing with three severity cases (severe, moderate, and mild). The lymphocyte to CRP ratio (C-reactive protein blood test) and SpO2 (blood oxygen saturation level) indicators were ranked and used as prediction system attributes. A machine learning model based on SVMs is created. A total of 78 COVID-19 patients were recruited from the Azizia primary health care sector/Wasit Health Directorate/Ministry of Health to form different combinations of COVID-19 clinical dataset. The outcomes demonstrate that the proposed approach had an average accuracy of 82%. The established prediction system allows for the early identification of three severity cases, which reduces deaths. © 2022 Ahmed M. Dinar et al.

A Deep Neural Network for Classification of Thoracic Diseases on Chest Radiography and Covid-19 Detection

This paper presents COVID-Net, an Artificial intelligent system that can detect COVID-19 from chest X-rays based on machine learning. COVID-Net is a 3-stage machine learning (ML) system. COVID-Net is a system built on a convolutional neural network trained on over 10,000 frontal view X-ray images. The merit of this system is that it detects COVID-19 from other kinds of diseases and can be used to diagnose a new type of viral or bacterial pneumonia. © 2021 IEEE.

Early Detection of COVID-19 Patients using Chromagram Features of Cough Sound Recordings with Machine Learning Algorithms

This paper presents a cough sound-based fast, automated, and noninvasive COVID-19 detection system to discriminate the cough sounds of the COVID-19 patients from the healthy individuals. The proposed system extracts an acoustic feature called chromagram from the cough sound samples and applies it to the input of a classifier algorithm. Two artificial neural network (ANN) based classifiers namely convolutional neural network (CNN) and deep neural network (DNN) are modeled for this purpose. The simulation results show that the proposed system achieves an accuracy of 92.9% and 91.7% with CNN and DNN respectively. The performance comparison of the proposed system with two popular machine learning algorithms namely support vector machine (SVM) and k-nearest neighbor (kNN) are also presented in this work. © 2021 IEEE.