Deep learning techniques for detection and prediction of pandemic diseases: a systematic literature review

Multimedia Tools and Applications, Journal Year: 2023, Volume and Issue: 83(2), P. 5893 - 5927

Published: May 29, 2023

Abstract Deep learning (DL) is becoming a fast-growing field in the medical domain and it helps timely detection of any infectious disease (IDs) essential to management diseases prediction future occurrences. Many scientists scholars have implemented DL techniques for pandemics, IDs other healthcare-related purposes, these outcomes are with various limitations research gaps. For purpose achieving an accurate, efficient less complicated DL-based system therefore, this study carried out systematic literature review (SLR) on pandemics using techniques. The survey anchored by four objectives state-of-the-art forty-five papers seven hundred ninety retrieved from different scholarly databases was analyze evaluate trend application areas pandemics. This used tables graphs extracted related articles online repositories analysis showed that good tool pandemic prediction. Scopus Web Science given attention current because they contain suitable scientific findings subject area. Finally, presents forty-four (44) studies technique performances. challenges identified include low performance model due computational complexities, improper labeling absence high-quality dataset among others. suggests possible solutions such as development improved or reduction output layer architecture pandemic-prone considerations.

Language: Английский

---

CataractEyeNet: A Novel Deep Learning Approach to Detect Eye Cataract Disorder

Lecture notes in networks and systems, Journal Year: 2023, Volume and Issue: unknown, P. 63 - 75

Published: Jan. 1, 2023

Language: Английский

---

An Improved COVID-19 Classification Model on Chest Radiography by Dual-Ended Multiple Attention Learning

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2023, Volume and Issue: 28(1), P. 145 - 156

Published: Oct. 13, 2023

As a highly contagious disease, COVID-19 has not only had great impact on the life, study and work of hundreds millions people around world, but also huge global health care system. Therefore, any technical tool that allows for rapid screening high-precision diagnosis infections can be vital help. In order to reduce burden system, computer-aided become current research hotspot. X-ray imaging is common low-cost help with diagnosis. The data used this 15,153 CXR images, containing 10,192 normal lungs, 3,631 positive cases 1,345 images viral pneumonia. For task, we propose dual-ended multiple attention learning model (DMAL). incorporates into both networks, two networks are linked using an integration module. Specifically, in backbone network extract features branch captures local area information; module combines multi-stage features; element, channel spatial prompts focus multi-scale information relevant disease. We evaluate proposed DMAL competitive methods as well ten advanced deep models image domain obtain best performance 99.67%, 99.53%, 99.66%, 99.60% 99.76% terms Accuracy, Precision, Sensitivity, F1 Scores Specificity. method will COVID-19, given general trend such severe infections. Our code available [https://github.com/Graziagh/DMALNet].

Language: Английский

---

Alzr-Net: A Novel Approach to Detect Alzheimer Disease

Published: May 17, 2023

Alzheimer disease is the early stage of dementia that leads to loss memory and other working skills mostly in elderly people. There currently no specific treatment available for Alzheimer's disease, however, detection can prevent worsening symptoms patients. In this work, we used a transfer learning approach accurate patients through MRI scans. We proposed customized approach, named as Alzr-Net, which based on Inception v3 examine effectiveness Alzr-Net diseases. performed extensive experimentation using pretrained models compared performance both type models. The obtained an accuracy, precision, recall, Fl-score 94.38%, 97.24%, 95.49%, 96.36% respectively. also with modern techniques detection, signified model. results above-mentioned metrics illustrated effective technique be employed patients, system reliable implement real-time environments.

Language: Английский

---

An AI healthcare ecosystem framework for Covid-19 detection and forecasting using CronaSona

Medical & Biological Engineering & Computing, Journal Year: 2024, Volume and Issue: 62(7), P. 1959 - 1979

Published: March 13, 2024

Abstract The primary purpose of this paper is to establish a healthcare ecosystem framework for COVID-19, CronaSona. Unlike some studies that focus solely on detection or forecasting, CronaSona aims provide holistic solution, managing data and/or knowledge, incorporating detection, expert advice, treatment recommendations, real-time tracking, and finally visualizing results. innovation lies in creating comprehensive an application not only aids COVID-19 diagnosis but also addresses broader health challenges. main objective introduce novel designed simplify the development construction applications by standardizing essential components required focused addressing diseases. includes two parts, which are stakeholders shared components, four subsystems: (1) management information subsystem, (2) (3) forecasting (4) mobile tracker subsystem. In proposed framework, app. was built try put virus under control. It reactive all users, especially patients doctors. reliable diagnostic tool using deep learning techniques, accelerating referral processes, focuses transmission COVID-19. subsystem monitoring potential carriers minimizing spread. compete with other help people face virus. Upon receiving developed validate test framework’s functionalities. aim application, app., develop techniques avoid increasing spread disease as much possible accelerate detecting features from their chest X-ray images. By CronaSona, human saved stress reduced knowing everything about performs highest accuracy, F1-score, precision, consecutive values 97%, 97.6%, 96.6%. Graphical

Language: Английский

---

Machine Learning-Based Methods for Pneumonia Disease Detection in Health Industry

BENTHAM SCIENCE PUBLISHERS eBooks, Journal Year: 2024, Volume and Issue: unknown, P. 234 - 246

Published: March 26, 2024

Due to partial medical facilities accessible in some developing nations such as India, early disease prediction is challenging. Pneumonia a deadly and widespread respiratory infection affecting the distal airways alveoli. responsible for high mortality rates short- long-term persons of all age groups. The spread mainly depends on immune response system human beings. symptoms vary from person also severity this disease. In 21st century, Artificial Intelligence (AI) recommended one early-stage diagnosis methods. This chapter discusses uses AI subdomains, which Machine learning challenges issues that researchers face while diagnosing pneumonia

Language: Английский

---

Prediction of cross-border spread of the COVID-19 pandemic: A predictive model for imported cases outside China

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(4), P. e0301420 - e0301420

Published: April 9, 2024

The COVID-19 pandemic has been present globally for more than three years, and cross-border transmission played an important role in its spread. Currently, most predictions of spread are limited to a country (or region), models risk assessment remain lacking. Information on imported cases reported from March 2020 June 2022 was collected the National Health Commission China, epidemic data countries origin were websites such as WHO Our World Data. It is proposed establish prediction model suitable prevention control overseas importation COVID-19. Firstly, SIR used fit infection status where exported, r2 values fitted curves obtained above 0.75, which indicated that could well different region. After fitting exporting countries, this basis, SIR-multiple linear regression import combination established, can predict case importation, established overall P <0.05, adjusted R2 = 0.7, indicating SIR-multivariate obtain better results. effectively estimates abroad.

Language: Английский

---

BactPNet: A Novel Automated Detection Approach for Bacterial Pneumonia Patients

Published: March 3, 2023

Every year, a large number of people around the globe, particularly, children die due to pneumonia disease. Approximately, 1.2 million cases have been reported in age ranges from 1 5. Out million, 880,000 died 2016. Therefore, is considered major cause mortality among children, South Asia as well African countries. It top ten causes developed countries, namely, UK, USA, and other European However, an early diagnosis treatment can significantly minimize death rates those countries that high prevalence. The research community has worked diagnose patients using traditional deep learning (DL)-based methods; however, existing approaches various limitations terms accurate detection patients. address above problem, we presented novel DL-based framework, BactPNet, for bacterial Our approach achieved accuracy 91.98%, precision 90%, recall 84%, F1-score 86%. results our confirm it be utilized enhance chest x-ray images. By adopting quality correct prediction further improved. More specifically, experimental findings comparative assessment with techniques show BactPNet better detect adopted by medical experts hospitals.

Language: Английский

---

ViBaNet: A Novel Deep Learning Approach to Detect Bacterial and Viral Pneumonia

Published: Dec. 11, 2023

Viral and bacterial pneumonia mostly occurs in lungs of the humans are life-threatening diseases if timely treatment is ignored. In this regard, age a very crucial aspect as effects different people. Most commonly, infants, well aged people at risk due to these their badly affected. It really challenging, time-consuming task for diagnostician inspect lung's radiographic scans diagnose pneumonia. Binary classification such normal vs or simple, however, detecting viral tricky difficult task. Therefore, we implemented deep-learning method identify victims effectively by employing lung X-rays avoid wrong decisions radiologists. present study, proposed novel technique, ViBaNet, which based on customized residual neural network investigate validity ViBaNet complicated disorders. We conducted experiments uncustomed evaluated efficacy techniques. The obtained an accuracy, precision, recall, F1-score 92.56%, 95.65%, 96.35%, 96%, respectively. above-mentioned analysis results provide evidence that effective be utilized individuals. Moreover, comparison with other techniques has shown superior performance using augmented data.

Language: Английский

---