Indian Journal of Science and Technology
DOI: 10.17485/IJST/v16i34.1306
Year: 2023, Volume: 16, Issue: 34, Pages: 2730-2739
Original Article
Kavita Thakur1, Navneet Kaur Sandhu2, Yogesh Kumar3*
1Research Scholar, Desh Bhagat University, Mandi Gobindgarh, Punjab, India
2Assistant Professor, Desh Bhagat University, Mandi Gobindgarh, Punjab, India
3Department of CSE, School of Technology, Pandit Deendayal Energy University, Gandhinagar, Gujarat, India
*Corresponding Author
Email: [email protected]
Received Date:30 May 2023, Accepted Date:07 August 2023, Published Date:14 September 2023
Objectives: This study explores the potential of deep learning-based techniques to improve disease management and intervention by focusing on their use in infectious disease prediction and prognosis. Methods: The research used deep learning models EfficientNetB0, NASNetLarge, DenseNet169, ResNet152V2, and InceptionResNetV2. For this study, a dataset comprising 29,252 images of different diseases such as COVID-19, MERS, Pneumonia, SARS, and tuberculosis. To visualize pixel intensity, exploratory data analysis was performed on the pictures. Preprocessing eliminated disruptive signals via image augmentation and contrast enhancement. After that, Otsu thresholding and contour feature morphological values retrieved relevant features. Findings: The best successful model was found to be EfficientNetB0. During training, it obtained a 90.22% accuracy rate, a loss of 0.279, having an RMSE value of 0.578. However, InceptionResNetV2 showed the best accuracy, loss, and RMSE values throughout model testing. The precise accuracy, loss, and RMSE results were 88%, 0.399, and 0.631, respectively. Novelty: The novelty resides in exploring methods based on deep learning for predicting and prognosticating infectious diseases, with the potential for handling diseases, strategies for intervention, and public health decisions.
Keywords: Tuberculosis; Pneumonia; Infectious diseases; Deep learning; InceptionResNetV2
© 2023 Thakur et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Published By Indian Society for Education and Environment (iSee)
Subscribe now for latest articles and news.