Research Article
Automated Detection and Classification of Multiple Diseases Using Deep Neural Networks
@INPROCEEDINGS{10.4108/eai.19-12-2025.2362548, author={Ratna Priya Singh and Arnav Mazumdar and Arnav Agrawal and Sumit Kumar and Surbhi Vijh}, title={Automated Detection and Classification of Multiple Diseases Using Deep Neural Networks}, proceedings={Proceedings of the 1st International Conference on Computational Techniques in Data Science, IICTDS 2025, 19-20 December 2025, Chandigarh, India}, publisher={EAI}, proceedings_a={IICTDS}, year={2026}, month={11}, keywords={Brain Disease Skin Disease Lung Disease}, doi={10.4108/eai.19-12-2025.2362548} }- Ratna Priya Singh
Arnav Mazumdar
Arnav Agrawal
Sumit Kumar
Surbhi Vijh
Year: 2026
Automated Detection and Classification of Multiple Diseases Using Deep Neural Networks
IICTDS
EAI
DOI: 10.4108/eai.19-12-2025.2362548
Abstract
Diseases are health problems which negatively impact the normal functioning of body and may involve one or more organs. This research targets three groups of diseases: pulmonary diseases from X-rays, neurological conditions like brain tumours from MRI scans, and skin diseases from dermoscopic images. Identifying these conditions accurately helps doctors choose the right treatment more effectively. We aim to create a smart model using deep learning methods to identify different diseases. Current methods rely on separate neural networks, where a dataset with multiple data types is given to different CNN architectures. Given a single CNN may not extract the right features, we propose a multi-modal approach using specific CNNs for each data type. Our plan involves three specialised CNN architectures: DenseNet-121 for chest X-ray analysis, ResNet-50 for brain MRI interpretation, and EfficientNet-B3 for skin lesion classification. Each CNN will be then combined for better accuracy in the disease detection process.