Downloads: 0
Survey Paper | Computer Science | Volume 15 Issue 4, April 2026 | Pages: 270 - 274 | India
Explainable AI in Medical Imaging: A Review of Black-Box Deep Learning Models
Abstract: The increasing volume of medical image data has generated a significant requirement for automated and interpretable medical image diagnosis tools. In this research, a medical image detection system based on deep learning has been proposed for the detection of brain tumors, lung diseases, and cardiac diseases. Various types of medical images, including Brain Tumor MRI images, Cardiac MRI images (CAD), and COVID-19 Radiography images, are integrated to form a unified framework. Before training the model, image preprocessing and harmonization techniques are applied to the dataset. Advanced deep learning architectures, especially Convolutional Neural Networks (CNNs), have been employed to accurately classify medical images. In order to increase the transparency of the proposed model, Explainable Artificial Intelligence (XAI) techniques, including Grad-CAM, LIME, and SHAP, have been integrated. Both visual and interpretable explanations are provided by the proposed model. Furthermore, a user-friendly interface has been created using Streamlit to allow clinicians to upload medical images and visualize the results. The performance of the proposed system has been evaluated using clinical performance metrics, and the results show that the proposed system can be considered an effective and reliable solution.
Keywords: Brain MRI, X-ray Radiography, Cardiac MRI, Medical Imaging, Deep Learning, Explainable AI, Clinical Decision Support
How to Cite?: Mona, S. S. Sanjana, Sangati Ganga Mahija, Spandhana K Devadiga, "Explainable AI in Medical Imaging: A Review of Black-Box Deep Learning Models", Volume 15 Issue 4, April 2026, International Journal of Science and Research (IJSR), Pages: 270-274, https://www.ijsr.net/getabstract.php?paperid=SR26327113406, DOI: https://dx.dx.doi.org/10.21275/SR26327113406