Authors: Maryam Ahmed, Tooba Bibi, Rizwan Ahmed Khan, Sidra Nasir

Published on: April 05, 2024

Impact Score: 8.2

Arxiv code: Arxiv:2404.03892

Summary

• What is new: This study presents a novel framework that seamlessly integrates Convolutional Neural Networks (CNNs) and Explainable Artificial Intelligence (XAI) to improve the accuracy of breast cancer diagnosis while making the AI decision-making process understandable to healthcare professionals.
• Why this is important: The challenge in the automated diagnosis of breast cancer lies in not only accurately classifying mammographic images but also in providing interpretable explanations for the AI’s decisions to gain the trust of medical practitioners.
• What the research proposes: The research utilizes a fine-tuned ResNet50 architecture combined with XAI methods like Grad-CAM, LIME, and SHAP, aimed at making CNN decisions transparent and understandable for healthcare professionals.
• Results: The findings underscore the potent synergy between CNNs and XAI, leading to improved diagnostic methods for breast cancer and fostering better collaboration between AI systems and medical practitioners.

Technical Details

Technological frameworks used: ResNet50, VGG-16, DenseNet

Models used: CNN, Grad-CAM, LIME, SHAP

Data used: CBIS-DDSM dataset

Potential Impact

Healthcare diagnostics, Medical Imaging Companies, AI-based HealthTech Startups

Want to implement this idea in a business?

We have generated a startup concept here: MediXAI.