Muna Al-Khaifi, GP Oncologist at Sunnybrook and Assistant Professor at the University of Toronto, shared a post on LinkedIn:
“Detection of Breast Cancer Using Machine Learning and Explainable AI
A recent study by Arravalli et al., 2025 explores how machine learning (ML) and explainable artificial intelligence (XAI) can enhance the accuracy and interpretability of breast cancer diagnosis.
Using diagnostic features of patients, the researchers applied several ML models, and Random Forest delivered the highest performance with an F1-score of 84%, followed by a stacked ensemble model at 83%. What makes this work especially impactful is the use of XAI methods like SHAP, LIME, ELI5, Anchor, and QLattice to reveal the reasoning behind each prediction.
This transparency helps build trust in AI systems and supports clinicians in making more informed and accurate decisions, potentially reducing diagnostic errors in the detection of breast cancer.
This study highlights the growing role of interpretable AI in medical diagnostics, paving the way for safer and more effective AI-assisted healthcare.”
Title: Detection of breast cancer using machine learning and explainable artificial intelligence
Authors: Tharunya Arravalli, Krishnaraj Chadaga, H Muralikrishna, Niranjana Sampathila, D. Cenitta, Rajagopala Chadaga, K. S. Swathi
Read the Full Article in Nature.
More posts featuring Muna Al-Khaifi on OncoDaily.