Key Patches Are All You Need: A Multiple Instance Learning Framework For Robust Medical Diagnosis
This work addresses robustness and interpretability issues in medical diagnosis for applications like skin and breast cancer, though it is incremental as it builds on existing multiple instance learning methods.
The paper tackled the problem of deep learning models in medical image analysis being sensitive to spurious correlations and dataset bias, proposing a multiple instance learning framework that uses only a subset of patches for classification. The results showed that this approach maintains diagnostic performance on in-domain data while improving robustness to demographic shifts and providing detailed explanations.
Deep learning models have revolutionized the field of medical image analysis, due to their outstanding performances. However, they are sensitive to spurious correlations, often taking advantage of dataset bias to improve results for in-domain data, but jeopardizing their generalization capabilities. In this paper, we propose to limit the amount of information these models use to reach the final classification, by using a multiple instance learning (MIL) framework. MIL forces the model to use only a (small) subset of patches in the image, identifying discriminative regions. This mimics the clinical procedures, where medical decisions are based on localized findings. We evaluate our framework on two medical applications: skin cancer diagnosis using dermoscopy and breast cancer diagnosis using mammography. Our results show that using only a subset of the patches does not compromise diagnostic performance for in-domain data, compared to the baseline approaches. However, our approach is more robust to shifts in patient demographics, while also providing more detailed explanations about which regions contributed to the decision. Code is available at: https://github.com/diogojpa99/MedicalMultiple-Instance-Learning.