Part-prototype models are explainable-by-design image classifiers, and a promising alternative to black box AI. This paper explores the applicability and potential of interpretable machine learning, in particular PIP-Net, for automated diagnosis support on real-world medical imaging data. PIP-Net learns human-understandable prototypical image parts and we evaluate its accuracy and interpretability for fracture detection and skin cancer diagnosis. We find that PIP-Net's decision making process is in line with medical classification standards, while only provided with image-level class labels. Because of PIP-Net's unsupervised pretraining of prototypes, data quality problems such as undesired text in an X-ray or labelling errors can be easily identified. Additionally, we are the first to show that humans can manually correct the reasoning of PIP-Net by directly disabling undesired prototypes. We conclude that part-prototype models are promising for medical applications due to their interpretability and potential for advanced model debugging.

翻译：部分原型模型是设计上可解释的图像分类器，也是黑箱人工智能的有前景替代方案。本文探讨了可解释机器学习（尤其是PIP-Net）在真实医学影像数据上用于自动化诊断支持的适用性和潜力。PIP-Net学习人类可理解的原型图像部分，我们评估了其在骨折检测和皮肤癌诊断中的准确性和可解释性。研究发现，尽管仅提供图像级类别标签，PIP-Net的决策过程仍符合医学分类标准。由于PIP-Net对原型进行了无监督预训练，数据质量问题（如X光片中不需要的文本或标签错误）可被轻松识别。此外，我们首次证明人类可通过直接禁用不需要的原型来手动修正PIP-Net的推理过程。结论表明，部分原型模型因其可解释性及高级模型调试潜力，在医学应用中具有广阔前景。