Localization supervision of chest x-ray classifiers using label-specific eye-tracking annotation
This addresses the costly annotation issue for medical imaging interpretability, though it is incremental as it builds on existing eye-tracking methods.
The paper tackled the problem of limited annotated bounding boxes for chest x-ray classification by using eye-tracking data from radiologists to supervise localization, improving model interpretability without affecting classification performance.
Convolutional neural networks (CNNs) have been successfully applied to chest x-ray (CXR) images. Moreover, annotated bounding boxes have been shown to improve the interpretability of a CNN in terms of localizing abnormalities. However, only a few relatively small CXR datasets containing bounding boxes are available, and collecting them is very costly. Opportunely, eye-tracking (ET) data can be collected in a non-intrusive way during the clinical workflow of a radiologist. We use ET data recorded from radiologists while dictating CXR reports to train CNNs. We extract snippets from the ET data by associating them with the dictation of keywords and use them to supervise the localization of specific abnormalities. We show that this method improves a model's interpretability without impacting its image-level classification.