Learn like a Pathologist: Curriculum Learning by Annotator Agreement for Histopathology Image Classification
This work addresses the problem of improving classification accuracy for medical professionals in histopathology, but it is incremental as it adapts an existing curriculum learning concept to a specific domain.
The paper tackled the challenge of applying curriculum learning to histopathology image classification by using annotator agreement as a difficulty proxy, resulting in a 4.5% AUC improvement from 83.7% to 88.2% on colorectal polyp classification.
Applying curriculum learning requires both a range of difficulty in data and a method for determining the difficulty of examples. In many tasks, however, satisfying these requirements can be a formidable challenge. In this paper, we contend that histopathology image classification is a compelling use case for curriculum learning. Based on the nature of histopathology images, a range of difficulty inherently exists among examples, and, since medical datasets are often labeled by multiple annotators, annotator agreement can be used as a natural proxy for the difficulty of a given example. Hence, we propose a simple curriculum learning method that trains on progressively-harder images as determined by annotator agreement. We evaluate our hypothesis on the challenging and clinically-important task of colorectal polyp classification. Whereas vanilla training achieves an AUC of 83.7% for this task, a model trained with our proposed curriculum learning approach achieves an AUC of 88.2%, an improvement of 4.5%. Our work aims to inspire researchers to think more creatively and rigorously when choosing contexts for applying curriculum learning.