Improving Zero-Shot Detection of Low Prevalence Chest Pathologies using Domain Pre-trained Language Models
This work addresses the challenge of improving diagnostic accuracy for rare chest diseases in medical imaging, though it is incremental as it builds on existing CLIP-based methods.
The paper tackled the problem of detecting low-prevalence chest pathologies in zero-shot learning by using domain pre-trained language models like CXR-BERT, BlueBERT, and ClinicalBERT to replace BERT weights in CLIP-based models, resulting in improved performance on rare diseases despite a degradation on common pathologies.
Recent advances in zero-shot learning have enabled the use of paired image-text data to replace structured labels, replacing the need for expert annotated datasets. Models such as CLIP-based CheXzero utilize these advancements in the domain of chest X-ray interpretation. We hypothesize that domain pre-trained models such as CXR-BERT, BlueBERT, and ClinicalBERT offer the potential to improve the performance of CLIP-like models with specific domain knowledge by replacing BERT weights at the cost of breaking the original model's alignment. We evaluate the performance of zero-shot classification models with domain-specific pre-training for detecting low-prevalence pathologies. Even though replacing the weights of the original CLIP-BERT degrades model performance on commonly found pathologies, we show that pre-trained text towers perform exceptionally better on low-prevalence diseases. This motivates future ensemble models with a combination of differently trained language models for maximal performance.