A Textbook Remedy for Domain Shifts: Knowledge Priors for Medical Image Analysis

TL;DR

This paper introduces KnoBo, a knowledge-enhanced model using medical knowledge priors from natural language sources to improve deep network generalization across domain shifts in medical imaging.

Contribution

It proposes KnoBo, a novel concept bottleneck model that incorporates medical knowledge priors via retrieval-augmented language models for better domain robustness.

Findings

KnoBo outperforms fine-tuned models by 32.4% on confounded datasets.

PubMed knowledge source improves model robustness and diversity.

Knowledge priors reduce sensitivity to domain shifts in medical imaging.

Abstract

While deep networks have achieved broad success in analyzing natural images, when applied to medical scans, they often fail in unexcepted situations. We investigate this challenge and focus on model sensitivity to domain shifts, such as data sampled from different hospitals or data confounded by demographic variables such as sex, race, etc, in the context of chest X-rays and skin lesion images. A key finding we show empirically is that existing visual backbones lack an appropriate prior from the architecture for reliable generalization in these settings. Taking inspiration from medical training, we propose giving deep networks a prior grounded in explicit medical knowledge communicated in natural language. To this end, we introduce Knowledge-enhanced Bottlenecks (KnoBo), a class of concept bottleneck models that incorporates knowledge priors that constrain it to reason with clinically relevant factors found in medical textbooks or PubMed. KnoBo uses retrieval-augmented language models to design an appropriate concept space paired with an automatic training procedure for recognizing the concept. We evaluate different resources of knowledge and recognition architectures on a broad range of domain shifts across 20 datasets. In our comprehensive evaluation with two imaging modalities, KnoBo outperforms fine-tuned models on confounded datasets by 32.4% on average. Finally, evaluations reveal that PubMed is a promising resource for making medical models less sensitive to domain shift, outperforming other resources on both diversity of information and final prediction performance.