Minimum Data, Maximum Impact: 20 annotated samples for explainable lung nodule classification

TL;DR

This paper demonstrates that using a small set of 20 annotated lung nodule images to train a generative model can produce synthetic data that significantly improves the performance of explainable classification models in medical imaging.

Contribution

The study introduces a method to synthesize attribute-annotated medical images using a diffusion model trained on only 20 samples, enhancing explainable lung nodule classification.

Findings

Attribute prediction accuracy increased by 13.4%.

Target prediction accuracy increased by 1.8%.

Synthetic data effectively compensates for limited real annotations.

Abstract

Classification models that provide human-interpretable explanations enhance clinicians' trust and usability in medical image diagnosis. One research focus is the integration and prediction of pathology-related visual attributes used by radiologists alongside the diagnosis, aligning AI decision-making with clinical reasoning. Radiologists use attributes like shape and texture as established diagnostic criteria and mirroring these in AI decision-making both enhances transparency and enables explicit validation of model outputs. However, the adoption of such models is limited by the scarcity of large-scale medical image datasets annotated with these attributes. To address this challenge, we propose synthesizing attribute-annotated data using a generative model. We enhance the Diffusion Model with attribute conditioning and train it using only 20 attribute-labeled lung nodule samples from the LIDC-IDRI dataset. Incorporating its generated images into the training of an explainable model boosts performance, increasing attribute prediction accuracy by 13.4% and target prediction accuracy by 1.8% compared to training with only the small real attribute-annotated dataset. This work highlights the potential of synthetic data to overcome dataset limitations, enhancing the applicability of explainable models in medical image analysis.