Transition-Aware best-of-N sampling for Longitudinal Chest X-ray Reports
For radiologists and clinicians, this work improves automated chest X-ray report generation by incorporating longitudinal context, but the gains are incremental as it builds on existing best-of-N sampling.
The paper introduces a training-free best-of-N sampling method for longitudinal chest X-ray report generation that explicitly models the transition between prior and current exams. The method outperforms random selection across multiple generators and prompts, with the largest gains on the Impression section.
In longitudinal clinical practice, every chest X-ray is read in the context of the patients prior exam, and much of what the radiologist communicates is the change from one visit to the next. To the best of our knowledge, we present the first training-free best-of-N sampling scheme for pre-trained chest X-ray report generators that is explicitly aware of this longitudinal prior to current transition. We call it transition-aware best-of-N sampling, each report is split into sentences and embedded into an unordered set in Rd; each (prior, current) pair is reduced to a fixed-dim directional vector via a set-to-set distance designed to encode the change between the two sets; and candidates are scored by cosine distance from their candidate transition vector to a cached bank of ground-truth training transition vectors, aggregated as min or kNN. We instantiate the framework with four directional set distances (mean-shift, novelty residual, directed-Hausdorff anchor, and cost-weighted optimal transport) and evaluate on a multi-visit AP-PA cohort, running inference under three prompts on three vision-language generators. Transition-aware best-of-N outperforms random selection across the board, with the largest relative gains on the Impression section.