Seeing Through Multiple Views: Parameter-Efficient Fine-Tuning via Selective Neurons for Consistent Radiology Report Generation

2026-06-29

AI Summary

This work addresses the clinical inconsistency in radiology report generation (RRG) arising from viewpoint discrepancies in multi-view X-ray images. To tackle this challenge, the authors propose View-PNDF, a novel framework that, for the first time, leverages neuron selectivity to identify and enhance neurons specifically sensitive to individual imaging views while preserving view-invariant representations. By integrating these refined features with large language models such as GPT-4o, the method generates comprehensive and clinically coherent reports. View-PNDF employs a parameter-efficient strategy through view-specific neuron detection and selective fine-tuning, enabling effective multi-view fusion. Evaluated on two established medical RRG benchmarks, the approach significantly improves both view-specific reporting quality and overall performance, with further validation from LLM-based assessments confirming its clinical reliability.

Abstract

Recent years have seen substantial advances in radiology report generation (RRG), yet existing approaches predominantly adopt direct feature fusion when handling multi-view X-ray images. Such approaches overlook the potential clinical inconsistencies and inaccuracies arising when a single model processes different views, adversely impacting performance and clinical reliability. To this end, we introduce View-PNDF (View-specific Pattern Neuron Detection and Fine-tuning), a parameter-efficient framework that fosters view-consistent report generation from a neuronal perspective. Specifically, View-PNDF comprises: (i) a view-specific neuron detection module identifying neurons responsive to particular views, (ii) a verification module quantifying the existence of these neurons, and (iii) a selective fine-tuning strategy strengthening detected neurons while preserving view-agnostic representations. By updating only view-specific neurons, View-PNDF achieves consistent diagnoses across different views with reduced computational costs. Subsequently, we employ Large Language Models (LLMs) to consolidate the view-specific reports into a complete radiology report. Furthermore, we use traditional Natural Language Generation (NLG) metrics-based assessment on integrated reports for baseline comparison and employ LLM-based assessment (e.g., GPT-4o) on view-specific reports to capture clinical significance. Extensive experiments on two medical RRG benchmarks demonstrate that View-PNDF substantially improves view-specific chest X-ray report generation quality while maintaining robust general-view performance.

Problem

• radiology report generation
• multi-view X-ray
• clinical inconsistency
• view-specific processing
• medical imaging

Innovation

• Parameter-Efficient Fine-Tuning
• View-Specific Neurons
• Radiology Report Generation
• Multi-View X-ray
• Large Language Models