A DeepSeek-Powered AI System for Automated Chest Radiograph Interpretation in Clinical Practice

2025-12-23

AI Summary

Global radiologist shortages and heavy burdens on frontline chest X-ray (CXR) interpretation persist, while existing large language models lack prospective clinical validation. To address this, we propose Janus-Pro-CXR—a lightweight, domain-optimized CXR interpretation system built upon the DeepSeek Janus-Pro (1B) architecture, incorporating domain-specific fine-tuning, model compression, and seamless integration into clinical workflows. We conducted the first multicenter prospective clinical trial (NCT07117266), demonstrating fully automated report generation and detection of six critical radiographic findings—despite having far fewer parameters than models like ChatGPT-4o. Results show a statistically significant 18.3% reduction in interpretation time (*P* < 0.001), improved report quality, and a 54.3% expert preference rate over conventional reporting. The open-source architecture enhances transparency and facilitates trustworthy AI deployment in resource-constrained primary care settings.

Technology Category

• Computer Vision: Large Vision Models
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

A global shortage of radiologists has been exacerbated by the significant volume of chest X-ray workloads, particularly in primary care. Although multimodal large language models show promise, existing evaluations predominantly rely on automated metrics or retrospective analyses, lacking rigorous prospective clinical validation. Janus-Pro-CXR (1B), a chest X-ray interpretation system based on DeepSeek Janus-Pro model, was developed and rigorously validated through a multicenter prospective trial (NCT07117266). Our system outperforms state-of-the-art X-ray report generation models in automated report generation, surpassing even larger-scale models including ChatGPT 4o (200B parameters), while demonstrating reliable detection of six clinically critical radiographic findings. Retrospective evaluation confirms significantly higher report accuracy than Janus-Pro and ChatGPT 4o. In prospective clinical deployment, AI assistance significantly improved report quality scores, reduced interpretation time by 18.3% (P < 0.001), and was preferred by a majority of experts in 54.3% of cases. Through lightweight architecture and domain-specific optimization, Janus-Pro-CXR improves diagnostic reliability and workflow efficiency, particularly in resource-constrained settings. The model architecture and implementation framework will be open-sourced to facilitate the clinical translation of AI-assisted radiology solutions.

Problem

• Addresses radiologist shortage by automating chest X-ray interpretation
• Validates AI system for clinical use through prospective multicenter trial
• Improves report accuracy and reduces interpretation time in practice

Innovation

• Developed Janus-Pro-CXR system using DeepSeek Janus-Pro model
• Conducted multicenter prospective trial for rigorous clinical validation
• Open-sourced model architecture to facilitate AI clinical translation