Objectives <p>High-resolution vessel wall imaging (HR-VWI) is essential for diagnosing vulnerable intracranial atherosclerotic plaques, but its interpretation requires expertise. This study investigates the integration of large language models (LLMs) and deep learning (DL) for more efficient diagnosis.</p> Materials and methods <p>A retrospective study of symptomatic intracranial atherosclerotic stenosis patients (June 2018–June 2024) was conducted. LLMs (ChatGPT-4o, DeepSeek-V3, and Moonshot AI) were trained on HR-VWI reports to extract diagnostic insights. Additionally, DL models, ResNet50 and Vision Transformer (ViT), were used to classify vulnerable plaques. Diagnostic accuracy, sensitivity, specificity, and time efficiency were evaluated with both junior and senior doctors.</p> Results <p>A total of 1806 plaques from 726 patients were analyzed. ChatGPT-4o exhibited the highest diagnostic performance (AUC: 0.874). Among DL models, ViT outperformed ResNet50 (AUC: 0.913 vs. 0.845). LLMs and ViT significantly improved junior doctors’ diagnostic accuracy and reduced plaque assessment time (from 301 s to 174 s, <i>p</i> < 0.05).</p> Conclusion <p>The integration of LLMs and DL models enhanced diagnostic performance and efficiency, especially for junior doctors. This approach could reduce the burden on healthcare systems, particularly in resource-limited settings, by improving diagnostic accuracy and reducing the time required for plaque analysis.</p> Key Points <p><Emphasis Type="BoldItalic">Question</Emphasis> <i>How can AI models assist less experienced doctors in accurately identifying vulnerable intracranial plaques on high-resolution vessel wall imaging (HR-VWI) and improve diagnostic efficiency?</i></p> <p><Emphasis Type="BoldItalic">Findings</Emphasis> <i>The integration of large language models (LLMs) and deep learning (ViT) significantly improves diagnostic accuracy and efficiency, reducing assessment time for vulnerable plaques.</i></p> <p><Emphasis Type="BoldItalic">Clinical relevance</Emphasis> <i>This study demonstrates that combining LLMs and deep learning enables junior doctors to achieve near-expert accuracy in diagnosing vulnerable plaques, potentially reducing stroke risk and easing diagnostic burdens in resource-limited healthcare environments.</i></p> Graphical Abstract <p></p>

AI-driven diagnosis of vulnerable intracranial atherosclerotic plaques using large language models and vision transformers: a multi-center study

期刊论文
发表日期：

2025年10月22日

查看全文

Zi-Ang Li,
Yu Gao,
Kai Ji,
Kai-Yue Zhang,
Qiang Zhang,
Jie Wang,
Lin Han,
Xiao-Yang Zhai,
Wen-Peng Wang,
Wen-Ling Liu,
Han-Yu Wei,
Rui-Jing Qin,
Yong-Dong Li,
Hong-Ling Zhao,
Rui-Fang Yan,
Hong-Kai Cui

摘要

Objectives

High-resolution vessel wall imaging (HR-VWI) is essential for diagnosing vulnerable intracranial atherosclerotic plaques, but its interpretation requires expertise. This study investigates the integration of large language models (LLMs) and deep learning (DL) for more efficient diagnosis.

Materials and methods

A retrospective study of symptomatic intracranial atherosclerotic stenosis patients (June 2018–June 2024) was conducted. LLMs (ChatGPT-4o, DeepSeek-V3, and Moonshot AI) were trained on HR-VWI reports to extract diagnostic insights. Additionally, DL models, ResNet50 and Vision Transformer (ViT), were used to classify vulnerable plaques. Diagnostic accuracy, sensitivity, specificity, and time efficiency were evaluated with both junior and senior doctors.

Results

A total of 1806 plaques from 726 patients were analyzed. ChatGPT-4o exhibited the highest diagnostic performance (AUC: 0.874). Among DL models, ViT outperformed ResNet50 (AUC: 0.913 vs. 0.845). LLMs and ViT significantly improved junior doctors’ diagnostic accuracy and reduced plaque assessment time (from 301 s to 174 s, p < 0.05).

Conclusion

The integration of LLMs and DL models enhanced diagnostic performance and efficiency, especially for junior doctors. This approach could reduce the burden on healthcare systems, particularly in resource-limited settings, by improving diagnostic accuracy and reducing the time required for plaque analysis.

Key Points

Question How can AI models assist less experienced doctors in accurately identifying vulnerable intracranial plaques on high-resolution vessel wall imaging (HR-VWI) and improve diagnostic efficiency?

Findings The integration of large language models (LLMs) and deep learning (ViT) significantly improves diagnostic accuracy and efficiency, reducing assessment time for vulnerable plaques.

Clinical relevance This study demonstrates that combining LLMs and deep learning enables junior doctors to achieve near-expert accuracy in diagnosing vulnerable plaques, potentially reducing stroke risk and easing diagnostic burdens in resource-limited healthcare environments.

Graphical Abstract