视觉语言模型评估乳腺癌可疑腋窝淋巴结的超声表现
回顾性纳入718例原发性浸润性乳腺癌患者,以超声引导下细针/粗针穿刺为参考标准,评估灰度及多普勒图像。比较三种视觉语言模型与不同年资放射科医生的诊断效能,结果显示高年资医生准确性最高,GPT-5.2与低年资医生准确性无显著差异但特异度更高。
要点
- GPT-5.2准确性78%,特异度80%,高于低年资医生的64%,可用于辅助降低假阳性
- 低年资医生敏感度92%但特异度仅64%,易将良性淋巴结误判为转移
- 高年资医生准确性83%、特异度89%,仍优于所有视觉语言模型
核心结论
- GPT-5.2是准确性最高的视觉语言模型,其准确性与低年资医生无显著差异,但敏感度较低且特异度较高
- GPT-5.2准确性低于高年资放射科医生,尚不能替代高年资医生的判读
研究设计与样本
- 回顾性研究纳入718例原发性浸润性乳腺癌患者,平均年龄53.1±11.4岁
- 所有患者在2025年1月至6月间接受超声引导下可疑腋窝淋巴结细针或粗针穿刺活检
- 每例患者留存单枚活检淋巴结的灰度图像及可获取的多普勒图像
- 两名低年资医生与两名高年资医生分别阅片并各组达成共识,同时将图像输入三种视觉语言模型
- 以细针/粗针穿刺病理结果为参考标准,采用Bonferroni校正进行组间比较
关键影像表现/征象
- 输入图像为可疑腋窝淋巴结的灰度图像及可获取的多普勒图像,模型与医生仅凭图像进行二分类判断
诊断标准与关键数值
- 低年资医生准确性76%,敏感度92%,特异度64%
- 高年资医生准确性83%,敏感度75%,特异度89%
- GPT-5.2准确性78%,敏感度76%,特异度80%
- Gemini-3-Pro准确性73%,敏感度79%,特异度68%
- Claude-Opus-4.5准确性67%,敏感度63%,特异度70%
临床应用与阅片要点
- 高年资医生准确性显著高于低年资医生及三种模型,低年资医生准确性显著高于Claude-Opus-4.5
- 低年资医生敏感度显著高于高年资医生及三种模型,高年资医生敏感度显著高于Claude-Opus-4.5
- 高年资医生特异度显著高于低年资医生及三种模型,GPT-5.2特异度显著高于低年资医生
- 视觉语言模型可探索作为低年资医生的监督辅助决策工具,帮助减少假阳性判读
局限性
- 原文未报告局限性信息
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
Background: Axillary management in patients with breast cancer is becoming increasingly individualized, requiring accurate nodal characterization. Objective: To evaluate the performance of vision-language models (VLMs) with image-only diagnostic input for detecting malignancy in biopsied axillary lymph nodes in patients with breast cancer and to compare performance with radiologists of different experience levels. Methods: This retrospective study included 718 patients (mean age, 53.1±11.4 years) with primary invasive breast cancer who underwent ultrasound-guided fine-needle aspiration or core-needle biopsy (FNA/CNB) of a suspicious target axillary lymph node from January 2025 through June 2025. A grayscale image and corresponding Doppler image, if available, were saved for a single biopsied node in each patient. Two inexperienced radiologists (both with only residency exposure to breast ultrasound) and two experienced radiologists (both breast imaging radiologists) reviewed images to classify nodes as metastatic or nonmetastatic; each pair reached consensus. The images were also inputted to three VLMs (GPT-5.2, Gemini-3-Pro, and Claude-Opus-4.5), with a text prompt requesting binary outputs (metastatic or nonmetastatic). Performance was compared between the reader groups and between the models and reader groups using Bonferroni adjustment and FNA/CNB as reference. Results: Accuracy, sensitivity, and specificity for inexperienced radiologists were 76%, 92%, and 64%; for experienced radiologists were 83%, 75%, and 89%; for GPT-5.2 were 78%, 76%, and 80%; for Gemini-3-Pro were 73%, 79%, and 68%; and for Claude-Opus-4.5 were 67%, 63%, and 70%, respectively. Accuracy was significantly higher for experienced radiologists compared with inexperienced radiologists and all three models; and for inexperienced radiologists compared with Claude-Opus-4.5. Sensitivity was significantly higher for inexperienced radiologists compared with experienced radiologists and all three models and for experienced radiologists compared with Claude-Opus-4.5. Specificity was significantly higher for experienced radiologists compared with inexperienced radiologists and all three models and for GPT-5.2 compared with inexperienced radiologists. Other comparisons were not significant. Conclusion: GPT-5.2 (the VLM with highest accuracy) showed no significant difference in accuracy versus inexperienced radiologists, albeit had lower sensitivity and greater specificity. GPT-5.2 had lower accuracy than experienced radiologists. Clinical Impact: The results support investigation of VLMs as a supervised decision-support tool to help reduce false-positive assessments by inexperienced radiologists.
引用
- He, Chen, Dong, et al. Vision Language Models for Ultrasound Assessment of Suspicious Axillary Lymph Nodes in Breast Cancer. AJR Am J Roentgenol. 2026. PMID: 42485405. doi:10.2214/AJR.26.35227
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。