Hsu Wen-Chi, Wang Yuli, Wu Yu-Fu, Chen Ruohua, Afyouni Shadi, Liu Jhehong, Vin Somasundaram, Shi Victoria, Imami Maliha, Chotiyanonta Jill S, Zandieh Ghazal, Cai Yeyu, Leal Jeffrey P, Oishi Kenichi, Zaheer Atif, Ward Robert C, Zhang Paul J L, Wu Jing, Jiao Zhicheng, Kamel Ihab R, Lin Gigin, Bai Harrison X
Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, Md.
Department of Medical Imaging and Intervention, Chang Gung Memorial Hospital at Linkou, No. 5 Fuxing St, Guishan Dist, Taoyuan 33305, Taiwan.
Radiology. 2025 Aug;316(2):e243412. doi: 10.1148/radiol.243412.
Background Artificial intelligence may enhance diagnostic accuracy in classifying ovarian lesions on MRI scans; however, its applicability across diverse datasets is uncertain. Purpose To develop an efficient, generalizable pipeline for MRI-based ovarian lesion characterization. Materials and Methods In this retrospective study, multiparametric MRI datasets of patients with ovarian lesions from a primary institution (January 2008 to January 2019) and two external institutions (January 2010 to October 2020) were analyzed. Lesions were automatically segmented using Meta's Segment Anything Model (SAM). A DenseNet-121 deep learning (DL) model incorporating both imaging and clinical data was then trained and validated externally for ovarian lesion classification. Lesions were evaluated by radiologists using the Ovarian-Adnexal Reporting and Data System for MRI and subjective assessment, classifying them as benign or malignant. The classification performances of the DL model and radiologists were compared using the DeLong test. Results The primary dataset included 534 lesions from 448 women (mean age, 52 years ± 15 [SD]) from institution A (United States), whereas the external datasets included 58 lesions from 55 women (mean age, 51 years ± 19) from institution B (United States) and 29 lesions from 29 women (mean age, 49 years ± 10) from institution C (Taiwan). SAM-assisted segmentation had a Dice coefficient of 0.86-0.88, reducing the processing time per lesion by 4 minutes compared with manual segmentation. The DL classification model achieved an area under the receiver operating characteristic curve (AUC) of 0.85 (95% CI: 0.85, 0.85) on the internal test and 0.79 (95% CI: 0.79, 0.79 and 0.78, 0.79) across both external datasets with SAM-segmented images, comparable with the radiologists' performance (AUC: 0.84-0.93; all > .05). Conclusion These results describe an accurate, efficient pipeline that integrates SAM with DL-based classification for differentiating malignant from benign ovarian lesions on MRI scans. It reduced segmentation time and achieved classification performance comparable with that of radiologists. © RSNA, 2025 See also the editorial by Bhayana and Wang in this issue.
背景 人工智能可能会提高磁共振成像（MRI）扫描中卵巢病变分类的诊断准确性；然而，其在不同数据集上的适用性尚不确定。目的 开发一种基于MRI的卵巢病变特征描述的高效、通用流程。材料与方法 在这项回顾性研究中，分析了来自一家主要机构（2008年1月至2019年1月）以及两家外部机构（2010年1月至2020年10月）的卵巢病变患者的多参数MRI数据集。使用Meta的分割一切模型（SAM）对病变进行自动分割。然后训练并在外部验证一个结合了影像和临床数据的DenseNet-121深度学习（DL）模型用于卵巢病变分类。放射科医生使用卵巢附件MRI报告和数据系统以及主观评估对病变进行评估，将其分类为良性或恶性。使用德龙检验比较DL模型和放射科医生的分类性能。结果 主要数据集包括来自机构A（美国）的448名女性（平均年龄52岁±15[标准差]）的534个病变，而外部数据集包括来自机构B（美国）的55名女性（平均年龄51岁±19）的58个病变以及来自机构C（台湾）的29名女性（平均年龄49岁±10）的29个病变。SAM辅助分割的骰子系数为0.86 - 0.88，与手动分割相比，每个病变的处理时间减少了4分钟。DL分类模型在内部测试中的受试者操作特征曲线下面积（AUC）为0.85（95%置信区间：0.85, 0.85），在两个外部数据集中使用SAM分割图像时的AUC为0.79（95%置信区间：0.79, 0.79和0.78, 0.79），与放射科医生的表现相当（AUC：0.84 - 0.93；均P >.05）。结论 这些结果描述了一种准确、高效的流程，该流程将SAM与基于DL的分类相结合，用于在MRI扫描中区分卵巢良恶性病变。它减少了分割时间，并实现了与放射科医生相当的分类性能。©RSNA，2025 另见本期Bhayana和Wang的社论。
