Chieh-Ju Chao, Yunqi Richard Gu, Wasan Kumar, Tiange Xiang, Lalith Appari, Justin Wu, Juan M. Farina, Rachael Wraith, Jiwoon Jeong, Reza Arsanjani, Garvan C. Kane, Jae K. Oh, Curtis P. Langlotz, Imon Banerjee, Li Fei-Fei, Ehsan Adeli
{"title":"心脏超声左心室分割的基础与领域特异性模型","authors":"Chieh-Ju Chao, Yunqi Richard Gu, Wasan Kumar, Tiange Xiang, Lalith Appari, Justin Wu, Juan M. Farina, Rachael Wraith, Jiwoon Jeong, Reza Arsanjani, Garvan C. Kane, Jae K. Oh, Curtis P. Langlotz, Imon Banerjee, Li Fei-Fei, Ehsan Adeli","doi":"10.1038/s41746-025-01730-y","DOIUrl":null,"url":null,"abstract":"<p>The Segment Anything Model (SAM) was fine-tuned on the EchoNet-Dynamic dataset and evaluated on external transthoracic echocardiography (TTE) and Point-of-Care Ultrasound (POCUS) datasets from CAMUS (University Hospital of St Etienne) and Mayo Clinic (99 patients: 58 TTE, 41 POCUS). Fine-tuned SAM was superior or comparable to MedSAM. The fine-tuned SAM also outperformed EchoNet and U-Net models, demonstrating strong generalization, especially on apical 2-chamber (A2C) images (fine-tuned SAM vs. EchoNet: CAMUS-A2C: DSC 0.891 ± 0.040 vs. 0.752 ± 0.196, p < 0.0001) and POCUS (DSC 0.857 ± 0.047 vs. 0.667 ± 0.279, p < 0.0001). Additionally, SAM-enhanced workflow reduced annotation time by 50% (11.6 ± 4.5 sec vs. 5.7 ± 1.7 sec, p < 0.0001) while maintaining segmentation quality. We demonstrated an effective strategy for fine-tuning a vision foundation model for enhancing clinical workflow efficiency and supporting human-AI collaboration.</p>","PeriodicalId":19349,"journal":{"name":"NPJ Digital Medicine","volume":"14 1","pages":""},"PeriodicalIF":12.4000,"publicationDate":"2025-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Foundation versus domain-specific models for left ventricular segmentation on cardiac ultrasound\",\"authors\":\"Chieh-Ju Chao, Yunqi Richard Gu, Wasan Kumar, Tiange Xiang, Lalith Appari, Justin Wu, Juan M. Farina, Rachael Wraith, Jiwoon Jeong, Reza Arsanjani, Garvan C. Kane, Jae K. Oh, Curtis P. Langlotz, Imon Banerjee, Li Fei-Fei, Ehsan Adeli\",\"doi\":\"10.1038/s41746-025-01730-y\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>The Segment Anything Model (SAM) was fine-tuned on the EchoNet-Dynamic dataset and evaluated on external transthoracic echocardiography (TTE) and Point-of-Care Ultrasound (POCUS) datasets from CAMUS (University Hospital of St Etienne) and Mayo Clinic (99 patients: 58 TTE, 41 POCUS). Fine-tuned SAM was superior or comparable to MedSAM. The fine-tuned SAM also outperformed EchoNet and U-Net models, demonstrating strong generalization, especially on apical 2-chamber (A2C) images (fine-tuned SAM vs. EchoNet: CAMUS-A2C: DSC 0.891 ± 0.040 vs. 0.752 ± 0.196, p < 0.0001) and POCUS (DSC 0.857 ± 0.047 vs. 0.667 ± 0.279, p < 0.0001). Additionally, SAM-enhanced workflow reduced annotation time by 50% (11.6 ± 4.5 sec vs. 5.7 ± 1.7 sec, p < 0.0001) while maintaining segmentation quality. We demonstrated an effective strategy for fine-tuning a vision foundation model for enhancing clinical workflow efficiency and supporting human-AI collaboration.</p>\",\"PeriodicalId\":19349,\"journal\":{\"name\":\"NPJ Digital Medicine\",\"volume\":\"14 1\",\"pages\":\"\"},\"PeriodicalIF\":12.4000,\"publicationDate\":\"2025-06-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"NPJ Digital Medicine\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1038/s41746-025-01730-y\",\"RegionNum\":1,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"HEALTH CARE SCIENCES & SERVICES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"NPJ Digital Medicine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1038/s41746-025-01730-y","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"HEALTH CARE SCIENCES & SERVICES","Score":null,"Total":0}
引用次数: 0
摘要
片段任何模型(SAM)在EchoNet-Dynamic数据集上进行微调,并在CAMUS(圣埃蒂安大学医院)和Mayo诊所的外部经胸超声心动图(TTE)和护理点超声(POCUS)数据集上进行评估(99例患者:58例TTE, 41例POCUS)。微调SAM优于MedSAM或与MedSAM相当。微调后的SAM也优于EchoNet和U-Net模型,表现出很强的泛化能力,特别是在根尖2室(A2C)图像上(微调后的SAM vs EchoNet: CAMUS-A2C: DSC 0.891±0.040 vs 0.752±0.196,p < 0.0001)和POCUS (DSC 0.857±0.047 vs 0.667±0.279,p < 0.0001)。此外,sam增强的工作流在保持分割质量的同时减少了50%的注释时间(11.6±4.5秒vs. 5.7±1.7秒,p < 0.0001)。我们展示了一种有效的策略来微调视觉基础模型,以提高临床工作流程效率并支持人类与人工智能的协作。
Foundation versus domain-specific models for left ventricular segmentation on cardiac ultrasound
The Segment Anything Model (SAM) was fine-tuned on the EchoNet-Dynamic dataset and evaluated on external transthoracic echocardiography (TTE) and Point-of-Care Ultrasound (POCUS) datasets from CAMUS (University Hospital of St Etienne) and Mayo Clinic (99 patients: 58 TTE, 41 POCUS). Fine-tuned SAM was superior or comparable to MedSAM. The fine-tuned SAM also outperformed EchoNet and U-Net models, demonstrating strong generalization, especially on apical 2-chamber (A2C) images (fine-tuned SAM vs. EchoNet: CAMUS-A2C: DSC 0.891 ± 0.040 vs. 0.752 ± 0.196, p < 0.0001) and POCUS (DSC 0.857 ± 0.047 vs. 0.667 ± 0.279, p < 0.0001). Additionally, SAM-enhanced workflow reduced annotation time by 50% (11.6 ± 4.5 sec vs. 5.7 ± 1.7 sec, p < 0.0001) while maintaining segmentation quality. We demonstrated an effective strategy for fine-tuning a vision foundation model for enhancing clinical workflow efficiency and supporting human-AI collaboration.
期刊介绍:
npj Digital Medicine is an online open-access journal that focuses on publishing peer-reviewed research in the field of digital medicine. The journal covers various aspects of digital medicine, including the application and implementation of digital and mobile technologies in clinical settings, virtual healthcare, and the use of artificial intelligence and informatics.
The primary goal of the journal is to support innovation and the advancement of healthcare through the integration of new digital and mobile technologies. When determining if a manuscript is suitable for publication, the journal considers four important criteria: novelty, clinical relevance, scientific rigor, and digital innovation.