人工智能辅助[18F]FDG PET/CT 定量：大 B 细胞淋巴瘤中一种节省时间的方法

IF 0.2 4区医学 Q4 PATHOLOGY

Medecine Nucleaire-Imagerie Fonctionnelle et Metabolique Pub Date : 2025-03-01 DOI:10.1016/j.mednuc.2025.01.142

V. Micheletti , E. Marchal , M. Gambiez , L. Dercle , M.E. Meyer , X. Palard Novello , A. Girard

{"title":"人工智能辅助[18F]FDG PET/CT 定量：大 B 细胞淋巴瘤中一种节省时间的方法","authors":"V. Micheletti , E. Marchal , M. Gambiez , L. Dercle , M.E. Meyer , X. Palard Novello , A. Girard","doi":"10.1016/j.mednuc.2025.01.142","DOIUrl":null,"url":null,"abstract":"<div><h3>Purpose</h3><div>To assess if AI-assisted segmentation on [18F]FDG PET/CT significantly reduces time in quantifying Total Metabolic Tumor Volume (TMTV) and the greatest distance between two lesions (Dmax) in large B-cell lymphoma (LBCL) patients.</div></div><div><h3>Methods</h3><div>A retrospective study at two institutions included 56 LBCL patients (Ann-Arbor staging: 14% with no disease, 25% stage I-II, 61% stage III-IV). Two methods were compared: AI-assisted segmentation (Pionus® v1.1.0, PAIRE, Paris) automatically identified lesions with expert correction of AI errors, while expert-led segmentation was fully manual. Both methods used a 41% SUVmax threshold.</div></div><div><h3>Results</h3><div>AI-assisted segmentation was 41% faster, saving an average of 104seconds per patient (152±134seconds vs. 255±270seconds, <em>P</em>  =0.67, <em>P</em>         <0.001), with time saved correlating with TMTV (<em>r</em>  <0.001). No significant differences in TMTV or Dmax were found between methods. Intraclass correlation coefficients were high for both TMTV (0.91 and 0.95) and Dmax (0.97 and 0.95), with strong inter-reader agreement.</div></div><div><h3>Conclusion</h3><div>AI-assisted segmentation significantly reduces time for TMTV and Dmax measurement in LBCL patients, while maintaining high reliability and inter-reader agreement compared to manual methods.</div></div>\",\"PeriodicalId\":49841,\"journal\":{\"name\":\"Medecine Nucleaire-Imagerie Fonctionnelle et Metabolique\",\"volume\":\"49 2\",\"pages\":\"Page 113\"},\"PeriodicalIF\":0.2000,\"publicationDate\":\"2025-03-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Medecine Nucleaire-Imagerie Fonctionnelle et Metabolique\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0928125825001421\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"PATHOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Medecine Nucleaire-Imagerie Fonctionnelle et Metabolique","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0928125825001421","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"PATHOLOGY","Score":null,"Total":0}

引用次数: 0

摘要

目的评估ai辅助的[18F]FDG PET/CT分割是否显著缩短了大b细胞淋巴瘤（LBCL）患者总代谢肿瘤体积（TMTV）和两个病变之间最大距离（Dmax）的量化时间。方法在两家机构进行回顾性研究，纳入56例LBCL患者（Ann-Arbor分期：14%无疾病，25% I-II期，61% III-IV期）。比较了两种方法：人工智能辅助分割（Pionus®v1.1.0， PAIRE, Paris）通过专家纠正人工智能错误自动识别病变，而专家主导的分割则完全是手动的。两种方法都使用41%的SUVmax阈值。结果ai辅助分割速度快41%，平均每位患者节省104秒（152±134秒vs 255±270秒）；0.001)，节省的时间与TMTV相关(r = 0.67, P <；0.001)。两种方法的TMTV和Dmax均无显著差异。TMTV（0.91和0.95）和Dmax（0.97和0.95）的类内相关系数较高，具有较强的读者间一致性。结论人工智能辅助分割显著减少了LBCL患者TMTV和Dmax测量的时间，同时与人工方法相比保持了较高的可靠性和读者间一致性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

AI-assisted [18F]FDG PET/CT quantification: A time-saving approach in large B-cell lymphoma

Purpose

To assess if AI-assisted segmentation on [18F]FDG PET/CT significantly reduces time in quantifying Total Metabolic Tumor Volume (TMTV) and the greatest distance between two lesions (Dmax) in large B-cell lymphoma (LBCL) patients.

Methods

A retrospective study at two institutions included 56 LBCL patients (Ann-Arbor staging: 14% with no disease, 25% stage I-II, 61% stage III-IV). Two methods were compared: AI-assisted segmentation (Pionus® v1.1.0, PAIRE, Paris) automatically identified lesions with expert correction of AI errors, while expert-led segmentation was fully manual. Both methods used a 41% SUVmax threshold.

Results

AI-assisted segmentation was 41% faster, saving an average of 104 seconds per patient (152 ± 134 seconds vs. 255 ± 270 seconds, P < 0.001), with time saved correlating with TMTV (r = 0.67, P < 0.001). No significant differences in TMTV or Dmax were found between methods. Intraclass correlation coefficients were high for both TMTV (0.91 and 0.95) and Dmax (0.97 and 0.95), with strong inter-reader agreement.

Conclusion

AI-assisted segmentation significantly reduces time for TMTV and Dmax measurement in LBCL patients, while maintaining high reliability and inter-reader agreement compared to manual methods.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Medecine Nucleaire-Imagerie Fonctionnelle et Metabolique PATHOLOGY-

CiteScore

0.30

自引率

0.00%

发文量

160

审稿时长

19.8 weeks

期刊介绍： Le but de Médecine nucléaire - Imagerie fonctionnelle et métabolique est de fournir une plate-forme d''échange d''informations cliniques et scientifiques pour la communauté francophone de médecine nucléaire, et de constituer une expérience pédagogique de la rédaction médicale en conformité avec les normes internationales.