新型人工智能辅助子宫肌瘤分割算法的评价

IF 3.3 2区医学 Q1 OBSTETRICS & GYNECOLOGY

Journal of minimally invasive gynecology Pub Date : 2025-10-22 DOI:10.1016/j.jmig.2025.09.007

S Naval , DM Anagani , DU BR , S Kothamachu

{"title":"新型人工智能辅助子宫肌瘤分割算法的评价","authors":"S Naval , DM Anagani , DU BR , S Kothamachu","doi":"10.1016/j.jmig.2025.09.007","DOIUrl":null,"url":null,"abstract":"<div><h3>Study Objective</h3><div>To evaluate uterine and fibroid segmentation performed by an AI-assisted algorithm (Nesa Medtech, Bengaluru - 560085, India) by three experienced clinicians.</div></div><div><h3>Design</h3><div>This study was a prospective validation study. A structured questionnaire was designed to validate the parameters of the uterus and fibroid and to capture the satisfaction rate of AI segmentation among the clinicians.</div></div><div><h3>Setting</h3><div>Uterine fibroid scanned data from different ultrasonography machines (irrespective of make) with less than a maximum fibroid size not exceeding 5 cm with a maximum of 4 fibroids, were segmented by an AI-assisted algorithm.</div></div><div><h3>Patients or Participants</h3><div>A total of 100 patients with uterine fibroids were included in the study.</div></div><div><h3>Interventions</h3><div>NA</div></div><div><h3>Measurements and Primary Results</h3><div>The acquired imaging data were segmented using the AI-assisted algorithm. The uterus and fibroid segmentation of 100 cases was successfully done by this algorithm. The precise size of the uterus and accurate mapping (size, location, and FIGO-type) of the fibroid were appreciated when these segmentation results were validated by 3 experienced clinicians with expertise in ultrasonography. Clinician -1 - was satisfied in 98% of cases, Clinician -2 - was satisfied in 97% of cases, and Clinician -3 - was satisfied in 99% of cases.</div></div><div><h3>Conclusion</h3><div>The AI-assisted algorithm demonstrated strong agreement with expert analysis of the segmentation of the uterus and fibroid by ultrasonography. It appears this novel AI algorithm is promising in fibroid segmentation and its accuracy should be analyzed in future studies. In the future, this AI algorithm could be used for surgical planning.</div></div>","PeriodicalId":16397,"journal":{"name":"Journal of minimally invasive gynecology","volume":"32 11","pages":"Pages S2-S3"},"PeriodicalIF":3.3000,"publicationDate":"2025-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Evaluation of Novel AI Assisted Algorithm for Segmentation of Uterine Fibroids\",\"authors\":\"S Naval , DM Anagani , DU BR , S Kothamachu\",\"doi\":\"10.1016/j.jmig.2025.09.007\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Study Objective</h3><div>To evaluate uterine and fibroid segmentation performed by an AI-assisted algorithm (Nesa Medtech, Bengaluru - 560085, India) by three experienced clinicians.</div></div><div><h3>Design</h3><div>This study was a prospective validation study. A structured questionnaire was designed to validate the parameters of the uterus and fibroid and to capture the satisfaction rate of AI segmentation among the clinicians.</div></div><div><h3>Setting</h3><div>Uterine fibroid scanned data from different ultrasonography machines (irrespective of make) with less than a maximum fibroid size not exceeding 5 cm with a maximum of 4 fibroids, were segmented by an AI-assisted algorithm.</div></div><div><h3>Patients or Participants</h3><div>A total of 100 patients with uterine fibroids were included in the study.</div></div><div><h3>Interventions</h3><div>NA</div></div><div><h3>Measurements and Primary Results</h3><div>The acquired imaging data were segmented using the AI-assisted algorithm. The uterus and fibroid segmentation of 100 cases was successfully done by this algorithm. The precise size of the uterus and accurate mapping (size, location, and FIGO-type) of the fibroid were appreciated when these segmentation results were validated by 3 experienced clinicians with expertise in ultrasonography. Clinician -1 - was satisfied in 98% of cases, Clinician -2 - was satisfied in 97% of cases, and Clinician -3 - was satisfied in 99% of cases.</div></div><div><h3>Conclusion</h3><div>The AI-assisted algorithm demonstrated strong agreement with expert analysis of the segmentation of the uterus and fibroid by ultrasonography. It appears this novel AI algorithm is promising in fibroid segmentation and its accuracy should be analyzed in future studies. In the future, this AI algorithm could be used for surgical planning.</div></div>\",\"PeriodicalId\":16397,\"journal\":{\"name\":\"Journal of minimally invasive gynecology\",\"volume\":\"32 11\",\"pages\":\"Pages S2-S3\"},\"PeriodicalIF\":3.3000,\"publicationDate\":\"2025-10-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of minimally invasive gynecology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1553465025003449\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"OBSTETRICS & GYNECOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of minimally invasive gynecology","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1553465025003449","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"OBSTETRICS & GYNECOLOGY","Score":null,"Total":0}

引用次数: 0

摘要

研究目的评估由三名经验丰富的临床医生使用人工智能辅助算法（Nesa Medtech, Bengaluru - 560085, India）进行的子宫和肌瘤分割。本研究为前瞻性验证研究。设计一份结构化问卷来验证子宫和肌瘤的参数，并了解临床医生对人工智能分割的满意度。使用人工智能辅助算法对最大肌瘤大小小于5厘米，最大肌瘤大小不超过4个的不同超声机（不论型号）的子宫肌瘤扫描数据进行分割。患者或参与者共纳入100例子宫肌瘤患者。干预测量和初步结果采用人工智能辅助算法对获取的成像数据进行分割。应用该算法成功分割了100例子宫和肌瘤。子宫的精确大小和子宫肌瘤的精确定位（大小、位置和figo型）得到了3名经验丰富的超声专业医生的验证。临床医生1 -的满意率为98%，临床医生2 -的满意率为97%，临床医生3 -的满意率为99%。结论人工智能辅助算法与超声对子宫和肌瘤分割的专家分析结果吻合较好。这种新的人工智能算法在肌瘤分割中具有广阔的应用前景，其准确性有待于进一步的研究。在未来，这种人工智能算法可以用于手术计划。

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

Evaluation of Novel AI Assisted Algorithm for Segmentation of Uterine Fibroids

Study Objective

To evaluate uterine and fibroid segmentation performed by an AI-assisted algorithm (Nesa Medtech, Bengaluru - 560085, India) by three experienced clinicians.

Design

This study was a prospective validation study. A structured questionnaire was designed to validate the parameters of the uterus and fibroid and to capture the satisfaction rate of AI segmentation among the clinicians.

Setting

Uterine fibroid scanned data from different ultrasonography machines (irrespective of make) with less than a maximum fibroid size not exceeding 5 cm with a maximum of 4 fibroids, were segmented by an AI-assisted algorithm.

Patients or Participants

A total of 100 patients with uterine fibroids were included in the study.

Interventions

Measurements and Primary Results

The acquired imaging data were segmented using the AI-assisted algorithm. The uterus and fibroid segmentation of 100 cases was successfully done by this algorithm. The precise size of the uterus and accurate mapping (size, location, and FIGO-type) of the fibroid were appreciated when these segmentation results were validated by 3 experienced clinicians with expertise in ultrasonography. Clinician -1 - was satisfied in 98% of cases, Clinician -2 - was satisfied in 97% of cases, and Clinician -3 - was satisfied in 99% of cases.

Conclusion

The AI-assisted algorithm demonstrated strong agreement with expert analysis of the segmentation of the uterus and fibroid by ultrasonography. It appears this novel AI algorithm is promising in fibroid segmentation and its accuracy should be analyzed in future studies. In the future, this AI algorithm could be used for surgical planning.

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

Journal of minimally invasive gynecology 医学-妇产科学

CiteScore

5.00

自引率

7.30%

发文量

272

审稿时长

37 days

期刊介绍： The Journal of Minimally Invasive Gynecology, formerly titled The Journal of the American Association of Gynecologic Laparoscopists, is an international clinical forum for the exchange and dissemination of ideas, findings and techniques relevant to gynecologic endoscopy and other minimally invasive procedures. The Journal, which presents research, clinical opinions and case reports from the brightest minds in gynecologic surgery, is an authoritative source informing practicing physicians of the latest, cutting-edge developments occurring in this emerging field.