{"title":"应用深度学习对甲状腺乳头状癌超声图像上的侧淋巴结进行自动分割和分类","authors":"","doi":"10.1016/j.asjsur.2024.02.140","DOIUrl":null,"url":null,"abstract":"<div><h3>Purpose</h3><p>It is crucial to preoperatively diagnose lateral cervical lymph node (LN) metastases (LNMs) in papillary thyroid carcinoma (PTC) patients. This study aims to develop deep-learning models for the automatic segmentation and classification of LNM on original ultrasound images.</p></div><div><h3>Methods</h3><p>This study included 1000 lateral cervical LN ultrasound images (consisting of 512 benign and 558 metastatic LNs) collected from 728 patients at the Chongqing General Hospital between March 2022 and July 2023. Three instance segmentation models (MaskRCNN, SOLO and Mask2Former) were constructed to segment and classify ultrasound images of lateral cervical LNs by recognizing each object individually and in a pixel-by-pixel manner. The segmentation and classification results of the three models were compared with an experienced sonographer in the test set.</p></div><div><h3>Results</h3><p>Upon completion of a 200-epoch learning cycle, the loss among the three unique models became negligible. To evaluate the performance of the deep-learning models, the intersection over union threshold was set at 0.75. The mean average precision scores for MaskRCNN, SOLO and Mask2Former were 88.8%, 86.7% and 89.5%, respectively. The segmentation accuracies of the MaskRCNN, SOLO, Mask2Former models and sonographer were 85.6%, 88.0%, 89.5% and 82.3%, respectively. The classification AUCs of the MaskRCNN, SOLO, Mask2Former models and sonographer were 0.886, 0.869, 0.90.2 and 0.852 in the test set, respectively.</p></div><div><h3>Conclusions</h3><p>The deep learning models could automatically segment and classify lateral cervical LNs with an AUC of 0.92. This approach may serve as a promising tool to assist sonographers in diagnosing lateral cervical LNMs among patients with PTC.</p></div>","PeriodicalId":55454,"journal":{"name":"Asian Journal of Surgery","volume":null,"pages":null},"PeriodicalIF":3.5000,"publicationDate":"2024-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S1015958424004019/pdfft?md5=90c6ce5c58b86ae128c55d810a53fb2b&pid=1-s2.0-S1015958424004019-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Application of deep-learning to the automatic segmentation and classification of lateral lymph nodes on ultrasound images of papillary thyroid carcinoma\",\"authors\":\"\",\"doi\":\"10.1016/j.asjsur.2024.02.140\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Purpose</h3><p>It is crucial to preoperatively diagnose lateral cervical lymph node (LN) metastases (LNMs) in papillary thyroid carcinoma (PTC) patients. This study aims to develop deep-learning models for the automatic segmentation and classification of LNM on original ultrasound images.</p></div><div><h3>Methods</h3><p>This study included 1000 lateral cervical LN ultrasound images (consisting of 512 benign and 558 metastatic LNs) collected from 728 patients at the Chongqing General Hospital between March 2022 and July 2023. Three instance segmentation models (MaskRCNN, SOLO and Mask2Former) were constructed to segment and classify ultrasound images of lateral cervical LNs by recognizing each object individually and in a pixel-by-pixel manner. The segmentation and classification results of the three models were compared with an experienced sonographer in the test set.</p></div><div><h3>Results</h3><p>Upon completion of a 200-epoch learning cycle, the loss among the three unique models became negligible. To evaluate the performance of the deep-learning models, the intersection over union threshold was set at 0.75. The mean average precision scores for MaskRCNN, SOLO and Mask2Former were 88.8%, 86.7% and 89.5%, respectively. The segmentation accuracies of the MaskRCNN, SOLO, Mask2Former models and sonographer were 85.6%, 88.0%, 89.5% and 82.3%, respectively. The classification AUCs of the MaskRCNN, SOLO, Mask2Former models and sonographer were 0.886, 0.869, 0.90.2 and 0.852 in the test set, respectively.</p></div><div><h3>Conclusions</h3><p>The deep learning models could automatically segment and classify lateral cervical LNs with an AUC of 0.92. This approach may serve as a promising tool to assist sonographers in diagnosing lateral cervical LNMs among patients with PTC.</p></div>\",\"PeriodicalId\":55454,\"journal\":{\"name\":\"Asian Journal of Surgery\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.5000,\"publicationDate\":\"2024-09-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S1015958424004019/pdfft?md5=90c6ce5c58b86ae128c55d810a53fb2b&pid=1-s2.0-S1015958424004019-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Asian Journal of Surgery\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1015958424004019\",\"RegionNum\":3,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"SURGERY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Asian Journal of Surgery","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1015958424004019","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"SURGERY","Score":null,"Total":0}
Application of deep-learning to the automatic segmentation and classification of lateral lymph nodes on ultrasound images of papillary thyroid carcinoma
Purpose
It is crucial to preoperatively diagnose lateral cervical lymph node (LN) metastases (LNMs) in papillary thyroid carcinoma (PTC) patients. This study aims to develop deep-learning models for the automatic segmentation and classification of LNM on original ultrasound images.
Methods
This study included 1000 lateral cervical LN ultrasound images (consisting of 512 benign and 558 metastatic LNs) collected from 728 patients at the Chongqing General Hospital between March 2022 and July 2023. Three instance segmentation models (MaskRCNN, SOLO and Mask2Former) were constructed to segment and classify ultrasound images of lateral cervical LNs by recognizing each object individually and in a pixel-by-pixel manner. The segmentation and classification results of the three models were compared with an experienced sonographer in the test set.
Results
Upon completion of a 200-epoch learning cycle, the loss among the three unique models became negligible. To evaluate the performance of the deep-learning models, the intersection over union threshold was set at 0.75. The mean average precision scores for MaskRCNN, SOLO and Mask2Former were 88.8%, 86.7% and 89.5%, respectively. The segmentation accuracies of the MaskRCNN, SOLO, Mask2Former models and sonographer were 85.6%, 88.0%, 89.5% and 82.3%, respectively. The classification AUCs of the MaskRCNN, SOLO, Mask2Former models and sonographer were 0.886, 0.869, 0.90.2 and 0.852 in the test set, respectively.
Conclusions
The deep learning models could automatically segment and classify lateral cervical LNs with an AUC of 0.92. This approach may serve as a promising tool to assist sonographers in diagnosing lateral cervical LNMs among patients with PTC.
期刊介绍:
Asian Journal of Surgery, launched in 1978, is the official peer-reviewed open access journal of the Asian Surgical Association, the Taiwan Robotic Surgery Association, and the Taiwan Society of Coloproctology. The Journal is published monthly by Elsevier and is indexed in SCIE, Medline, ScienceDirect, Scopus, Embase, Current Contents, PubMed, Current Abstracts, BioEngineering Abstracts, SIIC Data Bases, CAB Abstracts, and CAB Health.
ASJSUR has a growing reputation as an important medium for the dissemination of cutting-edge developments in surgery and its related disciplines in the Asia-Pacific region and beyond. Studies on state-of-the-art surgical innovations across the entire spectrum of clinical and experimental surgery are particularly welcome.
The journal publishes original articles, review articles, and case reports that are of exceptional and unique importance. The journal publishes original articles, review articles, and case reports that are of exceptional and unique importance.