结合图像相似性和人工智能预测模型，减少甲状腺结节诊断中的主观性并改善恶性肿瘤预测

IF 3.7 3区医学 Q2 ENDOCRINOLOGY & METABOLISM

Endocrine Practice Pub Date : 2024-11-01 DOI:10.1016/j.eprac.2024.08.001

Govind Nair , Aishwarya Vedula , Ethan Thomas Johnson , Johnson Thomas MD , Rajshree Patel MD , Jennifer Cheng DO , Ramya Vedula DO, MPH

{"title":"结合图像相似性和人工智能预测模型，减少甲状腺结节诊断中的主观性并改善恶性肿瘤预测","authors":"Govind Nair , Aishwarya Vedula , Ethan Thomas Johnson , Johnson Thomas MD , Rajshree Patel MD , Jennifer Cheng DO , Ramya Vedula DO, MPH","doi":"10.1016/j.eprac.2024.08.001","DOIUrl":null,"url":null,"abstract":"<div><h3>Objectives</h3><div>To evaluate the efficacy of combining predictive artificial intelligence (AI) and image similarity model to risk stratify thyroid nodules, using retrospective external validation study.</div></div><div><h3>Methods</h3><div>Two datasets were used to determine efficacy of the AI application. One was Stanford dataset ultrasound images of 192 nodules between April 2017 and May 2018 and the second was private practice consisting of 118 thyroid nodule images between January 2018 and December 2023. The nodules had definitive diagnosis by cytology or surgical pathology. The AI application was used to predict the diagnosis and American College of Radiology Thyroid Imaging and Data System (ACR TI-RADS) score.</div></div><div><h3>Results</h3><div>In the Stanford dataset, the AI application predicted malignancies with sensitivity of 1.0 and specificity of 0.55. Positive predictive value (PPV) was 0.18 and negative predictive value (NPV) was 1.0. The Area Under the Curve - Receiver Operating Characteristic was 0.78. ACR TI-RADS based clinical recommendation had a polychoric correlation of 0.67. In the private dataset, the AI application predicted malignancies with sensitivity of 0.91 and specificity of 0.95. PPV was 0.8 and NPV was 0.98. The area under the curve - receiver operating characteristic was 0.93 and accuracy was 0.94. ACR TI-RADS based score had a polychoric correlation of 0.94.</div></div><div><h3>Conclusion</h3><div>The AI application showed good performance for sensitivity and NPV between the two datasets and demonstrated potential for 61.5% reduction in the need for fine needle aspiration and strong correlation to ACR TI-RADS. However, PPV was variable between the datasets possibly from variability in image selection and prevalence of malignancy. If implemented widely and consistently among various clinical settings, this could lead to decreased patient burden associated with an invasive procedure and possibly to decreased health care spending.</div></div>","PeriodicalId":11682,"journal":{"name":"Endocrine Practice","volume":"30 11","pages":"Pages 1031-1037"},"PeriodicalIF":3.7000,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Combining Image Similarity and Predictive Artificial Intelligence Models to Decrease Subjectivity in Thyroid Nodule Diagnosis and Improve Malignancy Prediction\",\"authors\":\"Govind Nair , Aishwarya Vedula , Ethan Thomas Johnson , Johnson Thomas MD , Rajshree Patel MD , Jennifer Cheng DO , Ramya Vedula DO, MPH\",\"doi\":\"10.1016/j.eprac.2024.08.001\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Objectives</h3><div>To evaluate the efficacy of combining predictive artificial intelligence (AI) and image similarity model to risk stratify thyroid nodules, using retrospective external validation study.</div></div><div><h3>Methods</h3><div>Two datasets were used to determine efficacy of the AI application. One was Stanford dataset ultrasound images of 192 nodules between April 2017 and May 2018 and the second was private practice consisting of 118 thyroid nodule images between January 2018 and December 2023. The nodules had definitive diagnosis by cytology or surgical pathology. The AI application was used to predict the diagnosis and American College of Radiology Thyroid Imaging and Data System (ACR TI-RADS) score.</div></div><div><h3>Results</h3><div>In the Stanford dataset, the AI application predicted malignancies with sensitivity of 1.0 and specificity of 0.55. Positive predictive value (PPV) was 0.18 and negative predictive value (NPV) was 1.0. The Area Under the Curve - Receiver Operating Characteristic was 0.78. ACR TI-RADS based clinical recommendation had a polychoric correlation of 0.67. In the private dataset, the AI application predicted malignancies with sensitivity of 0.91 and specificity of 0.95. PPV was 0.8 and NPV was 0.98. The area under the curve - receiver operating characteristic was 0.93 and accuracy was 0.94. ACR TI-RADS based score had a polychoric correlation of 0.94.</div></div><div><h3>Conclusion</h3><div>The AI application showed good performance for sensitivity and NPV between the two datasets and demonstrated potential for 61.5% reduction in the need for fine needle aspiration and strong correlation to ACR TI-RADS. However, PPV was variable between the datasets possibly from variability in image selection and prevalence of malignancy. If implemented widely and consistently among various clinical settings, this could lead to decreased patient burden associated with an invasive procedure and possibly to decreased health care spending.</div></div>\",\"PeriodicalId\":11682,\"journal\":{\"name\":\"Endocrine Practice\",\"volume\":\"30 11\",\"pages\":\"Pages 1031-1037\"},\"PeriodicalIF\":3.7000,\"publicationDate\":\"2024-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Endocrine Practice\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1530891X24006475\",\"RegionNum\":3,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENDOCRINOLOGY & METABOLISM\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Endocrine Practice","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1530891X24006475","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}

引用次数: 0

摘要

目的通过回顾性外部验证研究，评估结合预测性人工智能（AI）和图像相似性模型对甲状腺结节进行风险分层的效果：方法：使用两个数据集来确定人工智能应用的有效性。一个是2017年4月至2018年5月期间192个结节的斯坦福数据集超声图像，第二个是2018年1月至2023年12月期间118个甲状腺结节图像的私人实践。这些结节经细胞学或手术病理学明确诊断。人工智能应用被用于预测诊断和美国放射学会甲状腺成像和数据系统（ACR TI-RADS）评分：在斯坦福数据集中，人工智能应用预测恶性肿瘤的灵敏度为 1.0，特异性为 0.55。阳性预测值（PPV）为 0.18，阴性预测值（NPV）为 1.0。曲线下面积--接收者操作特征（AUC-ROC）为 0.78。基于 ACR TI-RADS 的临床建议的多变量相关性为 0.67。在私人数据集中，人工智能应用预测恶性肿瘤的灵敏度为 0.91，特异性为 0.95。PPV 为 0.8，NPV 为 0.98。AUC-ROC为0.93，准确率为0.94。基于 ACR TI-RADS 的评分的多变量相关性为 0.94：人工智能应用在两个数据集之间显示出良好的灵敏度和 NPV 性能，并有可能减少 61.5% 的细针穿刺 (FNA)，与 ACR TI-RADS 有很强的相关性。不过，可能由于图像选择和恶性肿瘤发生率的不同，两个数据集之间的 PPV 存在差异。如果能在不同的临床环境中广泛一致地实施，就能减轻患者与侵入性手术相关的负担，并可能减少医疗支出。

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

Combining Image Similarity and Predictive Artificial Intelligence Models to Decrease Subjectivity in Thyroid Nodule Diagnosis and Improve Malignancy Prediction

Objectives

To evaluate the efficacy of combining predictive artificial intelligence (AI) and image similarity model to risk stratify thyroid nodules, using retrospective external validation study.

Methods

Two datasets were used to determine efficacy of the AI application. One was Stanford dataset ultrasound images of 192 nodules between April 2017 and May 2018 and the second was private practice consisting of 118 thyroid nodule images between January 2018 and December 2023. The nodules had definitive diagnosis by cytology or surgical pathology. The AI application was used to predict the diagnosis and American College of Radiology Thyroid Imaging and Data System (ACR TI-RADS) score.

Results

In the Stanford dataset, the AI application predicted malignancies with sensitivity of 1.0 and specificity of 0.55. Positive predictive value (PPV) was 0.18 and negative predictive value (NPV) was 1.0. The Area Under the Curve - Receiver Operating Characteristic was 0.78. ACR TI-RADS based clinical recommendation had a polychoric correlation of 0.67. In the private dataset, the AI application predicted malignancies with sensitivity of 0.91 and specificity of 0.95. PPV was 0.8 and NPV was 0.98. The area under the curve - receiver operating characteristic was 0.93 and accuracy was 0.94. ACR TI-RADS based score had a polychoric correlation of 0.94.

Conclusion

The AI application showed good performance for sensitivity and NPV between the two datasets and demonstrated potential for 61.5% reduction in the need for fine needle aspiration and strong correlation to ACR TI-RADS. However, PPV was variable between the datasets possibly from variability in image selection and prevalence of malignancy. If implemented widely and consistently among various clinical settings, this could lead to decreased patient burden associated with an invasive procedure and possibly to decreased health care spending.

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

Endocrine Practice ENDOCRINOLOGY & METABOLISM-

CiteScore

7.60

自引率

2.40%

发文量

546

审稿时长

41 days

期刊介绍： Endocrine Practice (ISSN: 1530-891X), a peer-reviewed journal published twelve times a year, is the official journal of the American Association of Clinical Endocrinologists (AACE). The primary mission of Endocrine Practice is to enhance the health care of patients with endocrine diseases through continuing education of practicing endocrinologists.