Evaluation of pathologically confirmed benign inflammatory breast diseases using artificial intelligence on ultrasound images

IF 0.2 Q4 OBSTETRICS & GYNECOLOGY

Revista de Senologia y Patologia Mamaria Pub Date : 2023-11-07 DOI:10.1016/j.senol.2023.100558

Irmak Durur-Subasi , Abdulkadir Eren , Fatma Zeynep Gungoren , Pelin Basim , Fazli Cem Gezen , Asli Cakir , Cengiz Erol , Ilker Ozgur Koska

{"title":"Evaluation of pathologically confirmed benign inflammatory breast diseases using artificial intelligence on ultrasound images","authors":"Irmak Durur-Subasi , Abdulkadir Eren , Fatma Zeynep Gungoren , Pelin Basim , Fazli Cem Gezen , Asli Cakir , Cengiz Erol , Ilker Ozgur Koska","doi":"10.1016/j.senol.2023.100558","DOIUrl":null,"url":null,"abstract":"<div><h3>Objectives</h3><p>It was aimed to use AI retrospectively to evaluate US images of pathologically confirmed benign inflammatory lesions, to compare the results of AI with our US reports, and to test the reliability of AI in itself.</p></div><div><h3>Methods</h3><p>US images of 71 histopathologically confirmed benign inflammatory breast lesions were analysed by the FDA-approved AI programme (Koios Decision Support) using 2 orthogonal projections. The lesions' probability of malignancy based on AI and BI-RADS categories of the lesion based on initial US interpretations were recorded. Categories obtained by both systems were divided into 2 groups as unsuspicious and suspicious in terms of malignancy and compared statistically. Reliability of AI was also evaluated.</p></div><div><h3>Results</h3><p>No statistically significant difference was found in the lesions' likelihood of malignancy based on the AI and initial US interpretations (<em>P</em> = .512). Additionally, a positive and substantial association (τ-b = 0.458, <em>P</em> < .001) between the levels of suspicion by AI and the initial US interpretation reports was discovered, as per Kendall-b correlation analysis. With a Cronbach alpha correlation coefficient of 0.727, the reliability was high for AI.</p></div><div><h3>Conclusions</h3><p>Benign inflammatory breast lesions may show suspicious appearances in terms of malignancy with US and AI. Artificial intelligence produces results comparable to radiologists' US reports for benign inflammatory diseases. AI has high reliability within itself.</p></div>","PeriodicalId":38058,"journal":{"name":"Revista de Senologia y Patologia Mamaria","volume":"37 1","pages":"Article 100558"},"PeriodicalIF":0.2000,"publicationDate":"2023-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Revista de Senologia y Patologia Mamaria","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0214158223000907","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"OBSTETRICS & GYNECOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Objectives

It was aimed to use AI retrospectively to evaluate US images of pathologically confirmed benign inflammatory lesions, to compare the results of AI with our US reports, and to test the reliability of AI in itself.

Methods

US images of 71 histopathologically confirmed benign inflammatory breast lesions were analysed by the FDA-approved AI programme (Koios Decision Support) using 2 orthogonal projections. The lesions' probability of malignancy based on AI and BI-RADS categories of the lesion based on initial US interpretations were recorded. Categories obtained by both systems were divided into 2 groups as unsuspicious and suspicious in terms of malignancy and compared statistically. Reliability of AI was also evaluated.

Results

No statistically significant difference was found in the lesions' likelihood of malignancy based on the AI and initial US interpretations (P = .512). Additionally, a positive and substantial association (τ-b = 0.458, P < .001) between the levels of suspicion by AI and the initial US interpretation reports was discovered, as per Kendall-b correlation analysis. With a Cronbach alpha correlation coefficient of 0.727, the reliability was high for AI.

Conclusions

Benign inflammatory breast lesions may show suspicious appearances in terms of malignancy with US and AI. Artificial intelligence produces results comparable to radiologists' US reports for benign inflammatory diseases. AI has high reliability within itself.

查看原文本刊更多论文

超声图像人工智能对病理证实的乳腺良性炎性疾病的评价

目的回顾性地使用人工智能评估病理证实的良性炎性病变的超声图像，将人工智能的结果与我们的美国报告进行比较，并测试人工智能本身的可靠性。方法对71例经组织病理学证实的乳腺良性炎性病变的影像进行分析，采用fda批准的AI程序(Koios Decision Support)，采用2个正交投影法。记录基于AI的病变恶性概率和基于初始US判读的病变BI-RADS分类。将两种系统获得的分类分为恶性程度不可疑和可疑两组，并进行统计学比较。对人工智能的可靠性也进行了评价。结果基于AI和初始US解释的病变恶性可能性无统计学差异(P = .512)。此外，根据Kendall-b相关分析，发现人工智能的怀疑程度与美国最初的口译报告之间存在正相关关系(τ-b = 0.458,P < .001)。Cronbach α相关系数为0.727，信度较高。结论乳腺良性炎性病变在US和AI表现为可疑的恶性肿瘤。人工智能产生的结果与美国放射科医生对良性炎症疾病的报告相当。人工智能本身具有很高的可靠性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊