Sociodemographic biases in a commercial AI model for intracranial hemorrhage detection.

IF 1.7 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Emergency Radiology Pub Date : 2024-10-01 Epub Date: 2024-07-22 DOI:10.1007/s10140-024-02270-w

Annie Trang, Kristin Putman, Dharmam Savani, Devina Chatterjee, Jerry Zhao, Peter Kamel, Jean J Jeudy, Vishwa S Parekh, Paul H Yi

{"title":"Sociodemographic biases in a commercial AI model for intracranial hemorrhage detection.","authors":"Annie Trang, Kristin Putman, Dharmam Savani, Devina Chatterjee, Jerry Zhao, Peter Kamel, Jean J Jeudy, Vishwa S Parekh, Paul H Yi","doi":"10.1007/s10140-024-02270-w","DOIUrl":null,"url":null,"abstract":"Purpose: To evaluate whether a commercial AI tool for intracranial hemorrhage (ICH) detection on head CT exhibited sociodemographic biases.Methods: Our retrospective study reviewed 9736 consecutive, adult non-contrast head CT scans performed between November 2021 and February 2022 in a single healthcare system. Each CT scan was evaluated by a commercial ICH AI tool and a board-certified neuroradiologist; ground truth was defined as final radiologist determination of ICH presence/absence. After evaluating the AI tool's aggregate diagnostic performance, sub-analyses based on sociodemographic groups (age, sex, race, ethnicity, insurance status, and Area of Deprivation Index [ADI] scores) assessed for biases. χ2 test or Fisher's exact tests evaluated for statistical significance with p ≤ 0.05.Results: Our patient population was 50% female (mean age 60 ± 19 years). The AI tool had an aggregate accuracy of 93% [9060/9736], sensitivity of 85% [1140/1338], specificity of 94% [7920/ 8398], positive predictive value (PPV) of 71% [1140/1618] and negative predictive value (NPV) of 98% [7920/8118]. Sociodemographic biases were identified, including lower PPV for patients who were females (67.3% [62,441/656] vs. 72.7% [699/962], p = 0.02), Black (66.7% [454/681] vs. 73.2% [686/937], p = 0.005), non-Hispanic/non-Latino (69.7% [1038/1490] vs. 95.4% [417/437]), p = 0.009), and who had Medicaid/Medicare (69.9% [754/1078]) or Private (66.5% [228/343]) primary insurance (p = 0.003). Lower sensitivity was seen for patients in the third quartile of national (78.8% [241/306], p = 0.001) and state ADI scores (79.0% [22/287], p = 0.001).Conclusions: In our healthcare system, a commercial AI tool had lower performance for ICH detection than previously reported and demonstrated several sociodemographic biases.","PeriodicalId":11623,"journal":{"name":"Emergency Radiology","volume":null,"pages":null},"PeriodicalIF":1.7000,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Emergency Radiology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1007/s10140-024-02270-w","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/7/22 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}

引用次数: 0

Abstract

Purpose: To evaluate whether a commercial AI tool for intracranial hemorrhage (ICH) detection on head CT exhibited sociodemographic biases.

Methods: Our retrospective study reviewed 9736 consecutive, adult non-contrast head CT scans performed between November 2021 and February 2022 in a single healthcare system. Each CT scan was evaluated by a commercial ICH AI tool and a board-certified neuroradiologist; ground truth was defined as final radiologist determination of ICH presence/absence. After evaluating the AI tool's aggregate diagnostic performance, sub-analyses based on sociodemographic groups (age, sex, race, ethnicity, insurance status, and Area of Deprivation Index [ADI] scores) assessed for biases. χ² test or Fisher's exact tests evaluated for statistical significance with p ≤ 0.05.

Results: Our patient population was 50% female (mean age 60 ± 19 years). The AI tool had an aggregate accuracy of 93% [9060/9736], sensitivity of 85% [1140/1338], specificity of 94% [7920/ 8398], positive predictive value (PPV) of 71% [1140/1618] and negative predictive value (NPV) of 98% [7920/8118]. Sociodemographic biases were identified, including lower PPV for patients who were females (67.3% [62,441/656] vs. 72.7% [699/962], p = 0.02), Black (66.7% [454/681] vs. 73.2% [686/937], p = 0.005), non-Hispanic/non-Latino (69.7% [1038/1490] vs. 95.4% [417/437]), p = 0.009), and who had Medicaid/Medicare (69.9% [754/1078]) or Private (66.5% [228/343]) primary insurance (p = 0.003). Lower sensitivity was seen for patients in the third quartile of national (78.8% [241/306], p = 0.001) and state ADI scores (79.0% [22/287], p = 0.001).

Conclusions: In our healthcare system, a commercial AI tool had lower performance for ICH detection than previously reported and demonstrated several sociodemographic biases.

查看原文本刊更多论文

颅内出血检测商业人工智能模型中的社会人口偏差。

目的：评估用于检测头部 CT 颅内出血（ICH）的商业人工智能工具是否存在社会人口学偏差：我们的回顾性研究回顾了 2021 年 11 月至 2022 年 2 月期间在一个医疗系统中进行的 9736 次连续成人非对比头部 CT 扫描。每张 CT 扫描均由一个商用 ICH AI 工具和一位经委员会认证的神经放射科医师进行评估；地面实况被定义为放射科医师对 ICH 存在/不存在的最终判断。在评估了人工智能工具的总体诊断性能后，根据社会人口群体（年龄、性别、种族、民族、保险状况和贫困地区指数 [ADI] 评分）进行了子分析，以评估是否存在偏差。采用χ2检验或费雪精确检验评估统计学意义，P≤0.05：50%的患者为女性（平均年龄为 60±19 岁）。人工智能工具的总准确率为 93% [9060/9736]，灵敏度为 85% [1140/1338]，特异性为 94% [7920/8398]，阳性预测值 (PPV) 为 71% [1140/1618]，阴性预测值 (NPV) 为 98% [7920/8118]。005）、非西班牙裔/非拉丁裔（69.7% [1038/1490] vs. 95.4% [417/437]），p = 0.009），以及拥有医疗补助/医疗保险（69.9% [754/1078]）或私人保险（66.5% [228/343]）（p = 0.003）的患者。全国（78.8% [241/306]，p = 0.001）和州 ADI 评分（79.0% [22/287]，p = 0.001）处于第三四分位数的患者敏感性较低：结论：在我们的医疗系统中，商业人工智能工具的 ICH 检测性能低于之前的报告，并表现出一些社会人口学偏差。

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

Emergency Radiology RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING-

CiteScore

4.60

自引率

4.50%

发文量

期刊介绍： To advance and improve the radiologic aspects of emergency careTo establish Emergency Radiology as an area of special interest in the field of diagnostic imagingTo improve methods of education in Emergency RadiologyTo provide, through formal meetings, a mechanism for presentation of scientific papers on various aspects of Emergency Radiology and continuing educationTo promote research in Emergency Radiology by clinical and basic science investigators, including residents and other traineesTo act as the resource body on Emergency Radiology for those interested in emergency patient care Members of the American Society of Emergency Radiology (ASER) receive the Emergency Radiology journal as a benefit of membership!