2024 , Vol. 18 >Issue 11: 1061 - 1065
DOI: https://doi.org/10.3877/cma.j.issn.1674-0785.2024.11.013
人工智能辅助乳腺影像学检查造成的自动偏差
收稿日期: 2024-09-27
网络出版日期: 2025-03-06
基金资助
陕西省卫生健康科研基金项目（2022A010）陕西省重点研发计划（2022SF-010）
版权
Automated bias caused by artificial intelligence-assisted breast imaging screening
Received date: 2024-09-27
Online published: 2025-03-06
Copyright
乳腺癌是女性最常见的恶性肿瘤，严重威胁着女性的身心健康。大量循证医学证据表明，乳腺癌早诊、早治可改善预后以及降低病死率。近年来，人工智能技术在乳腺肿瘤图像分析领域受到前所未有的研究热情。大量研究表明，人工智能辅助下的影像学检查具有高灵敏度、低疲劳性等优点，使得人工智能相关产品逐渐上市并被应用到各大医院，但在使用人工智能时仍会因为过度依赖等原因导致自动偏差，自动偏差的出现会影响影像科医生的决策，从而导致漏诊或误诊，造成适得其反的后果，本文将对人工智能辅助乳腺影像学检查所导致的自动偏差进行综述。
梁博宇, 宋宏萍, 宋张骏, 巨艳, 王虎霞, 薛文欣 . 人工智能辅助乳腺影像学检查造成的自动偏差[J]. 中华临床医师杂志(电子版), 2024 , 18(11) : 1061 -1065 . DOI: 10.3877/cma.j.issn.1674-0785.2024.11.013
Breast cancer is the most common malignant tumor in women, which seriously threatens women's physical and mental health. Evidence-based medical evidence shows that early diagnosis and treatment of breast cancer can improve the prognosis and reduce the mortality rate. In recent years,artificial intelligence (AI) technology has received unprecedented research enthusiasm in the field of breast tumor image analysis. Numerous studies have shown that AI-assisted imaging has the advantages of high sensitivity and low fatigue, which makes AI-related products gradually marketed and applied to major hospitals. However, when using AI, automated bias can still occur due to excessive dependence and other reasons. The occurrence of automated bias can affect the decision-making of radiologists, leading to missed or misdiagnosed cases and causing counterproductive consequences. This article will review the automated bias caused by AI-assisted breast imaging examinations.
| 1 | Siegel RL, Miller KD, Fuchs H, et al. Cancer Statistics, 2021 [J]. CA Cancer J Clin, 2021, 71: 7-33. |
| 2 | Lee HJ, Kim EK, Kim MJ, et al. Observer variability of Breast Imaging Reporting and Data System (BI-RADS) for breast ultrasound[J]. Eur J Radiol, 2008, 65(2): 293-298. |
| 3 | Abdullah N, Mesurolle B, El-Khoury M, et al. Breast imaging reporting and data system lexicon for US: interobserver agreement for assessment of breast masses [J]. Radiology, 2009, 252(3): 665-672. |
| 4 | 邱凯锋, 王则远, 何志超, 等. 人工智能技术在超说明书用药循证中的应用研究 [J/OL]. 中华临床医师杂志(电子版), 2023, 17(12):1212-1218. |
| 5 | Yang S, Gao X, Liu L, et al. Performance and reading time of automated breast US with or without computer-aided detection [J].Radiology, 2019, 3: 540-549. |
| 6 | Benjamens S, Dhunnoo P, Meskó B. The state of artificial intelligence based FDA-approved medical devices and algorithms: an online database [J]. NPJ Digit Med, 2020, 3(1): 118. |
| 7 | Tam MDBS, Dyer T, Dissez G, et al. Augmenting lung cancer diagnosis on chest radiographs: positioning artificial intelligence to improve radiologist performance [J]. Clin Radiol, 2021, 76(8): 607-614. |
| 8 | Finck T, Moosbauer J, Probst M, et al. Faster and better: How anomaly detection can accelerate and improve reporting of head computed tomography [J]. Diagnostics (Basel), 2022, 12(2): 452. |
| 9 | Sele D, Chugunova M. Putting a human in the loop: Increasing uptake,but decreasing accuracy of automated decision-making [J]. PLoS One,2024, 19(2): e0298037. |
| 10 | Parasuraman R, Manzey DH. Complacency and bias in human use of automation: an attentional integration [J]. Hum Factors, 2010, 52(3):381-410. |
| 11 | Chan HP，Samala RK，Hadjiiski LM. CAD and AI for breast cancerrecent development and challenges [J]. Br J Radiol, 2020, 93(1108):20190580. |
| 12 | Lehman CD, Wellman RD, Buist DSM, et al. Diagnostic accuracy of digital screening mammography with and without computer-aided detection [J]. JAMA Intern Med, 2015, 75(11): 1828-1837. |
| 13 | Hofvind S, Lee CI. A warning about warning signals for interpreting mammograms [J]. Radiology, 2022, 302(2): 284-285. |
| 14 | Alberdi RZ, Llanes ABF, Ortega RA, et al. Effect of radiologist experience on the risk of false-positive results in breast cancer screening programs [J]. Eur Radiol, 2011, 21(10): 2083-2090. |
| 15 | Skitka LJ, Mosier K, Burdick MD. Accountability and automation bias[J]. Int J Hum Comput Stud, 2000, 52(4): 701-717. |
| 16 | Goddard K, Roudsari A, Wyatt JC. Automation bias: a systematic review of frequency, effect mediators, and mitigators [J]. J Am Med Inform Assoc, 2012, 19(1): 121-127. |
| 17 | Cabitza F, Rasoini R, Gensini GF. Unintended consequences of machine learning in medicine [J]. JAMA, 2017, 318(6): 517-518. |
| 18 | Mosier KL, Skitka LJ, Heers S, et al. Automation bias: decision making and performance in high-tech cockpits [J]. Int J Aviat Psychol,1997, 8(1): 47-63. |
| 19 | Kwon MR, Youn I, Lee MY, et al. Diagnostic performance of artificial intelligence-based computer-aided detection software for automated breast ultrasound [J]. Acad Radiol, 2024 31(2): 480-491. |
| 20 | Dratsch T, Chen X, Rezazade Mehrizi M, et al. Automation bias in mammography: the impact of artificial intelligence BI-RADS suggestions on reader performance [J]. Radiology, 2023, 4: e222176. |
| 21 | Alberdi E, Povykalo A, Strigini L, et al. Effects of incorrect computeraided detection (CAD) output on human decision-making in mammography [J]. Acad Radiol, 2004, 11(8): 909-918. |
| 22 | Lyell D, Coiera E. Automation bias and verification complexity: a systematic review [J]. J Am Med Inform Assoc, 2017, 2: 423-431. |
| 23 | McGuirl JM, Sarter NB. Supporting trust calibration and the effective use of decision aids by presenting dynamic system confidence information [J]. Hum Factors, 2006, 48(4): 656-665. |
| 24 | Dzindolet MT, Peterson SA, Pomranky RA, et al. The role of trust in automation reliance [J]. Int J Hum Comput Stud, 2003, 58(6): 697-718. |
| 25 | Mosier KL, Skitka LJ, Heers S, et al. Automation bias: decision making and performance in high-tech cockpits [J]. Int J Aviat Psychol,2017, 271-287. |
| 26 | Xu X, Bao L, Tan Y, et al. 1000-case reader study of radiologists’performance in interpretation of automated breast volume scanner images with a computer-aided detection system [J]. Ultrasound Med Biol, 2018, 8: 1694-1702. |
| 27 | Skitka LJ, Mosier K, Burdick MD. Accountability and automation bias[J]. International J Hum Comp Stud, 2000, 52(4): 701-717. |
| 28 | Bagheri N, Jamieson GA. The impact of context-related reliability on automation failure detection and scanning behaviour [J]. Conf Proc IEEE Int Conf Syst Man Cybern, 2004, 2004: 212-217. |
| 29 | Coiera E, Westbrook J, Wyatt J. The safety and quality of decision support systems [J]. Methods Inf Med, 2006, 45(Suppl 1): 20-25. |
| 30 | Singh AL, Tiwari T, Singh IL. Effects of automation reliability and training on automation-induced complacency and perceived mental workload [J]. J Indian Acad Appl Psychol, 2009, 35(Spec iss): 9-22. |
| 31 | Bahner J, Huper A-D, Manzey D. Misuse of automated decision aids:complacency, automation bias and the impact of training experience [J].Int J Hum Comp Stud, 2008, 66(9): 688-699. |
/
| 〈 |
|
〉 |