Amponsah Daniel, Thamman Ritu, Brandt Eric, James Cornelius, Spector-Bagdady Kayte, Yong Celina M
Division of Cardiovascular Medicine and Cardiovascular Institute, Stanford University, Stanford, CA, USA.
University of Pittsburgh School of Medicine, Pittsburgh, PA, USA.
Curr Cardiovasc Risk Rep. 2024 Nov;18(11):153-162. doi: 10.1007/s12170-024-00745-6. Epub 2024 Aug 20.
The integration of artificial intelligence (AI) in medicine holds promise for transformative advancements aimed at improving healthcare outcomes. Amidst this promise, AI has been envisioned as a tool to detect and mitigate racial and ethnic inequity known to plague current cardiovascular care. However, this enthusiasm is dampened by the recognition that AI itself can harbor and propagate biases, necessitating a careful approach to ensure equity. This review highlights topics in the landscape of AI in cardiology, its role in identifying and addressing healthcare inequities, promoting diversity in research, concerns surrounding its applications, and proposed strategies for fostering equitable utilization.
Artificial intelligence has proven to be a valuable tool for clinicians in diagnosing and mitigating racial and ethnic inequities in cardiology, as well as the promotion of diversity in research. This promise is counterbalanced by the cautionary reality that AI can inadvertently perpetuate existent biases stemming from limited diversity in training data, inherent biases within datasets, and inadequate bias detection and monitoring mechanisms. Recognizing these concerns, experts emphasize the need for rigorous efforts to address these limitations in the development and deployment of AI within medicine.
Implementing AI in cardiovascular care to identify and address racial and ethnic inequities requires careful design and execution, beginning with meticulous data collection and a thorough review of training datasets. Furthermore, ensuring equitable performance involves rigorous testing and continuous surveillance of algorithms. Lastly, the promotion of diversity in the AI workforce and engagement of stakeholders are crucial to the advancement of equity to ultimately realize the potential for artificial intelligence for cardiovascular health equity.
将人工智能（AI）整合到医学中有望带来变革性进展，旨在改善医疗保健结果。在此前景下，人工智能被视为一种工具，用于检测和缓解困扰当前心血管护理的种族和民族不平等问题。然而，由于认识到人工智能本身可能存在并传播偏见，因此需要谨慎对待以确保公平性。本综述重点介绍了心脏病学领域人工智能的相关主题、其在识别和解决医疗保健不平等问题中的作用、促进研究多样性、围绕其应用的担忧以及促进公平利用的建议策略。
事实证明，人工智能对临床医生来说是一种有价值的工具，可用于诊断和缓解心脏病学中的种族和民族不平等问题，以及促进研究多样性。这一前景被一个警示性的现实所抵消，即人工智能可能会无意中延续因训练数据多样性有限、数据集中的固有偏见以及偏见检测和监测机制不足而产生的现有偏见。认识到这些问题，专家们强调需要做出严格努力，以解决医学领域人工智能开发和部署中的这些局限性。
在心血管护理中应用人工智能以识别和解决种族和民族不平等问题需要精心设计和执行，首先要进行细致的数据收集并对训练数据集进行全面审查。此外，确保公平性能需要对算法进行严格测试和持续监测。最后，促进人工智能领域劳动力的多样性以及让利益相关者参与进来对于推进公平性至关重要，最终实现人工智能在心血管健康公平方面的潜力。
