Virk Abdullah, Alasmari Safanah, Patel Deepkumar, Allison Karen
Department of Ophthalmology, Flaum Eye Institute, University of Rochester, Rochester, USA.
School of Health Sciences and Practice, New York Medical College, New York, USA.
Cureus. 2025 Mar 16;17(3):e80676. doi: 10.7759/cureus.80676. eCollection 2025 Mar.
The landscape of healthcare is rapidly changing with the increasing usage of machine and deep learning artificial intelligence and digital tools to assist in various sectors. This study aims to analyze the feasibility of the implementation of artificial intelligence (AI) models into healthcare systems. This review included English-language publications from databases such as SCOPUS, PubMed, and Google Scholar between 2000 and 2024. AI integration in healthcare systems will assist in large-scale dataset analysis, access to healthcare information, surgery data and simulation, and clinical decision-making in addition to many other healthcare services. However, with the reliance on AI, issues regarding medical liability, cybersecurity, and health disparities can form. This necessitates updates and transparency on health policy, AI training, and cybersecurity measures. To support the implementation of AI in healthcare, transparency regarding AI algorithm training and analytical approaches is key to allowing physicians to trust and make informed decisions about the applicability of AI results. Transparency will also allow healthcare systems to adapt appropriately, provide AI services, and create viable security measures. Furthermore, the increased diversity of data used in AI algorithm training will allow for greater generalizability of AI solutions in patient care. With the growth of AI usage and interaction with patient data, security measures and safeguards, such as system monitoring and cybersecurity training, should take precedence. Stricter digital policy and data protection guidelines will add additional layers of security for patient data. This collaboration will further bolster security measures amongst different regions and healthcare systems in addition to providing more means to innovative care. With the growing digitization of healthcare, advancing cybersecurity will allow effective and safe implementation of AI and other digital systems into healthcare and can improve the safety of patients and their personal health information.
随着机器学习和深度学习人工智能以及数字工具在各个领域的使用日益增加,医疗保健领域正在迅速变化。本研究旨在分析将人工智能(AI)模型应用于医疗保健系统的可行性。这篇综述纳入了2000年至2024年间来自SCOPUS、PubMed和谷歌学术等数据库的英文出版物。除了许多其他医疗保健服务外,人工智能在医疗保健系统中的整合将有助于大规模数据集分析、获取医疗保健信息、手术数据和模拟以及临床决策。然而,随着对人工智能的依赖,可能会出现医疗责任、网络安全和健康差距等问题。这就需要在健康政策、人工智能培训和网络安全措施方面进行更新并保持透明。为了支持人工智能在医疗保健中的应用,人工智能算法训练和分析方法的透明度是让医生信任并就人工智能结果的适用性做出明智决策的关键。透明度还将使医疗保健系统能够适当地调整、提供人工智能服务并制定可行的安全措施。此外,人工智能算法训练中使用的数据多样性增加将使人工智能解决方案在患者护理中具有更大的通用性。随着人工智能使用的增加以及与患者数据的交互,安全措施和保障,如系统监控和网络安全培训,应优先考虑。更严格的数字政策和数据保护指南将为患者数据增加额外的安全层。这种合作除了提供更多创新护理手段外,还将进一步加强不同地区和医疗保健系统之间的安全措施。随着医疗保健数字化程度的不断提高,推进网络安全将使人工智能和其他数字系统能够有效且安全地应用于医疗保健,并可提高患者及其个人健康信息的安全性。