This study conducts a thorough examination of the research stream focusing on AI risks in healthcare, aiming to explore the distinct genres within this domain. A selection criterion was employed to carefully analyze 39 articles to identify three primary genres of AI risks prevalent in healthcare: clinical data risks, technical risks, and socio-ethical risks. Selection criteria was based on journal ranking and impact factor. The research seeks to provide a valuable resource for future healthcare researchers, furnishing them with a comprehensive understanding of the complex challenges posed by AI implementation in healthcare settings. By categorizing and elucidating these genres, the study aims to facilitate the development of empirical qualitative and quantitative research, fostering evidence-based approaches to address AI-related risks in healthcare effectively. This endeavor contributes to building a robust knowledge base that can inform the formulation of risk mitigation strategies, ensuring safe and efficient integration of AI technologies in healthcare practices. Thus, it is important to study AI risks in healthcare to build better and efficient AI systems and mitigate risks.

翻译：本研究对医疗领域人工智能风险的研究脉络进行了系统性审视，旨在探索该领域中的不同风险类别。通过采用基于期刊排名与影响因子的筛选标准，我们严格分析了39篇文献，识别出医疗领域三大主要人工智能风险类别：临床数据风险、技术风险以及社会伦理风险。本研究旨在为未来的医疗研究人员提供宝贵资源，使其全面理解人工智能在医疗场景中应用所带来的复杂挑战。通过对这些风险类别进行归类与阐释，本研究有助于推动实证性质性与定量研究的发展，促进基于证据的方法以有效应对医疗领域与人工智能相关的风险。此项工作有助于构建稳健的知识基础，为制定风险缓解策略提供依据，从而确保人工智能技术在医疗实践中的安全高效整合。因此，研究医疗领域的人工智能风险对于构建更优、更高效的人工智能系统及降低风险具有重要意义。