Search strings and search results.

The healthcare sector is rapidly evolving with the integration of Artificial Intelligence (AI). As AI technologies shift from rule-based expert systems to deep learning architectures, AI-based chatbots have emerged as innovative solutions to persistent challenges in the health domain. Given the growing concerns about their effectiveness and ethical implications, as well as the demand to optimise their potential in facilitating health outcomes, this study conducts a systematic review of existing research on AI-based chatbots, focusing on their applications and evaluation. A total of 348 articles, collected from eight databases—PubMed/MEDLINE, EMBASE, PsycINFO, CINAHL, IEEE, the ACM Digital Library, Scopus, and Web of Science - 20 of which were analysed. This review identifies four main research areas concerning AI-based chatbots: text quality, clinical efficacy, user engagement, and safety. It also highlights the lack of randomised controlled trials (RCTs) and the limited use of theoretical frameworks in evaluating their performance. Future research directions and practical solutions are discussed.医疗保健领域正随着人工智能（Artificial Intelligence）的融入而快速演进。随着人工智能技术从基于规则的专家系统向深度学习架构转型，基于人工智能的聊天机器人已成为解决医疗领域长期存在的各类难题的创新解决方案。鉴于学界与公众对其有效性及伦理影响的担忧日益加剧，同时为优化其在促进健康结局方面的应用潜力，本研究针对现有基于人工智能聊天机器人的相关研究开展了系统综述，重点聚焦其应用场景与评估方法。本研究从PubMed/MEDLINE、EMBASE、PsycINFO、CINAHL、IEEE、ACM Digital Library、Scopus及Web of Science共8个数据库中检索收集到348篇文献，其中20篇经分析后纳入本次综述。本综述明确了基于人工智能聊天机器人的四大核心研究方向：文本质量、临床疗效、用户参与度与安全性。同时，本综述也指出了当前研究存在的显著短板：随机对照试验（Randomised Controlled Trials，简称RCTs）的严重匮乏，以及在评估其性能时理论框架的应用极为有限。本研究最后探讨了未来的研究方向与可行的实践解决方案。