Suppr超能文献

应对医疗保健领域人工智能实施的障碍与促进因素:一项范围综述

Navigating the barriers and facilitators to implementation of AI in healthcare : a scoping review.

作者信息

Khattak Mohammed, Bowness James S, Yonis Rusul, Kierkegaard Patrick, McGregor Alison, Perry Daniel C

机构信息

Faculty of Health and Life Science, University of Liverpool, Liverpool, UK.

Trauma & Orthopaedics Department, Alder Hey Children's Hospital, Liverpool, UK.

出版信息

Bone Joint J. 2025 Jul 1;107-B(7):666-672. doi: 10.1302/0301-620X.107B7.BJJ-2024-1570.R1.

Abstract

AIMS

There is increasing emphasis on applying AI techniques to enhance healthcare delivery and decision-making. However, despite much interest and early promise, a major challenge is translation into clinical practice. To address the challenges of AI deployment, optimize implementation, and establish strategies for effective utilization of AI technology in healthcare, we aimed to answer the question: what are the key determinants influencing effective deployment of AI technology in healthcare?

METHODS

We followed PRISMA-ScR and the Joanna Briggs Institute Methodology guidelines for scoping reviews; the research protocol was published prospectively on Open Science Framework. We searched PubMed, Cochrane, Ovid MEDLINE, Scopus, and IEEE Xplore for papers published in English from 2000, including systematic/scoping reviews and meta-analyses with full text available.

RESULTS

The initial search was limited to AI medical imaging technology. It identified 1,511 papers, of which 523 met the eligibility criteria based on title and abstract screening. A total of 488 papers were excluded due to context or irrelevant content, leaving 35 papers for full-text review. No systematic/scoping reviews specifically addressing the deployment of AI medical imaging solutions were identified, prompting the inclusion criteria to be broadened to encompass any study designs related to all relevant technology. Overall, 15 papers were included in the final scoping review.

CONCLUSION

The successful deployment of AI in healthcare is challenging, due to barriers which can be ethical, technological, regulatory, financial, or patient- and workforce-related. Facilitators to drive successful implementation include planning, organizational culture, patient involvement, stakeholder engagement, education, and leadership. Leveraging these essential barriers and facilitators provides a foundation for developing implementation strategies that streamline the deployment of AI technology in healthcare.

摘要

目的

越来越强调应用人工智能技术来改善医疗服务和决策。然而,尽管人们对此兴趣浓厚且初期前景乐观,但一个主要挑战是将其转化为临床实践。为应对人工智能部署的挑战、优化实施并制定在医疗保健中有效利用人工智能技术的策略,我们旨在回答以下问题:影响人工智能技术在医疗保健中有效部署的关键决定因素是什么?

方法

我们遵循PRISMA - ScR和乔安娜·布里格斯研究所方法指南进行范围审查;研究方案已前瞻性地发表在开放科学框架上。我们在PubMed、Cochrane、Ovid MEDLINE、Scopus和IEEE Xplore中搜索2000年以来以英文发表的论文,包括系统评价/范围审查和可获取全文的荟萃分析。

结果

最初的搜索仅限于人工智能医学成像技术。共识别出1511篇论文,其中523篇基于标题和摘要筛选符合纳入标准。由于背景或无关内容,共排除488篇论文,剩下35篇进行全文审查。未找到专门针对人工智能医学成像解决方案部署的系统评价/范围审查,因此将纳入标准扩大到涵盖与所有相关技术相关的任何研究设计。总体而言,最终范围审查纳入了15篇论文。

结论

由于存在伦理、技术、监管、财务或与患者及劳动力相关的障碍,人工智能在医疗保健中的成功部署具有挑战性。推动成功实施的促进因素包括规划、组织文化、患者参与、利益相关者参与、教育和领导力。利用这些关键障碍和促进因素为制定实施策略奠定了基础,这些策略可简化人工智能技术在医疗保健中的部署。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验