Suppr超能文献

人工智能在细菌感染控制中的应用:一项范围综述

Artificial Intelligence in Bacterial Infections Control: A Scoping Review.

作者信息

Abu-El-Ruz Rasha, AbuHaweeleh Mohannad Natheef, Hamdan Ahmad, Rajha Humam Emad, Sarah Jood Mudar, Barakat Kaoutar, Zughaier Susu M

机构信息

College of Health Sciences, QU Health, Qatar University, Doha P.O. Box 2713, Qatar.

College of Medicine, QU Health, Qatar University, Doha P.O. Box 2713, Qatar.

出版信息

Antibiotics (Basel). 2025 Mar 2;14(3):256. doi: 10.3390/antibiotics14030256.

Abstract

: Artificial intelligence has made significant strides in healthcare, contributing to diagnosing, treating, monitoring, preventing, and testing various diseases. Despite its broad adoption, clinical consensus on AI's role in infection control remains uncertain. This scoping review aims to understand the characteristics of AI applications in bacterial infection control. : This review examines the characteristics of AI applications in bacterial infection control, analyzing 54 eligible studies across 5 thematic scopes. The search from 3 databases yielded a total of 1165 articles, only 54 articles met the eligibility criteria and were extracted and analyzed. Five thematic scopes were synthesized from the extracted data; countries, aim, type of AI, advantages, and limitations of AI applications in bacterial infection prevention and control. The majority of articles were reported from high-income countries, mainly by the USA. The most common aims are pathogen identification and infection risk assessment. The most common AI used in infection control is machine learning. The commonest reported advantage is predictive modeling and risk assessment, and the commonest disadvantage is generalizability of the models. : This scoping review was developed according to Arksey and O'Malley frameworks. A comprehensive search across PubMed, Embase, and Web of Science was conducted using broad search terms, with no restrictions. Publications focusing on AI in infection control and prevention were included. Citations were managed via EndNote, with initial title and abstract screening by two authors. Data underwent comprehensive narrative mapping and categorization, followed by the construction of thematic scopes. Artificial intelligence applications in infection control need to be strengthened for low-income countries. More efforts should be dedicated to investing in models that have proven their effectiveness in infection control, to maximize their utilization and tackle challenges.

摘要

人工智能在医疗保健领域取得了重大进展,有助于诊断、治疗、监测、预防和检测各种疾病。尽管其被广泛采用,但关于人工智能在感染控制中的作用的临床共识仍不明确。本范围综述旨在了解人工智能在细菌感染控制中的应用特点。

本综述研究了人工智能在细菌感染控制中的应用特点,分析了5个主题范围内的54项符合条件的研究。从3个数据库检索共获得1165篇文章,只有54篇文章符合纳入标准并被提取和分析。从提取的数据中综合出5个主题范围:国家、目的、人工智能类型、人工智能在细菌感染预防和控制中的应用优势及局限性。大多数文章来自高收入国家,主要是美国。最常见的目的是病原体识别和感染风险评估。感染控制中最常用的人工智能是机器学习。最常报道的优势是预测建模和风险评估,最常见的劣势是模型的通用性。

本范围综述是根据阿克西和奥马利框架制定的。使用宽泛的检索词对PubMed、Embase和科学网进行了全面检索,没有限制。纳入了关注人工智能在感染控制和预防方面的出版物。通过EndNote管理引文,由两位作者进行初步的标题和摘要筛选。数据经过全面的叙述性映射和分类,随后构建主题范围。低收入国家在感染控制方面的人工智能应用需要加强。应更加努力投资于已证明在感染控制中有效的模型,以最大限度地利用它们并应对挑战。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/194d/11939793/e89cbaaceff5/antibiotics-14-00256-g001.jpg

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