Abstract
Objectives: The exponential increase in published articles makes a thorough and
expedient review of literature increasingly challenging. Artificial Intelligent (AI)
platforms have been developed to attempt to address this problem, with them
progressively being incorporated into practice. This review delineated the common automated tools and platforms that employ AI approaches and evaluated the
reported benefits and challenges in using such methods.
Methods: A search was
conducted in 4 databases (Medline, Embase, Cochrane database of systematic reviews, and Epistemonikos) up to April 2021 for systematic reviews and other
related reviews implementing AI methods. To be included, an AI-assisted review
must use any form of AI method, including machine learning, deep learning, neural
network, or any other applications that are used to enable the full or semiautonomous performance of one or more stages in the development of evidence
synthesis.
Results: From a total of 4911 records identified, 87 articles underwent
full-text screening, and 12 reviews were included. These reviews used nine
different tools to implement 15 different AI methods. Eleven methods were used in
the screening stages of the review (73%). The rest were divided: two in data
extraction (13%) and two in risk of bias assessment (13%). The ambiguous benefits
of the data extractions, combined with the reported advantages from 10 reviews,
indicate that AI platforms have taken hold with varying success in evidence synthesis. The results are qualified by the reliance on the self-reporting of the review
authors.
Conclusions: Extensive human validation still appears required at this
stage in the implementation of AI/ML methods, though further evaluation is
required to define the overall contribution of such platforms in enhancing efficiency and quality in evidence synthesis.
expedient review of literature increasingly challenging. Artificial Intelligent (AI)
platforms have been developed to attempt to address this problem, with them
progressively being incorporated into practice. This review delineated the common automated tools and platforms that employ AI approaches and evaluated the
reported benefits and challenges in using such methods.
Methods: A search was
conducted in 4 databases (Medline, Embase, Cochrane database of systematic reviews, and Epistemonikos) up to April 2021 for systematic reviews and other
related reviews implementing AI methods. To be included, an AI-assisted review
must use any form of AI method, including machine learning, deep learning, neural
network, or any other applications that are used to enable the full or semiautonomous performance of one or more stages in the development of evidence
synthesis.
Results: From a total of 4911 records identified, 87 articles underwent
full-text screening, and 12 reviews were included. These reviews used nine
different tools to implement 15 different AI methods. Eleven methods were used in
the screening stages of the review (73%). The rest were divided: two in data
extraction (13%) and two in risk of bias assessment (13%). The ambiguous benefits
of the data extractions, combined with the reported advantages from 10 reviews,
indicate that AI platforms have taken hold with varying success in evidence synthesis. The results are qualified by the reliance on the self-reporting of the review
authors.
Conclusions: Extensive human validation still appears required at this
stage in the implementation of AI/ML methods, though further evaluation is
required to define the overall contribution of such platforms in enhancing efficiency and quality in evidence synthesis.
Original language | English |
---|---|
Pages (from-to) | S517 |
Number of pages | 1 |
Journal | Value in Health |
Volume | 25 |
Issue number | 7 |
Early online date | 25 Jun 2022 |
DOIs | |
Publication status | Published - 1 Jul 2022 |
Event | ISPOR 2022: The Future of HEOR in Patient-Driven Digital Healthcare Systems - Washington DC , United States Duration: 10 May 2022 → 18 May 2022 https://www.ispor.org/conferences-education/conferences/past-conferences/ispor-2022 |