Systematic reviews of machine learning in healthcare: a literature review

ABSTRACT

Introduction

The increasing availability of data and computing power has made machine learning (ML) a viable approach to faster, more efficient healthcare delivery.

Methods

A systematic literature review (SLR) of published SLRs evaluating ML applications in healthcare settings published between1 January 2010 and 27 March 2023 was conducted.

Results

In total 220 SLRs covering 10,462 ML algorithms were reviewed. The main application of AI in medicine related to the clinical prediction and disease prognosis in oncology and neurology with the use of imaging data. Accuracy, specificity, and sensitivity were provided in 56%, 28%, and 25% SLRs respectively. Internal and external validation was reported in 53% and less than 1% of the cases respectively. The most common modeling approach was neural networks (2,454 ML algorithms), followed by support vector machine and random forest/decision trees (1,578 and 1,522 ML algorithms, respectively).

Expert opinion

The review indicated considerable reporting gaps in terms of the ML’s performance, both internal and external validation. Greater accessibility to healthcare data for developers can ensure the faster adoption of ML algorithms into clinical practice.

KEYWORDS:

• SLRs
• machine learning
• artificial intelligence
• healthcare
• healthcare data

Article highlights

• Artificial Intelligence and Machine Learning (ML) have to the potential to improve health outcomes and increase healthcare system’s efficiency.

• A systematic literature review (SLR) identified 220 published SLRs evaluating ML applications in healthcare settings covering 10,462 ML.

• The most common modeling approach was neural networks (2,454 ML algorithms), followed by support vector machine and random forest/decision trees (1,578 and 1,522 ML algorithms, respectively).

• Internal validation was reported in 53% of the ML algorithms and external validation in less than 1% of cases. The lack of assessment of the AI performance should be overcome to facilitate the application of AI/ML in healthcare.

Declaration of interests

This paper was not funded. The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Authors contribution

KK and SP participated in the design and execution of the SRL and oversaw studies selection and the synthesis of the results obtained; they also finalized the discussion and the conclusions. JEP participated in the design of the SRL and actively contributed to the selection of studies and data extraction. BA, MHV, KJK contributed data extraction as well as to the drafting of the manuscript.

All authors read and approved the final version of the manuscript for publication.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14737167.2023.2279107

Additional information

Funding

This paper was not funded.