JMIR Publications

ABSTRACT

Background:

Artificial intelligence (AI) has been extensively used in a range of medical fields to promote therapeutic development. The development of diverse AI techniques has also contributed to early detections, disease diagnoses and referral management. This arises along with the concerns about the value of advanced AI in disease diagnosis from healthcare professionals, medical service providers as well as health policy decision-makers.

Objective:

This review aimed to systematically examine the literature, in particular, on the performance comparison between advanced AI and human clinicians, and therefore provide an up-to-date summary with respect to what extent AI has been applied to disease diagnoses. By doing so, this review discussed the relationship between the current advanced AI development and clinicians as to disease diagnosis and thus therapeutic development in the long run.

Methods:

We systematically searched articles in Scopus, PubMed, CINAHL, Web of Science and the Cochrane Library published between January 2000 and March 2019, following the Preferred Reporting Items for Systematic reviews and Meta-Analysis. According to preset inclusion and exclusion criteria, only articles compared the medical performance between advanced AI and human experts were considered.

Results:

Nine articles were identified. A convolutional neural network was the commonly-applied advanced AI technology. Due to variation in medical fields, there is a distinction between individual studies in terms of AI’s classification/labeling, training process, dataset size and algorithm validation. Performance indices reported in articles included diagnostic accuracy, weighted errors, false positive rate, sensitivity and specificity (and/or the area under the receiver operating characteristic curve). The results showed that the performance of AI was on par with that of clinicians and exceeded those clinicians with less experience.

Conclusions:

Current AI development has diagnostic performance comparable with medical experts, especially in image-recognition related fields. Further studies can be extended to other types of medical imaging such as MRI and other image-unrelated medical practices. With the continued development of AI-assisted technologies, the clinical implications underpinned by clinicians’ experience and guided by patient-centered healthcare principle should be constantly considered in future AI-related and other technology-based medical research. Clinical Trial: NA