Objective:

To collect multicenter clinical studies of artificial intelligence (AI)-based multiparametric magnetic resonance imaging (mpMRI) in the diagnosis of prostate cancer and to analyze the diagnostic value of AI mpMRI for prostate cancer.

Methods:

PubMed

Chinese National Knowledge Infrastructure (CNKI)

Wanfang data

and China Biology Medicine disc (CBMdisc) were systematically searched. The mpMRI

prostate cancer

artificial intelligence

diagnosis

and multicenter studies were used as entry criteria to collect

relevant literature for review. The findings of the papers were analyzed in terms of study protocol design

patient population

AI-assisted methods

and diagnostic efficacy.

Results:

A total of 6 eligible papers were collected

and the first was published in 2018. In all 6 studies

the researchers invited multiple radiologists to interpretate the mpMRI images with the help of AI. Most studies concluded that there was an improvement in radiologists diagnostic efficacy with AI assistance. However

there was some variation in the study protocols reported in different literature

and some of the study designs were flawed.

Conclusion:

There are still few multicenter clinical studies of AI applied to prostate mpMRI

and the findings are inconsistent. Standardized clinical studies should be conducted in the future to further validate the value of AI in the diagnosis of mpMRI prostate cancer.