Deep Bio’s AI Algorithm Helps Risk Stratification for Prostate Cancer

Pioneer in medical AI for digital pathology and cancer diagnosis support software, Deep Bio has announced results of collaborative research with Johns Hopkins Medicine. 

The joint research compared human pathologists and an AI algorithm in grading prostate biopsy specimens to predict biochemical recurrence after radical prostatectomy. The study reassessed the diagnoses of 284 patients initially diagnosed as Gleason grade group 2, one of the most challenging cohorts to diagnose, who underwent radical prostatectomy at Johns Hopkins Medicine from 2000 to 2014. Approximately 16% of patients went through recurrence of prostate cancer, which can be tracked through almost 14 years of follow-up data. In order to compare DeepBio’s algorithm to human pathologist, biopsies were re-graded by two expert genitourinary pathologists, with a third expert to facilitate tie-breaks, as well as DeepBio’s software DeepDx® Prostate. 

Results showed poor agreement between the two pathologists (kappa 0.17), while grading between the consensus pathology read and the AI algorithm had a slightly higher agreement (kappa 0.33). 

The study indicated that the algorithm can act as a tool that stratifies patients for subsequent biochemical recurrence after radical prostatectomy. This is significant as it can potentially prevent unnecessary surgeries and aid in choosing between treatment modalities. 

“It is a meaningful milestone for Deep Bio to conduct a joint research with Johns Hopkins Medicine, one of the top global universities, and the findings are remarkable for prostate cancer diagnosis using AI,” said Sun-Woo Kim, CEO of Deep Bio. “As precise diagnosis and personalization have become the crucial factors in deciding medical decisions for medical professionals, we will focus on bringing innovative deep learning technologies that can unlock new insights in cancer diagnosis and treatment” added he.