It is deeply meaningful that Deep Bio introduces the world's first clearance of AI technology that aids grading the severity of prostate cancer in pathological tissue diagnosis.
SEOUL, South Korea (PRWEB) November 24, 2021
Deep Bio, a pioneer in medical AI specialized for cancer diagnostics, announced that it has received Korea MFDS (Ministry of Food and Drug Safety) approval for DeepDx®-Prostate Pro, its deep learning-based software that assists the histopathological diagnosis of prostate cancer as a Class III medical device.
This marks the world’s first regulatory clearance of an AI-powered medical device that analyzes digitized slide images of prostate specimens to classify the severity of prostate cancer based on the Gleason grading system, which is the most widely used system to grade prostate cancer. This milestone is Deep Bio’s second Korean MFDS regulatory approval, following the clearance of a model for prostate cancer detection last year. DeepDx®-Prostate Pro analyzes WSIs (Whole Slide Images) of H&E (Hematoxylin & Eosin) stained prostate core needle biopsies and automatically classifies the histological severity of prostate cancer. Analysis results automatically generate Gleason grades and Gleason scores based on the Gleason grading system. If a tissue is not categorized by the Gleason grading system, the result will show as ‘No Grade.’
DeepDx®-Prostate Pro, the world's only algorithm that supports severity analysis of prostate cancer, demonstrated robust performance in evaluation studies, showing 98.7% concordance in grade group classification and 96.9% concordance in the “No grade” group classification compared with reference standard created by three pathologists. DeepDx®-Prostate Pro also demonstrated high agreement with the pathologists. When compared with the analysis of three pathologists, the software analysis recorded a kappa score of 0.713 and a quadratic-weighted kappa score of 0.922. In addition, the software improved workflows by greatly reducing time spent on analyses. While it took 550 minutes for a pathologist to analyze the WSIs of the entire biopsy samples (593 cases) using the current standard of care, the same pathologist completed analyses of the same cases in 364 minutes with the software, marking 33.8% decrease in diagnostic turnaround time.
“It is deeply meaningful that Deep Bio introduces the world's first clearance of AI technology that aids grading the severity of prostate cancer in pathological tissue diagnosis,” stated Sun Woo Kim, the CEO of Deep Bio. “Although Gleason Grade results are critical for patient follow-up treatment and care, the diagnostic process suffers from high discordance between the pathologists. We are confident that through the use of our AI support software, pathologists can make their diagnoses faster and in a more consistent manner,” added Kim.
Tae Yeoung Kwak, the CTO of Deep Bio said “Deep Bio’s AI technology for analyzing prostate cancer has gained a global reputation following presentation in international conferences and academic journals. In addition to a model for prostate needle biopsies, Deep Bio is currently developing diagnostic algorithms to analyze other prostate tissue types obtained by transurethral prostatectomy (TURP) and prostatectomy, as well as pathologic tissue analysis technology for various cancer types such as breast cancer and bladder cancer. Deep Bio will make every effort to commercialize these technologies in the near future to improve existing diagnostic workflows.”
About Deep Bio
Deep Bio Inc. is an AI biotech company with in-house expertise in deep learning, pathology, life sciences, and pharmacotherapeutics. As the country’s first to obtain Korea’s MFDS (Ministry of Food and Drug Safety) approval of an AI-based cancer diagnostic support solution, Deep Bio envisions a suite of AI-based IVD SaMDs (In Vitro Diagnostics Software as a Medical Device) for diagnosis and prognosis of multiple cancers. Deep Bio is actively engaged in the research space and participating in ongoing collaborations with top US medical centers. To learn more, visit http://www.deepbio.co.kr.