An artificial intelligence model has been recently developed by Gangnam Severance to find the best method of treatment for prostate cancer.
The research team led by professor Koo Kyo-Cheol, Lee Kwang-Seok, and Chung Bunh-ha of the urology department at Gangnam Severance Hospital and the artificial intelligence company Cellavas developed a personalized survival prediction and optimal treatment model through AI analysis.
AI prediction model learns and analyzes clinical and pathological data of 7,267 prostate cancer patients to predict treatment response and survival rate.
According to the research team, " the accuracy of prediction is up to 91% which is higher than that of models using statistical analysis, which is about 80%," in addition "data can be collected that reflects the stage-by-step pathogenesis of the diagnosis and treatment process to analyze the effects of the condition on the progress of cancer." The explaination is tha this can predict the progress of the disease and provide optimal treatment at the time of diagnosis of prostate cancer.
Professor Koo Kyo-Cheol said, "While the existing prediction model did not reflect the time series, the newly developed model can analyze the entire progress of the disease and find the best method of treatment at the time of diagnosis, and also. "Based on this, Gangnam Severance have established an online platform customized treatment selection system that provides the most appropriate method of treatment depending on the individual characteristics of each patient."
This platform is being used to consult patients and establish treatment policies at the prostate cancer centre of Gangnam Severance Hospital.
According to professor Chung Byung-Ha, " the current standard treatment guildeline are based on Western clinical indicators, and also said " this study reflected the reality of Korea because it developed algorithms based on 30 years of treatment data on Korean patients."
The study was conducted with support from the Korean Research Foundation of Science and ICT, and the result were published in the recent issue of the World Journal of Urology.