Dr. Answer AI for Prostate Cancer: Predicting Biochemical Recurrence Following Radical Prostatectomy

• Published in: Technology in cancer research & treatment Vol. 20; p. 15330338211024660
• Main Authors: Park, Jihwan, Rho, Mi Jung, Moon, Hyong Woo, Kim, Jaewon, Lee, Chanjung, Kim, Dongbum, Kim, Choung-Soo, Jeon, Seong Soo, Kang, Minyong, Lee, Ji Youl
• Format: Journal Article
• Language: English
• Published: Los Angeles, CA SAGE Publications 2021; Sage Publications Ltd; SAGE Publishing
• Subjects: Adult; Aged; Aged, 80 and over; Area Under Curve; Artificial Intelligence; Cancer surgery; Decision Trees; Humans; Male; Middle Aged; Models, Biological; Neoplasm Recurrence, Local - blood; Original; Patients; Prediction models; Prostate cancer; Prostate-Specific Antigen - blood; Prostatectomy; Prostatic Neoplasms - blood; Prostatic Neoplasms - pathology; Prostatic Neoplasms - surgery; ROC Curve; Software; Urological surgery; Doctor’s Answer; prostate cancer; radical prostatectomy; PROMISE CLIP registry; biochemical recurrence; random forest; artificial intelligence
• ISSN: 1533-0346; 1533-0338
• DOI: 10.1177/15330338211024660

Objectives: To develop a model to predict biochemical recurrence (BCR) after radical prostatectomy (RP), using artificial intelligence (AI) techniques. Patients and Methods: This study collected data from 7,128 patients with prostate cancer (PCa) who received RP at 3 tertiary hospitals. After preprocessing, we used the data of 6,755 cases to generate the BCR prediction model. There were 16 input variables with BCR as the outcome variable. We used a random forest to develop the model. Several sampling techniques were used to address class imbalances. Results: We achieved good performance using a random forest with synthetic minority oversampling technique (SMOTE) using Tomek links, edited nearest neighbors (ENN), and random oversampling: accuracy = 96.59%, recall = 95.49%, precision = 97.66%, F1 score = 96.59%, and ROC AUC = 98.83%. Conclusion: We developed a BCR prediction model for RP. The Dr. Answer AI project, which was developed based on our BCR prediction model, helps physicians and patients to make treatment decisions in the clinical follow-up process as a clinical decision support system.