Premium
Development and validation of an explainable artificial intelligence‐based decision‐supporting tool for prostate biopsy
Author(s) -
Suh Jungyo,
Yoo Sangjun,
Park Juhyun,
Cho Sung Yong,
Cho Min Chul,
Son Hwancheol,
Jeong Hyeon
Publication year - 2020
Publication title -
bju international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.773
H-Index - 148
eISSN - 1464-410X
pISSN - 1464-4096
DOI - 10.1111/bju.15122
Subject(s) - medicine , receiver operating characteristic , confidence interval , prostate , calculator , prostate biopsy , prostate cancer , choquet integral , artificial intelligence , feature selection , urology , nuclear medicine , statistics , mathematics , computer science , cancer , operating system , fuzzy logic
Objectives To develop and validate a risk calculator for prostate cancer (PCa) and clinically significant PCa (csPCa) using explainable artificial intelligence (XAI). Patients and Methods We used data of 3791 patients to develop and validate the risk calculator. We initially divided the data into development and validation sets. An extreme gradient‐boosting algorithm was applied to the development calculator using five‐fold cross‐validation with hyperparameter tuning following feature selection in the development set. The model feature importance was determined based on the Shapley value. The area under the curve (AUC) of the receiver operating characteristic curve was analysed for each validation set of the calculator. Results Approximately 1216 (32.7%) and 562 (14.8%) patients were diagnosed with PCa and csPCa. The data of 2843 patients were used for development, whereas the data of 948 patients were used as a test set. We selected the variables for each PCa and csPCa risk calculation according to the least absolute shrinkage and selection operator regression. The AUC of the final PCa model was 0.869 (95% confidence interval [CI] 0.844–0.893), whereas that of the csPCa model was 0.945 (95% CI 0.927–0.963). The prostate‐specific antigen (PSA) level, free PSA level, age, prostate volume (both the transitional zone and total), hypoechoic lesions on ultrasonography, and testosterone level were found to be important parameters in the PCa model. The number of previous biopsies was not associated with the risk of csPCa, but was negatively associated with the risk of PCa. Conclusion We successfully developed and validated a decision‐supporting tool using XAI for calculating the probability of PCa and csPCa prior to prostate biopsy.