Machine Learning Approaches for the Prediction of Prostate Cancer according to Age and the Prostate-Specific Antigen Level
이재근, 양승우, 이승희, 현윤경, 김진범, Jin Long, 이지용, 박종목, 하태영, 신주현, 임재성, 나용길, 송기학,
소속 상세정보
이재근 ( Lee Jae-Geun )
Chungnam National University School of Medicine Department of Urology
양승우 ( Yang Seung-Woo )
Chungnam National University College of Medicine Department of Urology
이승희 ( Lee Seung-Hee )
National Institute for Mathematical Sciences Division of Medical Mathematics
현윤경 ( Hyon Yun-Kyong )
National Institute for Mathematical Sciences Division of Medical Mathematics
김진범 ( Kim Jin-Bum )
Konyang University College of Medicine Department of Urology
( Jin Long )
Chungnam National University College of Medicine Department of Urology
이지용 ( Lee Ji-Yong )
Chungnam National University College of Medicine Department of Urology
박종목 ( Park Jong-Mok )
Chungnam National University College of Medicine Department of Urology
하태영 ( Ha Tae-Young )
National Institute for Mathematical Sciences Division of Medical Mathematics
신주현 ( Shin Ju-Hyun )
Chungnam National University College of Medicine Department of Urology
임재성 ( Lim Jae-Sung )
Chungnam National University College of Medicine Department of Urology
나용길 ( Na Yong-Gil )
Chungnam National University College of Medicine Department of Urology
송기학 ( Song Ki-Hak )
Chungnam National University College of Medicine Department of Urology
Abstract
Purpose: The aim of this study was to evaluate the applicability of machine learning methods that combine data on age and prostate-specific antigen (PSA) levels for predicting prostate cancer.
Materials and Methods: We analyzed 943 patients who underwent transrectal ultrasonography (TRUS)-guided prostate biopsy at Chungnam National University Hospital between 2014 and 2018 because of elevated PSA levels and/or abnormal digital rectal examination and/or TRUS findings. We retrospectively reviewed the patients’ medical records, analyzed the prediction rate of prostate cancer, and identified 20 feature importances that could be compared with biopsy results using 5 different algorithms, viz., logistic regression (LR), support vector machine, random forest (RF), extreme gradient boosting, and light gradient boosting machine.
Results: Overall, the cancer detection rate was 41.8%. In patients younger than 75 years and with a PSA level less than 20 ng/mL, the best prediction model for prostate cancer detection was RF among the machine learning methods based on LR analysis. The PSA density was the highest scored feature importances in the same patient group.
Conclusions: These results suggest that the prediction rate of prostate cancer using machine learning methods not inferior to that using LR and that these methods may increase the detection rate for prostate cancer and reduce unnecessary prostate biopsy, as they take into consideration feature importances affecting the prediction rate for prostate cancer.
키워드
Prediction; Prostate cancer; Machine learning; Prostate biopsy
원문 및 링크아웃 정보
등재저널 정보