Artificial intelligence-aided detection for prostate cancer with multimodal routine health check-up data - an Asian multi-center study

Sep 24, 2026

Early detection of high-grade prostate cancer (HGPCa) is critical, yet existing strategies often cause high negative biopsy rates and substantial medical costs. This study develops an Asian Prostate Cancer Artificial intelligence (APCA) score using routine health check-up data, adding no extra cost beyond standard evaluations, to predict HGPCa risk. A total of 7,476 patients underwent biopsy across eight Asian referral centers between 2008 and 2021.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Background

Patients and methods

Results

Conclusions

Keywords

’Diagnostic Study Artiﬁcial intelligence-aided detection for prostate cancer with multimodal routine health check-up data: an Asian multi-center study Zijian Song, MDa,b , Wei Zhang, MDa , Qingchao Jiang, PhDc,d , Longxin Deng, MDa , Le Du, BAc,d , Weiming Mou, MDa,e , Yancheng Lai, MDa , Wenhui Zhang, MDa , Yang Yang, MDf, Jasmine Lim, MDg , Kang Liu, MDh , Jae Young Park, MDi , Chi-Fai Ng, MDh , Teng Aik Ong, MDg , Qiang Wei, MDj , Lei Li, MDk , Xuedong Wei, MDl , Ming Chen, MDm , Zhixing Cao, PhDc,d,* , Fubo Wang, MDn,o,p,* , Rui Chen, MDa,b,* Background: The early detection of high-grade prostate cancer (HGPCa) is of great importance. However, the current detection strategies result in a high rate of negative biopsies and high medical costs. In this study, the authors aimed to establish an Asian Prostate Cancer Artiﬁcial intelligence (APCA) score with no extra cost other than routine health check-ups to predict the risk of HGPCa. Patients and methods: A total of 7476 patients with routine health check-up data who underwent prostate biopsies from January 2008 to December 2021 in eight referral centres in Asia were screened. After data pre-processing and cleaning, 5037 patients and 117 features were analyzed. Seven AI-based algorithms were tested for feature selection and seven AI-based algorithms were tested for classiﬁcation, with the best combination applied for model construction. The APAC score was established in the CH cohort and validated in a multi-centre cohort and in each validation cohort to evaluate its generalizability in different Asian regions. The performance of the models was evaluated using area under the receiver operating characteristic curve (ROC), calibration plot, and decision curve analyses. Results: Eighteen features were involved in the APCA score predicting HGPCa, with some of these markers not previously used in prostate cancer diagnosis. The area under the curve (AUC) was 0.76 (95% CI:0.74–0.78) in the multi-centre validation cohort and the increment of AUC (APCA vs. PSA) was 0 . 16 (95% CI:0 .13–0.20) . The calibration plots yielded a high degree of coherence and the decision curve analysis yielded a higher net clinical beneﬁt. Applying the APCA score could reduce unnecessary biopsies by 20 .2% and 38 .4%, at the risk of missing 5 .0% and 10 .0% of HGPCa cases in the multi-centre validation cohort, respectively. Conclusions: The APCA score based on routine health check-ups could reduce unnecessary prostate biopsies without additional examinations in Asian populations. Further prospective population-based studies are warranted to conﬁrm these results. Keywords: artiﬁcial intelligence, diagnosis, prostate biopsy, Prostate cancer, risk prediction aDepartment of Urology, Shanghai Changhai Hospital, Second Military Medical University, bDepartment of Urology, Renji Hospital, Shanghai Jiao Tong University School of Medicine, cKey Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, dState Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai, eDepartment of Urology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, fDepartment of Clinical Laboratory, Nanjing Jinling Hospital, Nanjing University School of Medicine, gDepartment of Urology, University of Malaya Medical Centre, Kuala Lumpur, Malaysia, hSH Ho Urology Centre, Department of Surgery, The Chinese University of Hong Kong, Hong Kong SAR, China, iDepartment of Urology, Korea University Ansan Hospital, Soule, Korea, jDepartment of Urology, Institute of Urology, West China Hospital, Sichuan University, Chengdu, Sichuan, kDepartment of Urology, The First Afﬁliated Hospital of Xi’an Jiaotong University, Xi’an Shaanxi, lDepartment of Urology, The First Afﬁliated Hospital of Soochow University, Suzhou, mDepartment of Urology, Zhongda Hospital, Southeast University, Nanjing, nSchool of Life Sciences, Guangxi Medical University, Nanning, Gu

FAQs

What problem does the study address in prostate cancer detection?

The study targets the difficulty of early detection of high-grade prostate cancer, where current approaches can lead to many negative biopsies and high medical costs.

How was the APCA score developed and validated?

The authors screened 7,476 patients from eight Asian referral centers, analyzed 117 features after preprocessing, built the APCA score in a CH cohort, and validated it in a multi-centre cohort and each validation cohort.

What does the study report about model performance and clinical impact?

In multi-centre validation, the APCA score achieved an AUC of 0.76, with an AUC increase versus PSA of 0.16, improved calibration, and higher net clinical benefit; applying APCA could reduce unnecessary biopsies while risking missing some HGPCa cases.

Sep 24, 2026

Uploaded by Guten tag