Download This Paper
Preprints with The Lancet is a collaboration between The Lancet Group of journals and SSRN to facilitate the open sharing of preprints for early engagement, community comment, and collaboration. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early-stage research papers that have not been peer-reviewed. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. The findings should not be used for clinical or public health decision-making or presented without highlighting these facts. For more information, please see the FAQs.
Artificial Intelligence-Aided Detection for Prostate Cancer with Multi-Modal Routine Health Check-Up Data: An Asian Multi-Center Study
39 Pages Posted: 26 Jul 2023
More...Abstract
Background: The early detection of high-grade prostate cancer (HGPCa) is of great importance. However, current detection strategies cause a high rate of negative biopsies and high medical costs. In this study, we aimed to establish the Asian Prostate Cancer AI (APCA) score with no extra cost other than routine health check-ups (including blood routine tests, blood biochemistry tests, urine routine tests, abdominal ultrasounds, etc.) to predict the risk of HGPCa.
Methods: A total of 7476 patients who underwent prostate biopsy at eight hospitals in Asia were screened. After data pre-processing and cleaning, 5037 patients and 117 features were analyzed. A total of 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 performance of the models was evaluated with the area under the receiver operating characteristic curve (AUC), calibration plot, and decision curve analyses.
Findings: A total of 18 features were involved in the APCA score predicting HGPCa, with the AUC of 0·93, 0·83, 0·77, 0·75, 0·88, 0.72, 0·72 and 0·76 the internal cohort and 7 external cohorts being, respectively. The increment of AUC (APCA vs. PSA) were 0·16 in the multi-center validation cohort and 0·18, 0·17, 0·21, 0·36, 0·16, 0·08, 0·11 in the seven independent validation cohorts. The calibration plots yield a high degree of coherence, and the decision curve analysis yield higher net clinical benefit. Applying the APCA score could reduce 38·4% unnecessary biopsies at the cost of missing 10·0% HGPCa in the multi-center validation cohort.
Interpretation: The APCA score based on routine health check-ups could reduce unnecessary prostate biopsy at low cost in Asian populations. Future prospective population-based studies are warranted to confirm the results of this study.
Trial Registration: Registered on the Chinese Clinical Trial Registry (http://www.chictr.org.cn, ChiCTR2100048428).
Funding: This study is supported by the National Natural Science Foundation of China (82272905), the Rising-Star Program of the Science and Technology Commission of Shanghai Municipality (21QA1411500), and the Shanghai Action Plan for Technological Innovation Grant (No. 22ZR1478000, 22ZR1415300, 22511104000, 23S41900500).
Declaration of Interest: The authors declare no conflicts of interest.
Ethical Approval: The protocol was approved by the Institutional Ethics Committee of Shanghai Changhai Hospital (CHEC2020-157),
Keywords: Prostate cancer, diagnosis, prostate biopsy, artificial intelligence, risk prediction
Suggested Citation: Suggested Citation