No abstract

Purpose This study aimed to develop deep learning (DL) models based on multicentre biparametric magnetic resonance imaging (bpMRI) for the diagnosis of clinically significant prostate cancer (csPCa) and compare the performance of these models with that of the Prostate Imaging and Reporting and Data System (PI-RADS) assessment by expert radiologists based on multiparametric MRI (mpMRI). Methods We included 1861 consecutive male patients who underwent radical prostatectomy or biopsy at seven hospitals with mpMRI. These patients were divided into the training (1216 patients in three hospitals) and external validation cohorts (645 patients in four hospitals). PI-RADS assessment was performed by expert radiologists. We developed DL models for the classification between benign and malignant lesions (DL-BM) and that between csPCa and non-csPCa (DL-CS). An integrated model combining PI-RADS and the DL-CS model, abbreviated as PIDL-CS, was developed. The performances of the DL models and PIDL-CS were compared with that of PI-RADS. Results In each external validation cohort, the area under the receiver operating characteristic curve (AUC) values of the DL-BM and DL-CS models were not significantly different from that of PI-RADS (P > 0.05), whereas the AUC of PIDL-CS was superior to that of PI-RADS (P < 0.05), except for one external validation cohort (P > 0.05). The specificity of PIDL-CS for the detection of csPCa was much higher than that of PI-RADS (P < 0.05). Conclusion Our proposed DL models can be a potential non-invasive auxiliary tool for predicting csPCa. Furthermore, PIDL-CS greatly increased the specificity of csPCa detection compared with PI-RADS assessment by expert radiologists, greatly reducing unnecessary biopsies and helping radiologists achieve a precise diagnosis of csPCa.

The present study aimed to explore the potential of artificial intelligence (AI) methodology based on magnetic resonance (MR) images to aid in the management of prostate cancer (PCa). To this end, we reviewed and summarized the studies comparing the diagnostic and predictive performance for PCa between AI and common clinical assessment methods based on MR images and/or clinical characteristics, thereby investigating whether AI methods are generally superior to common clinical assessment methods for the diagnosis and prediction fields of PCa. First, we found that, in the included studies of the present study, AI methods were generally equal to or better than the clinical assessment methods for the risk assessment of PCa, such as risk stratification of prostate lesions and the prediction of therapeutic outcomes or PCa progression. In particular, for the diagnosis of clinically significant PCa, the AI methods achieved a higher summary receiver operator characteristic curve (SROC-AUC) than that of the clinical assessment methods (0.87 vs. 0.82). For the prediction of adverse pathology, the AI methods also achieved a higher SROC-AUC than that of the clinical assessment methods (0.86 vs. 0.75). Second, as revealed by the radiomics quality score (RQS), the studies included in the present study presented a relatively high total average RQS of 15.2 (11.0–20.0). Further, the scores of the individual RQS elements implied that the AI models in these studies were constructed with relatively perfect and standard radiomics processes, but the exact generalizability and clinical practicality of the AI models should be further validated using higher levels of evidence, such as prospective studies and open-testing datasets.

Purpose: To construct deep learning (DL) models based on multicentre biparametric MRI (bpMRI) for the diagnosis of clinically significant prostate cancer (csPCa), and compare the performance of these models with that of Prostate Imaging Reporting and Data System (PI-RADS) assessment of expert-level radiologists based on multiparametric MRI (mpMRI).Methods: This study included 1861 consecutive men with mpMRI from seven hospitals, who underwent radical prostatectomy or biopsy. These patients were divided into training cohort (3 hospitals, 1216 patients) and external validation cohorts (4 hospitals, 645 patients). PI-RADS assessment was performed by expert-level radiologists. The DL models were constructed for the classifications between benign and malignant lesions (DL-BM), and between csPCa and non-csPCa (DL-CS). The integrated model combining a DL-CS model and PI-RADS (PIDL-CS) was constructed. The performance of the deep learning models and PIDL-CS were compared with those of PI-RADS.Results: In each external validation cohort, the area under receiver operating characteristic curve (AUC) values of DL-BM and DL-CS models were not significantly different from that of PI-RADS (Ps > 0.05), whereas the AUC of PIDL-CS was superior to that of PI-RADS (Ps < 0.05) except one external validation cohort (P > 0.05). The specificity of the PIDL-CS for the detection of csPCa was much more than that of PI-RADS (Ps < 0.05).Conclusion: Our proposed DL model can be a potential non-invasive auxiliary tool to predict csPCa. Further, PIDL-CS greatly increased the specificity in the detection of csPCa compared with PI-RADS assessment of expert-level radiologists, greatly reducing the unnecessary biopsies and helping radiologists achieve a precise diagnosis of csPCa.