Abstract
BACKGROUND: Prostate cancer poses a significant risk to men's health. In this study, a model for differentiating benign and malignant nodules in the central region of the prostate was constructed by combining multi-parametric MRI and hematological lab values. METHODS: This retrospective study analyzed the data acquired from Lianyungang First People's Hospital and The Second Affiliated Hospital of Hainan Medical College from January 2018 to December 2021. We included 310 MRI-confirmed prostatic nodule patients. The data were split into a training set (260 cases) and an external validation set (50 cases), with the latter exclusively from The Second Affiliated Hospital of Hainan Medical College to test the model's generalizability. Univariate and multivariate logistic regression identified critical measurements for differentiating prostate cancer (PCa) from benign prostatic hyperplasia (BPH), which were then integrated into a nomogram model. RESULTS: The key indicators determined by multivariate logistic regression analysis included apparent diffusion coefficient (ADC), standard deviation (StDev), neutrophil to lymphocyte ratio (NLR), and prostate specific antigen (PSA). The nomogram's performance, as indicated by the area under the curve (AUC), was 0.844 (95% CI: 0.811-0.938) in the training set and 0.818 (95% CI: 0.644-0.980) in the external validation set. Calibration and decision curves demonstrated that the nomogram was well-calibrated and could serve as an effective tool in clinical practice. CONCLUSION: The nomogram model based on ADC, StDev, NLR and PSA may be helpful to identify PCa and BPH.