Quantitative analysis of the intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) parameters in the diagnosis of prostate cancer

To evaluate the diagnostic value of quantitative parameters of intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) in prostate cancer (PCa) by classifying prostate tissues into healthy, benign prostatic hyperplasia (BPH), and malignant as well as compare the parameters with Prostate Imaging-Reporting and Data System version 2.1 (PI-RADS v2.1) scoring.

The study cohort comprised of an MRI dataset of 32 patients suspected of having PCa. The IVIM-DWI parameters, true diffusion coefficient (D), pseudo-diffusion coefficient (D*), perfusion fraction (f), and apparent diffusion coefficient (ADC) of tissue regions of interest were determined and compared within different categories. Diagnostic performance of quantitative parameters and their combinations were evaluated using machine learning methods.

The parameters D, f, and ADC values in the malignant tissues group were substantially lower (p < 0.05) compared to the BPH and healthy group, however D* value was higher in malignant tissues. The parameters D, f, and ADC values ​​of malignant tissues were found to be negatively correlated with the PI-RADS scores, but the D* was positively correlated. The diagnostic efficiency of parameters ADC was higher (accuracy = 67.80 ± 1.10% and area under receiver operating characteristic curve (AU-ROC) = 0.79) than other parameters. The combination of all parameters D, D*, f and ADC were showed the higher diagnostic accuracy (accuracy = 83.00 ± 2.50% and AU-ROC = 0.88).

The incorporation of quantitative parameters derived from IVIM-DWI has the potential to enhance the accuracy of prostate cancer detection.