Aims and objectives
To evaluate radiomics features extracted from T2-weighted,
diffusion weighted imaging (DWI),
diffusion kurtosis imaging (DKI),
Toft model (TM),
and shutter-speed model (SSM) perfusion maps among prostate cancer (PCa),
benign prostatic hyperplasia (BPH),
and benign peripheral zone (PZ).
To compare the diagnostic performance of advanced prostate radiomics to PI-RADS v2 classification.
Methods and materials
40 foci of PCa,
48 BPH nodules,
and 36 benign PZ from 40 patients who underwent multiparametric MRI of prostate,
to eventually address a target-biopsy,
MRI exam was performed without endorectal coil with a 3T mMR Biograph scanner.
DWI was performed using 7 b-values (0-2500 s/mm 2 ): classical apparent diffusion coefficient (ADC) map was generated using b values up to 1500 s/mm 2 ; the entire range of b values was used to compute non-Gaussian diffusion coefficient (D) and...
Identified radiomic features differentiated PCa from benign PZ.
Prediction performances were higher for diffusion features (extracted for both DWI and DKI - Fig1) than for perfusion ones (extracted for both the TM and the SSM - Fig2).
These differences were confirmed independently by the number of features included in the logistic model (i.e.,
Intermodal approach lead to logistic regression models with very high discimination performances (AUC values close to 1) with best...
Among perfusion features,
parameters extracted by SSM have higher predictive performances than TM-derived. A classifier that includes features extracted from DKI significantly improves the accuracy in cancer detection.
- Monti S.,
DCE-MRI pharamacokinetic-based phenotyping of invasive ductal carcinoma: a radiomic study for prediction of histologic outcomes.
Contrast Media Mol Imaging (2018). - Incoronato M.,
Radiogenomic analysis of oncological data: a technical survey.
Int J Mol Sci.