The goal of this observational study is to assess the chance to predict prostate cancer with the analysis of specific values found in images obtained from RMmp (multiparametric magnetic resonance imaging) of patients who went through the procedure at IRCCS Azienda Ospedaliero-Universitaria hospital in the city of Bologna from September 2015 to March 2021.

The main question it aims to answer is:

- does the analysis of ADC values and radiomics values found in the RMmp images help with the detection of prostate cancer?

The radiologist involved in this observational study will enroll the participants, collect their informed consent and radiological data.

The urologist will collect the clinical data while the Anatomo-pathologist will take care of the hystological data.

The ADC values of the suspected lesions and healthy parenchyma will be extracted from the MR exams. For the evaluation of radiometric characteristics, images will first be de-identified, the downloaded and processed by the hospital's medical physicists.

The engineers at the University of Bologna, led by medical investigators, will extract the radiomic characteristics All data will be anonymised and collected in a database shared between the collaborating operational units.

Inclusion Criteria:

• Presence of one or more lesions classified as PI-RADSv2.1 ≥ 3 at a prostate RMmp at the IRCCS Azienda Ospedaliero-Universitaria in Bologna
• Indication for eco-guided targeted biopsy by fusion technique integrated with systematic biopsy
• Indication for radical prostatectomy
• Participants aged 18 at the time of examination
• Obtaining informed consent.

Exclusion Criteria:

• Impossibility of performing MRI (claustrophobic or obese participants etc.) or contraindications to the execution of MRI
• Technically sub-optimal investigations for the presence of artifacts (hip prosthesis, movement of the endorectal probe, etc.)

The study will include all participants who have performed a fusion biopsy with RMmp of the prostate in the period between the beginning of use of this technique (September 2015) and March 2021 whose images are available on RIS-PACS of the IRCCS University Hospital of Bologna will be included in the study . The sample size considered sufficient is 120 patients, of which 90 will be used for method development and 30 will be used as validation group. The latter will be used to evaluate how generalizable the results of the study are to the population, to avoid overestimating the model's capabilities in evaluating it on the same patients on which it was developed.