Deep learning algorithm performs similarly to radiologists in the assessment of prostate volume on MRI

Abstract

Objectives

Prostate volume (PV) in combination with prostate specific antigen (PSA) yields PSA density which is an increasingly important biomarker. Calculating PV from MRI is a time-consuming, radiologist-dependent task. The aim of this study was to assess whether a deep learning algorithm can replace PI-RADS 2.1 based ellipsoid formula (EF) for calculating PV.

Methods

Eight different measures of PV were retrospectively collected for each of 124 patients who underwent radical prostatectomy and preoperative MRI of the prostate (multicenter and multi-scanner MRI’s 1.5 and 3 T). Agreement between volumes obtained from the deep learning algorithm (PV _DL) and ellipsoid formula by two radiologists (PV _EF1 and PV _EF2) was evaluated against the reference standard PV obtained by manual planimetry by an expert radiologist (PV _MPE). A sensitivity analysis was performed using a prostatectomy specimen as the reference standard. Inter-reader agreement was evaluated between the radiologists using the ellipsoid formula and between the expert and inexperienced radiologists performing manual planimetry.

Results

PV _DL showed better agreement and precision than PV _EF1 and PV _EF2 using the reference standard PV _MPE (mean difference [95% limits of agreement] PV _DL: −0.33 [−10.80; 10.14], PV _EF1: −3.83 [−19.55; 11.89], PV _EF2: −3.05 [−18.55; 12.45]) or the PV determined based on specimen weight (PV _DL: −4.22 [−22.52; 14.07], PV _EF1: −7.89 [−30.50; 14.73], PV _EF2: −6.97 [−30.13; 16.18]). Inter-reader agreement was excellent between the two experienced radiologists using the ellipsoid formula and was good between expert and inexperienced radiologists performing manual planimetry.

Conclusion

Deep learning algorithm performs similarly to radiologists in the assessment of prostate volume on MRI.

Key Points

• A commercially available deep learning algorithm performs similarly to radiologists in the assessment of prostate volume on MRI.

• The deep-learning algorithm was previously untrained on this heterogenous multicenter day-to-day practice MRI data set.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-022-09239-8.