Performance of Deep Learning and Genitourinary Radiologists in Detection of Prostate Cancer Using 3-T Multiparametric Magnetic Resonance Imaging

Abstract

Background: Several deep learning-based techniques have been developed for prostate cancer (PCa) detection using multiparametric magnetic resonance imaging (mpMRI), but few of them have been rigorously evaluated relative to radiologists' performance or whole-mount histopathology (WMHP).

Purpose: To compare the performance of a previously proposed deep learning algorithm, FocalNet, and expert radiologists in the detection of PCa on mpMRI with WMHP as the reference.

Study type: Retrospective, single-center study.

Subjects: A total of 553 patients (development cohort: 427 patients; evaluation cohort: 126 patients) who underwent 3-T mpMRI prior to radical prostatectomy from October 2010 to February 2018.

Field strength/sequence: 3-T, T2-weighted imaging and diffusion-weighted imaging.

Assessment: FocalNet was trained on the development cohort to predict PCa locations by detection points, with a confidence value for each point, on the evaluation cohort. Four fellowship-trained genitourinary (GU) radiologists independently evaluated the evaluation cohort to detect suspicious PCa foci, annotate detection point locations, and assign a five-point suspicion score (1: least suspicious, 5: most suspicious) for each annotated detection point. The PCa detection performance of FocalNet and radiologists were evaluated by the lesion detection sensitivity vs. the number of false-positive detections at different thresholds on suspicion scores. Clinically significant lesions: Gleason Group (GG) ≥ 2 or pathological size ≥ 10 mm. Index lesions: the highest GG and the largest pathological size (secondary).

Statistical tests: Bootstrap hypothesis test for the detection sensitivity between radiologists and FocalNet.

Results: For the overall differential detection sensitivity, FocalNet was 5.1% and 4.7% below the radiologists for clinically significant and index lesions, respectively; however, the differences were not statistically significant (P = 0.413 and P = 0.282, respectively).

Data conclusion: FocalNet achieved slightly lower but not statistically significant PCa detection performance compared with GU radiologists. Compared with radiologists, FocalNet demonstrated similar detection performance for a highly sensitive setting (suspicion score ≥ 1) or a highly specific setting (suspicion score = 5), while lower performance in between.

Level of evidence: 3 TECHNICAL EFFICACY STAGE: 2.

Keywords: automatic cancer detection; deep learning; multiparametric MRI; prostate cancer.

Figure 1:

Illustration of the processes for FocalNet development (left) and the evaluation (right) of cancer detection performance by FocalNet and experienced genitourinary radiologists under the same setting. ADC: apparent diffusion coefficient. Confid: confidence value.

Figure 2:

Flowchart for study inclusion for the evaluation cohort.

Figure 3:

Examples of lesion detection. The left two columns show the input T2WI and ADC map, respectively. The right two columns show the FocalNet predicted lesion probability map and detection points (green crosses) with reference lesion annotation (red contours), respectively. a) patient at age 66, with a PCa lesion at left anterior peripheral zone with Gleason Group 5 (Gleason Score 4+5). b) patient at age 68, with a PCa lesion at left posterolateral peripheral zone with Gleason Group 2 (Gleason Score 3+4). c) patient at age 69, with a PCa lesion at right posterolateral peripheral zone with Gleason Group 3 (Gleason Score 4+3).

Figure 4:

Free-response receiver operating characteristics (FROC) analysis for index lesion detection for 126 patients in the evaluation cohort with detection sensitivity plotted as a function of the number of false-positive detections for each patient on average. The shaded area surrounding the FocalNet curve (blue) shows the 95% confidence interval for detection sensitivity by bootstrapping the patient population. Dots indicate each radiologist performance at suspicion score thresholds.

Figure 5:

Free-response receiver operating characteristics (FROC) analysis for clinically significant lesion detection. The shaded area surrounding the FocalNet curve (blue) shows the 95% confidence interval for detection sensitivity by bootstrapping the patient population. Dots indicate each radiologist performance at suspicion score thresholds.