Deep learning–assisted prostate cancer detection on bi-parametric MRI: minimum training data size requirements and effect of prior knowledge

Abstract

Objectives To assess Prostate Imaging Reporting and Data System (PI-RADS)–trained deep learning (DL) algorithm performance and to investigate the effect of data size and prior knowledge on the detection of clinically significant prostate cancer (csPCa) in biopsy-naïve men with a suspicion of PCa. Methods Multi-institution data included 2734 consecutive biopsy-naïve men with elevated PSA levels (≥ 3 ng/mL) that underwent multi-parametric MRI (mpMRI). mpMRI exams were prospectively reported using PI-RADS v2 by expert radiologists. A DL framework was designed and trained on center 1 data (n = 1952) to predict PI-RADS ≥ 4 (n = 1092) lesions from bi-parametric MRI (bpMRI). Experiments included varying the number of cases and the use of automatic zonal segmentation as a DL prior. Independent center 2 cases (n = 296) that included pathology outcome (systematic and MRI targeted biopsy) were used to compute performance for radiologists and DL. The performance of detecting PI-RADS 4–5 and Gleason > 6 lesions was assessed on 782 unseen cases (486 center 1, 296 center 2) using free-response ROC (FROC) and ROC analysis. Results The DL sensitivity for detecting PI-RADS ≥ 4 lesions was 87% (193/223, 95% CI: 82–91) at an average of 1 false positive (FP) per patient, and an AUC of 0.88 (95% CI: 0.84–0.91). The DL sensitivity for the detection of Gleason > 6 lesions was 85% (79/93, 95% CI: 77–83) @ 1 FP compared to 91% (85/93, 95% CI: 84–96) @ 0.3 FP for a consensus panel of expert radiologists. Data size and prior zonal knowledge significantly affected performance (4%, $$p<0.05$$ p < 0.05 ). Conclusion PI-RADS-trained DL can accurately detect and localize Gleason > 6 lesions. DL could reach expert performance using substantially more than 2000 training cases, and DL zonal segmentation. Key Points • AI for prostate MRI analysis depends strongly on data size and prior zonal knowledge. • AI needs substantially more than 2000 training cases to achieve expert performance.

Automatic zonal segmentation of the prostate from 2D and 3D T2-weighted MRI and evaluation for clinical use

“…This method can be applied on both 2D and 3D T2w MRI sequences. The obtained results not only are similar to the results from the state of the art but are also coherent with the results obtained by radiologists and globally preserving zonal locations and sectorial positions of the lesions, making our method suitable as a first step tool for an automated system dedicated to diagnosis and grading of PCa, as done by Hosseinzadeh et al 70 or by Mehta et al 71…”

Usefulness of Collaborative Work in the Evaluation of Prostate Cancer from MRI

“…Many studies have shown that deep learning strategies can achieve better outcomes than simpler systems that make use of pathology samples [ 29 ]. There are other examples of algorithms based on artificial intelligence and machine learning in PCa that could be an excellent addition to our work [ 30 , 31 , 32 ]. Finally, considering the difficulties to segment the prostate gland regions, a solution based on AI was proposed by Bardis et al [ 33 ].…”

Diffusion-Weighted MRI in the Genitourinary System

“…The expert panel and studies suggest a minimum of 100 training cases before obtaining an AUC on par with more experimented readers [ 70 , 71 ]. However, training requirements may be drastically modified by the introduction of new machine learning algorithms to assist prostate MRI analysis [ 72 , 73 ].…”