Artificial Intelligence in Magnetic Resonance Imaging-based Prostate Cancer Diagnosis: Where Do We Stand in 2021?

Men suspected of harboring prostate cancer (PCa) increasingly undergo multiparametric magnetic resonance imaging (mpMRI) and mpMRI-guided biopsy. The potential of mpMRI coupled to artificial intelligence (AI) methods to detect and classify PCa before decision-making requires investigation. To review the literature for studies addressing the diagnostic performance of combined mpMRI and AI approaches to detect and classify PCa, and to provide selection criteria for relevant articles having clinical significance. We performed a nonsystematic search of the English language literature using the PubMed-MEDLINE database up to October 30, 2020. We included all original studies addressing the diagnostic accuracy of mpMRI and AI to detect and classify PCa with histopathological analysis as a reference standard. Eleven studies assessed AI and mpMRI approaches for PCa detection and classification based on a ground truth that referred to the entire prostate either with radical prostatectomy specimens (RPS) or relocalization of positive systematic and/or targeted biopsy. Seven studies retrospectively annotated cancerous lesions onto mpMRI identified in whole-mount sections from RPS, three studies used a backward projection of histological prostate biopsy information, and one study used a combined cohort of both approaches. All studies cross-validated their data sets; only four used a test set and one a multisite validation scheme. Performance metrics for lesion detection ranged from 87.9% to 92% at a threshold specificity of 50%. The lesion classification accuracy of the algorithms was comparable to that of the Prostate Imaging-Reporting and Data System. For an algorithm to be implemented into radiological workflows and to be clinically applicable, it must be trained with a ground truth labeling that reflects histopathological information for the entire prostate and it must be externally validated. Lesion detection and classification performance metrics are promising but require prospective implementation and external validation for clinical significance. We reviewed the literature for studies on prostate cancer detection and classification using magnetic resonance imaging (MRI) and artificial intelligence algorithms. The main application is in supporting radiologists in interpreting MRI scans and improving the diagnostic performance, so that fewer unnecessary biopsies are carried out.

Suarez-Ibarrola R ，Sigle A ，Eklund M ，Eberli D ，Miernik A ，Benndorf M ，Bamberg F ，Gratzke C ... -  《European Urology Focus》

Prostate Cancer Risk Assessment in Biopsy-naïve Patients: The Rotterdam Prostate Cancer Risk Calculator in Multiparametric Magnetic Resonance Imaging-Transrectal Ultrasound (TRUS) Fusion Biopsy and Systematic TRUS Biopsy.

Mannaerts CK ，Gayet M ，Verbeek JF ，Engelbrecht MRW ，Savci-Heijink CD ，Jager GJ ，Gielens MPM ，van der Linden H ，Beerlage HP ，de Reijke TM ，Wijkstra H ，Roobol MJ ... -  《European Urology Oncology》

Evaluation of a Cascaded Deep Learning-based Algorithm for Prostate Lesion Detection at Biparametric MRI.

Lin Y ，Yilmaz EC ，Belue MJ ，Harmon SA ，Tetreault J ，Phelps TE ，Merriman KM ，Hazen L ，Garcia C ，Yang D ，Xu Z ，Lay NS ，Toubaji A ，Merino MJ ，Xu D ，Law YM ，Gurram S ，Wood BJ ，Choyke PL ，Pinto PA ，Turkbey B ... -  《-》

Magnetic resonance imaging in prostate cancer detection and management: a systematic review.

Monni F ，Fontanella P ，Grasso A ，Wiklund P ，Ou YC ，Randazzo M ，Rocco B ，Montanari E ，Bianchi G ... -  《-》

Multiparametric Magnetic Resonance Imaging (MRI) and MRI-Transrectal Ultrasound Fusion Biopsy for Index Tumor Detection: Correlation with Radical Prostatectomy Specimen.

Multiparametric magnetic resonance imaging (mpMRI) and MRI fusion targeted biopsy (FTB) detect significant prostate cancer (sPCa) more accurately than conventional biopsies alone. To evaluate the detection accuracy of mpMRI and FTB on radical prostatectomy (RP) specimen. From a cohort of 755 men who underwent transperineal MRI and transrectal ultrasound fusion biopsy under general anesthesia between 2012 and 2014, we retrospectively analyzed 120 consecutive patients who had subsequent RP. All received saturation biopsy (SB) in addition to FTB of lesions with Prostate Imaging Reporting and Data System (PI-RADS) score ≥2. The index lesion was defined as the lesion with extraprostatic extension, the highest Gleason score (GS), or the largest tumor volume (TV) if GS were the same, in order of priority. GS 3+3 and TV ≥1.3ml or GS ≥3+4 and TV ≥0.55ml were considered sPCa. We assessed the detection accuracy by mpMRI and different biopsy approaches and analyzed lesion agreement between mpMRI and RP specimen. Overall, 120 index and 71 nonindex lesions were detected. Overall, 107 (89%) index and 51 (72%) nonindex lesions harbored sPCa. MpMRI detected 110 of 120 (92%) index lesions, FTB (two cores per lesion) alone diagnosed 96 of 120 (80%) index lesions, and SB alone diagnosed 110 of 120 (92%) index lesions. Combined SB and FTB detected 115 of 120 (96%) index foci. FTB performed significantly less accurately compared with mpMRI (p=0.02) and the combination for index lesion detection (p=0.002). Combined FTB and SB detected 97% of all sPCa lesions and was superior to mpMRI (85%), FTB (79%), and SB (88%) alone (p<0.001 each). Spearman's rank correlation coefficient for index lesion agreement between mpMRI and RP was 0.87 (p<0.001). Limitations included the retrospective design, multiple operators, and nonblinding of radiologists. MpMRI identified 92% of index lesions compared with RP histopathology. The combination of FTB and SB was superior to both approaches alone, reliably detecting 97% of sPCa lesions. Multiparametric magnetic resonance imaging detects the index lesion accurately in 9 of 10 patients; however, the combined biopsy approach, while missing less significant cancer, comes at the cost of detecting more insignificant cancer.

Radtke JP ，Schwab C ，Wolf MB ，Freitag MT ，Alt CD ，Kesch C ，Popeneciu IV ，Huettenbrink C ，Gasch C ，Klein T ，Bonekamp D ，Duensing S ，Roth W ，Schueler S ，Stock C ，Schlemmer HP ，Roethke M ，Hohenfellner M ，Hadaschik BA ... -  《-》