Artificial intelligence and radiologists in prostate cancer detection on MRI (PI-CAI): an international, paired, non-inferiority, confirmatory study.

BACKGROUND: Artificial intelligence (AI) systems can potentially aid the diagnostic pathway of prostate cancer by alleviating the increasing workload, preventing overdiagnosis, and reducing the dependence on experienced radiologists. We aimed to investigate the performance of AI systems at detecting clinically significant prostate cancer on MRI in comparison with radiologists using the Prostate Imaging-Reporting and Data System version 2.1 (PI-RADS 2.1) and the standard of care in multidisciplinary routine practice at scale. METHODS: In this international, paired, non-inferiority, confirmatory study, we trained and externally validated an AI system (developed within an international consortium) for detecting Gleason grade group 2 or greater cancers using a retrospective cohort of 10 207 MRI examinations from 9129 patients. Of these examinations, 9207 cases from three centres (11 sites) based in the Netherlands were used for training and tuning, and 1000 cases from four centres (12 sites) based in the Netherlands and Norway were used for testing. In parallel, we facilitated a multireader, multicase observer study with 62 radiologists (45 centres in 20 countries; median 7 [IQR 5-10] years of experience in reading prostate MRI) using PI-RADS (2.1) on 400 paired MRI examinations from the testing cohort. Primary endpoints were the sensitivity, specificity, and the area under the receiver operating characteristic curve (AUROC) of the AI system in comparison with that of all readers using PI-RADS (2.1) and in comparison with that of the historical radiology readings made during multidisciplinary routine practice (ie, the standard of care with the aid of patient history and peer consultation). Histopathology and at least 3 years (median 5 [IQR 4-6] years) of follow-up were used to establish the reference standard. The statistical analysis plan was prespecified with a primary hypothesis of non-inferiority (considering a margin of 0·05) and a secondary hypothesis of superiority towards the AI system, if non-inferiority was confirmed. This study was registered at ClinicalTrials.gov, NCT05489341. FINDINGS: Of the 10 207 examinations included from Jan 1, 2012, through Dec 31, 2021, 2440 cases had histologically confirmed Gleason grade group 2 or greater prostate cancer. In the subset of 400 testing cases in which the AI system was compared with the radiologists participating in the reader study, the AI system showed a statistically superior and non-inferior AUROC of 0·91 (95% CI 0·87-0·94; p<0·0001), in comparison to the pool of 62 radiologists with an AUROC of 0·86 (0·83-0·89), with a lower boundary of the two-sided 95% Wald CI for the difference in AUROC of 0·02. At the mean PI-RADS 3 or greater operating point of all readers, the AI system detected 6·8% more cases with Gleason grade group 2 or greater cancers at the same specificity (57·7%, 95% CI 51·6-63·3), or 50·4% fewer false-positive results and 20·0% fewer cases with Gleason grade group 1 cancers at the same sensitivity (89·4%, 95% CI 85·3-92·9). In all 1000 testing cases where the AI system was compared with the radiology readings made during multidisciplinary practice, non-inferiority was not confirmed, as the AI system showed lower specificity (68·9% [95% CI 65·3-72·4] vs 69·0% [65·5-72·5]) at the same sensitivity (96·1%, 94·0-98·2) as the PI-RADS 3 or greater operating point. The lower boundary of the two-sided 95% Wald CI for the difference in specificity (-0·04) was greater than the non-inferiority margin (-0·05) and a p value below the significance threshold was reached (p<0·001). INTERPRETATION: An AI system was superior to radiologists using PI-RADS (2.1), on average, at detecting clinically significant prostate cancer and comparable to the standard of care. Such a system shows the potential to be a supportive tool within a primary diagnostic setting, with several associated benefits for patients and radiologists. Prospective validation is needed to test clinical applicability of this system. FUNDING: Health~Holland and EU Horizon 2020.

Weakly Supervised MRI Slice-Level Deep Learning Classification of Prostate Cancer Approximates Full Voxel- and Slice-Level Annotation: Effect of Increasing Training Set Size.

BACKGROUND: Weakly supervised learning promises reduced annotation effort while maintaining performance. PURPOSE: To compare weakly supervised training with full slice-wise annotated training of a deep convolutional classification network (CNN) for prostate cancer (PC). STUDY TYPE: Retrospective. SUBJECTS: One thousand four hundred eighty-nine consecutive institutional prostate MRI examinations from men with suspicion for PC (65 ± 8 years) between January 2015 and November 2020 were split into training (N = 794, enriched with 204 PROSTATEx examinations) and test set (N = 695). FIELD STRENGTH/SEQUENCE: 1.5 and 3T, T2-weighted turbo-spin-echo and diffusion-weighted echo-planar imaging. ASSESSMENT: Histopathological ground truth was provided by targeted and extended systematic biopsy. Reference training was performed using slice-level annotation (SLA) and compared to iterative training utilizing patient-level annotations (PLAs) with supervised feedback of CNN estimates into the next training iteration at three incremental training set sizes (N = 200, 500, 998). Model performance was assessed by comparing specificity at fixed sensitivity of 0.97 [254/262] emulating PI-RADS ≥ 3, and 0.88-0.90 [231-236/262] emulating PI-RADS ≥ 4 decisions. STATISTICAL TESTS: Receiver operating characteristic (ROC) and area under the curve (AUC) was compared using DeLong and Obuchowski test. Sensitivity and specificity were compared using McNemar test. Statistical significance threshold was P = 0.05. RESULTS: Test set (N = 695) ROC-AUC performance of SLA (trained with 200/500/998 exams) was 0.75/0.80/0.83, respectively. PLA achieved lower ROC-AUC of 0.64/0.72/0.78. Both increased performance significantly with increasing training set size. ROC-AUC for SLA at 500 exams was comparable to PLA at 998 exams (P = 0.28). ROC-AUC was significantly different between SLA and PLA at same training set sizes, however the ROC-AUC difference decreased significantly from 200 to 998 training exams. Emulating PI-RADS ≥ 3 decisions, difference between PLA specificity of 0.12 [51/433] and SLA specificity of 0.13 [55/433] became undetectable (P = 1.0) at 998 exams. Emulating PI-RADS ≥ 4 decisions, at 998 exams, SLA specificity of 0.51 [221/433] remained higher than PLA specificity at 0.39 [170/433]. However, PLA specificity at 998 exams became comparable to SLA specificity of 0.37 [159/433] at 200 exams (P = 0.70). DATA CONCLUSION: Weakly supervised training of a classification CNN using patient-level-only annotation had lower performance compared to training with slice-wise annotations, but improved significantly faster with additional training data. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Prostate cancer risk assessment and avoidance of prostate biopsies using fully automatic deep learning in prostate MRI: comparison to PI-RADS and integration with clinical data in nomograms.

OBJECTIVES: Risk calculators (RCs) improve patient selection for prostate biopsy with clinical/demographic information, recently with prostate MRI using the prostate imaging reporting and data system (PI-RADS). Fully-automated deep learning (DL) analyzes MRI data independently, and has been shown to be on par with clinical radiologists, but has yet to be incorporated into RCs. The goal of this study is to re-assess the diagnostic quality of RCs, the impact of replacing PI-RADS with DL predictions, and potential performance gains by adding DL besides PI-RADS. MATERIAL AND METHODS: One thousand six hundred twenty-seven consecutive examinations from 2014 to 2021 were included in this retrospective single-center study, including 517 exams withheld for RC testing. Board-certified radiologists assessed PI-RADS during clinical routine, then systematic and MRI/Ultrasound-fusion biopsies provided histopathological ground truth for significant prostate cancer (sPC). nnUNet-based DL ensembles were trained on biparametric MRI predicting the presence of sPC lesions (UNet-probability) and a PI-RADS-analogous five-point scale (UNet-Likert). Previously published RCs were validated as is; with PI-RADS substituted by UNet-Likert (UNet-Likert-substituted RC); and with both UNet-probability and PI-RADS (UNet-probability-extended RC). Together with a newly fitted RC using clinical data, PI-RADS and UNet-probability, existing RCs were compared by receiver-operating characteristics, calibration, and decision-curve analysis. RESULTS: Diagnostic performance remained stable for UNet-Likert-substituted RCs. DL contained complementary diagnostic information to PI-RADS. The newly-fitted RC spared 49% [252/517] of biopsies while maintaining the negative predictive value (94%), compared to PI-RADS ≥ 4 cut-off which spared 37% [190/517] (p < 0.001). CONCLUSIONS: Incorporating DL as an independent diagnostic marker for RCs can improve patient stratification before biopsy, as there is complementary information in DL features and clinical PI-RADS assessment. CLINICAL RELEVANCE STATEMENT: For patients with positive prostate screening results, a comprehensive diagnostic workup, including prostate MRI, DL analysis, and individual classification using nomograms can identify patients with minimal prostate cancer risk, as they benefit less from the more invasive biopsy procedure. KEY POINTS: The current MRI-based nomograms result in many negative prostate biopsies. The addition of DL to nomograms with clinical data and PI-RADS improves patient stratification before biopsy. Fully automatic DL can be substituted for PI-RADS without sacrificing the quality of nomogram predictions. Prostate nomograms show cancer detection ability comparable to previous validation studies while being suitable for the addition of DL analysis.

Contribution of Dynamic Contrast-enhanced and Diffusion MRI to PI-RADS for Detecting Clinically Significant Prostate Cancer.

Background Prostate Imaging Reporting and Data System (PI-RADS) version 2.0 requires multiparametric MRI of the prostate, including diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) imaging sequences; however, the contribution of DCE imaging remains unclear. Purpose To assess whether DCE imaging in addition to apparent diffusion coefficient (ADC) and normalized T2 values improves PI-RADS version 2.0 for prediction of clinically significant prostate cancer (csPCa). Materials and Methods In this retrospective study, clinically reported PI-RADS lesions in consecutive men who underwent 3-T multiparametric MRI (T2-weighted, DWI, and DCE MRI) from May 2015 to September 2016 were analyzed quantitatively and compared with systematic and targeted MRI-transrectal US fusion biopsy. The normalized T2 signal (nT2), ADC measurement, mean early-phase DCE signal (mDCE), and heuristic DCE parameters were calculated. Logistic regression analysis indicated the most predictive DCE parameters for csPCa (Gleason grade group ≥2). Receiver operating characteristic parameter models were compared using the Obuchowski test. Recursive partitioning analysis determined ADC and mDCE value ranges for combined use with PI-RADS. Results Overall, 260 men (median age, 64 years [IQR, 58-69 years]) with 432 lesions (csPCa [n = 152] and no csPCa [n = 280]) were included. The mDCE parameter was predictive of csPCa when accounting for the ADC and nT2 parameter in the peripheral zone (odds ratio [OR], 1.76; 95% CI: 1.30, 2.44; P = .001) but not the transition zone (OR, 1.17; 95% CI: 0.81, 1.69; P = .41). Recursive partitioning analysis selected an ADC cutoff of 0.897 × 10-3 mm2/sec (P = .04) as a classifier for peripheral zone lesions with a PI-RADS score assessed on the ADC map (hereafter, ADC PI-RADS) of 3. The mDCE parameter did not differentiate ADC PI-RADS 3 lesions (P = .11), but classified lesions with ADC PI-RADS scores greater than 3 with low ADC values (less than 0.903 × 10-3 mm2/sec, P < .001) into groups with csPCa rates of 70% and 97% (P = .008). A lesion size cutoff of 1.5 cm and qualitative DCE parameters were not defined as classifiers according to recursive partitioning (P > .05). Conclusion Quantitative or qualitative dynamic contrast-enhanced MRI was not relevant for Prostate Imaging Reporting and Data System (PI-RADS) 3 lesion risk stratification, while quantitative apparent diffusion coefficient (ADC) values were helpful in upgrading PI-RADS 3 and PI-RADS 4 lesions. Quantitative ADC measurement may be more important for risk stratification than current methods in future versions of PI-RADS. © RSNA, 2022 Online supplemental material is available for this article See also the editorial by Goh in this issue.

Application of a validated prostate MRI deep learning system to independent same-vendor multi-institutional data: demonstration of transferability.

OBJECTIVES: To evaluate a fully automatic deep learning system to detect and segment clinically significant prostate cancer (csPCa) on same-vendor prostate MRI from two different institutions not contributing to training of the system. MATERIALS AND METHODS: In this retrospective study, a previously bi-institutionally validated deep learning system (UNETM) was applied to bi-parametric prostate MRI data from one external institution (A), a PI-RADS distribution-matched internal cohort (B), and a csPCa stratified subset of single-institution external public challenge data (C). csPCa was defined as ISUP Grade Group ≥ 2 determined from combined targeted and extended systematic MRI/transrectal US-fusion biopsy. Performance of UNETM was evaluated by comparing ROC AUC and specificity at typical PI-RADS sensitivity levels. Lesion-level analysis between UNETM segmentations and radiologist-delineated segmentations was performed using Dice coefficient, free-response operating characteristic (FROC), and weighted alternative (waFROC). The influence of using different diffusion sequences was analyzed in cohort A. RESULTS: In 250/250/140 exams in cohorts A/B/C, differences in ROC AUC were insignificant with 0.80 (95% CI: 0.74-0.85)/0.87 (95% CI: 0.83-0.92)/0.82 (95% CI: 0.75-0.89). At sensitivities of 95% and 90%, UNETM achieved specificity of 30%/50% in A, 44%/71% in B, and 43%/49% in C, respectively. Dice coefficient of UNETM and radiologist-delineated lesions was 0.36 in A and 0.49 in B. The waFROC AUC was 0.67 (95% CI: 0.60-0.83) in A and 0.7 (95% CI: 0.64-0.78) in B. UNETM performed marginally better on readout-segmented than on single-shot echo-planar-imaging. CONCLUSION: For same-vendor examinations, deep learning provided comparable discrimination of csPCa and non-csPCa lesions and examinations between local and two independent external data sets, demonstrating the applicability of the system to institutions not participating in model training. CLINICAL RELEVANCE STATEMENT: A previously bi-institutionally validated fully automatic deep learning system maintained acceptable exam-level diagnostic performance in two independent external data sets, indicating the potential of deploying AI models without retraining or fine-tuning, and corroborating evidence that AI models extract a substantial amount of transferable domain knowledge about MRI-based prostate cancer assessment. KEY POINTS: • A previously bi-institutionally validated fully automatic deep learning system maintained acceptable exam-level diagnostic performance in two independent external data sets. • Lesion detection performance and segmentation congruence was similar on the institutional and an external data set, as measured by the weighted alternative FROC AUC and Dice coefficient. • Although the system generalized to two external institutions without re-training, achieving expected sensitivity and specificity levels using the deep learning system requires probability thresholds to be adjusted, underlining the importance of institution-specific calibration and quality control.

Magnetic Resonance Imaging-Guided Transurethral Ultrasound Ablation of Prostate Cancer.

PURPOSE: Magnetic resonance imaging-guided transurethral ultrasound ablation uses directional thermal ultrasound under magnetic resonance imaging thermometry feedback control for prostatic ablation. We report 12-month outcomes from a prospective multicenter trial (TACT). MATERIALS AND METHODS: A total of 115 men with favorable to intermediate risk prostate cancer across 13 centers were treated with whole gland ablation sparing the urethra and apical sphincter. The co-primary 12-month endpoints were safety and efficacy. RESULTS: In all, 72 (63%) had grade group 2 and 77 (67%) had NCCN® intermediate risk disease. Median treatment delivery time was 51 minutes with 98% (IQR 95-99) thermal coverage of target volume and spatial ablation precision of ±1.4 mm on magnetic resonance imaging thermometry. Grade 3 adverse events occurred in 9 (8%) men. The primary endpoint (U.S. Food and Drug Administration mandated) of prostate specific antigen reduction ≥75% was achieved in 110 of 115 (96%) with median prostate specific antigen reduction of 95% and nadir of 0.34 ng/ml. Median prostate volume decreased from 37 to 3 cc. Among 68 men with pretreatment grade group 2 disease, 52 (79%) were free of grade group 2 disease on 12-month biopsy. Of 111 men with 12-month biopsy data, 72 (65%) had no evidence of cancer. Erections (International Index of Erectile Function question 2 score 2 or greater) were maintained/regained in 69 of 92 (75%). Multivariate predictors of persistent grade group 2 at 12 months included intraprostatic calcifications at screening, suboptimal magnetic resonance imaging thermal coverage of target volume and a PI-RADS™ 3 or greater lesion at 12-month magnetic resonance imaging (p <0.05). CONCLUSIONS: The TACT study of magnetic resonance imaging-guided transurethral ultrasound whole gland ablation in men with localized prostate cancer demonstrated effective tissue ablation and prostate specific antigen reduction with low rates of toxicity and residual disease.

Magnetic resonance imaging-guided transurethral ultrasound ablation in patients with localised prostate cancer: 3-year outcomes of a prospective Phase I study.

OBJECTIVES: To report the 3-year follow-up of a Phase I study of magnetic resonance imaging (MRI)-guided transurethral ultrasound ablation (TULSA) in 30 men with localised prostate cancer. Favourable 12-month safety and ablation precision were previously described. PATIENTS AND METHODS: As a mandated safety criterion, TULSA was delivered as near whole-gland ablation, applying 3-mm margins sparing 10% of peripheral prostate tissue in 30 men. After 12-month biopsy and MRI, biannual follow-up included prostate-specific antigen (PSA), adverse events (AEs), and functional quality-of-life assessment, with repeat systematic biopsy at 3 years. RESULTS: A 3-year follow-up was completed by 22 patients. Between 1 and 3 years, there were no new serious or severe AEs. Urinary and bowel function remained stable. Erectile function recovered by 1 year and was stable at 3 years. The PSA level decreased 95% to a median (interquartile range) nadir of 0.33 (0.1-0.4) ng/mL, stable to 0.8 (0.4-1.6) ng/mL at 3 years. Serial biopsies identified clinically significant disease in 10/29 men (34%) and any cancer in 17/29 (59%). By 3 years, seven men had recurrence (four histological, three biochemical) and had undergone salvage therapy without complications (including six prostatectomies). At 3 years, three of 22 men refused biopsy, and two of the 22 (9%) had clinically significant disease (one new, one persistent). Predictors of salvage therapy requirement included less extensive ablation coverage and higher PSA nadir. CONCLUSION: With 3-year Phase I follow-up, TULSA demonstrates safe and precise ablation for men with localised prostate cancer, providing predictable PSA and biopsy outcomes, without affecting functional abilities or precluding salvage therapy.

Simulated clinical deployment of fully automatic deep learning for clinical prostate MRI assessment.

OBJECTIVES: To simulate clinical deployment, evaluate performance, and establish quality assurance of a deep learning algorithm (U-Net) for detection, localization, and segmentation of clinically significant prostate cancer (sPC), ISUP grade group ≥ 2, using bi-parametric MRI. METHODS: In 2017, 284 consecutive men in active surveillance, biopsy-naïve or pre-biopsied, received targeted and extended systematic MRI/transrectal US-fusion biopsy, after examination on a single MRI scanner (3 T). A prospective adjustment scheme was evaluated comparing the performance of the Prostate Imaging Reporting and Data System (PI-RADS) and U-Net using sensitivity, specificity, predictive values, and the Dice coefficient. RESULTS: In the 259 eligible men (median 64 [IQR 61-72] years), PI-RADS had a sensitivity of 98% [106/108]/84% [91/108] with a specificity of 17% [25/151]/58% [88/151], for thresholds at ≥ 3/≥ 4 respectively. U-Net using dynamic threshold adjustment had a sensitivity of 99% [107/108]/83% [90/108] (p > 0.99/> 0.99) with a specificity of 24% [36/151]/55% [83/151] (p > 0.99/> 0.99) for probability thresholds d3 and d4 emulating PI-RADS ≥ 3 and ≥ 4 decisions respectively, not statistically different from PI-RADS. Co-occurrence of a radiological PI-RADS ≥ 4 examination and U-Net ≥ d3 assessment significantly improved the positive predictive value from 59 to 63% (p = 0.03), on a per-patient basis. CONCLUSIONS: U-Net has similar performance to PI-RADS in simulated continued clinical use. Regular quality assurance should be implemented to ensure desired performance. KEY POINTS: • U-Net maintained similar diagnostic performance compared to radiological assessment of PI-RADS ≥ 4 when applied in a simulated clinical deployment. • Application of our proposed prospective dynamic calibration method successfully adjusted U-Net performance within acceptable limits of the PI-RADS reference over time, while not being limited to PI-RADS as a reference. • Simultaneous detection by U-Net and radiological assessment significantly improved the positive predictive value on a per-patient and per-lesion basis, while the negative predictive value remained unchanged.

Automated volumetric assessment with artificial neural networks might enable a more accurate assessment of disease burden in patients with multiple sclerosis.

OBJECTIVES: Patients with multiple sclerosis (MS) regularly undergo MRI for assessment of disease burden. However, interpretation may be time consuming and prone to intra- and interobserver variability. Here, we evaluate the potential of artificial neural networks (ANN) for automated volumetric assessment of MS disease burden and activity on MRI. METHODS: A single-institutional dataset with 334 MS patients (334 MRI exams) was used to develop and train an ANN for automated identification and volumetric segmentation of T2/FLAIR-hyperintense and contrast-enhancing (CE) lesions. Independent testing was performed in a single-institutional longitudinal dataset with 82 patients (266 MRI exams). We evaluated lesion detection performance (F1 scores), lesion segmentation agreement (DICE coefficients), and lesion volume agreement (concordance correlation coefficients [CCC]). Independent evaluation was performed on the public ISBI-2015 challenge dataset. RESULTS: The F1 score was maximized in the training set at a detection threshold of 7 mm3 for T2/FLAIR lesions and 14 mm3 for CE lesions. In the training set, mean F1 scores were 0.867 for T2/FLAIR lesions and 0.636 for CE lesions, as compared to 0.878 for T2/FLAIR lesions and 0.715 for CE lesions in the test set. Using these thresholds, the ANN yielded mean DICE coefficients of 0.834 and 0.878 for segmentation of T2/FLAIR and CE lesions in the training set (fivefold cross-validation). Corresponding DICE coefficients in the test set were 0.846 for T2/FLAIR lesions and 0.908 for CE lesions, and the CCC was ≥ 0.960 in each dataset. CONCLUSIONS: Our results highlight the capability of ANN for quantitative state-of-the-art assessment of volumetric lesion load on MRI and potentially enable a more accurate assessment of disease burden in patients with MS. KEY POINTS: • Artificial neural networks (ANN) can accurately detect and segment both T2/FLAIR and contrast-enhancing MS lesions in MRI data. • Performance of the ANN was consistent in a clinically derived dataset, with patients presenting all possible disease stages in MRI scans acquired from standard clinical routine rather than with high-quality research sequences. • Computer-aided evaluation of MS with ANN could streamline both clinical and research procedures in the volumetric assessment of MS disease burden as well as in lesion detection.

Automated quantitative tumour response assessment of MRI in neuro-oncology with artificial neural networks: a multicentre, retrospective study.

BACKGROUND: The Response Assessment in Neuro-Oncology (RANO) criteria and requirements for a uniform protocol have been introduced to standardise assessment of MRI scans in both clinical trials and clinical practice. However, these criteria mainly rely on manual two-dimensional measurements of contrast-enhancing (CE) target lesions and thus restrict both reliability and accurate assessment of tumour burden and treatment response. We aimed to develop a framework relying on artificial neural networks (ANNs) for fully automated quantitative analysis of MRI in neuro-oncology to overcome the inherent limitations of manual assessment of tumour burden. METHODS: In this retrospective study, we compiled a single-institution dataset of MRI data from patients with brain tumours being treated at Heidelberg University Hospital (Heidelberg, Germany; Heidelberg training dataset) to develop and train an ANN for automated identification and volumetric segmentation of CE tumours and non-enhancing T2-signal abnormalities (NEs) on MRI. Independent testing and large-scale application of the ANN for tumour segmentation was done in a single-institution longitudinal testing dataset from the Heidelberg University Hospital and in a multi-institutional longitudinal testing dataset from the prospective randomised phase 2 and 3 European Organisation for Research and Treatment of Cancer (EORTC)-26101 trial (NCT01290939), acquired at 38 institutions across Europe. In both longitudinal datasets, spatial and temporal tumour volume dynamics were automatically quantified to calculate time to progression, which was compared with time to progression determined by RANO, both in terms of reliability and as a surrogate endpoint for predicting overall survival. We integrated this approach for fully automated quantitative analysis of MRI in neuro-oncology within an application-ready software infrastructure and applied it in a simulated clinical environment of patients with brain tumours from the Heidelberg University Hospital (Heidelberg simulation dataset). FINDINGS: For training of the ANN, MRI data were collected from 455 patients with brain tumours (one MRI per patient) being treated at Heidelberg hospital between July 29, 2009, and March 17, 2017 (Heidelberg training dataset). For independent testing of the ANN, an independent longitudinal dataset of 40 patients, with data from 239 MRI scans, was collected at Heidelberg University Hospital in parallel with the training dataset (Heidelberg test dataset), and 2034 MRI scans from 532 patients at 34 institutions collected between Oct 26, 2011, and Dec 3, 2015, in the EORTC-26101 study were of sufficient quality to be included in the EORTC-26101 test dataset. The ANN yielded excellent performance for accurate detection and segmentation of CE tumours and NE volumes in both longitudinal test datasets (median DICE coefficient for CE tumours 0·89 [95% CI 0·86-0·90], and for NEs 0·93 [0·92-0·94] in the Heidelberg test dataset; CE tumours 0·91 [0·90-0·92], NEs 0·93 [0·93-0·94] in the EORTC-26101 test dataset). Time to progression from quantitative ANN-based assessment of tumour response was a significantly better surrogate endpoint than central RANO assessment for predicting overall survival in the EORTC-26101 test dataset (hazard ratios ANN 2·59 [95% CI 1·86-3·60] vs central RANO 2·07 [1·46-2·92]; p<0·0001) and also yielded a 36% margin over RANO (p<0·0001) when comparing reliability values (ie, agreement in the quantitative volumetrically defined time to progression [based on radiologist ground truth vs automated assessment with ANN] of 87% [266 of 306 with sufficient data] compared with 51% [155 of 306] with local vs independent central RANO assessment). In the Heidelberg simulation dataset, which comprised 466 patients with brain tumours, with 595 MRI scans obtained between April 27, and Sept 17, 2018, automated on-demand processing of MRI scans and quantitative tumour response assessment within the simulated clinical environment required 10 min of computation time (average per scan). INTERPRETATION: Overall, we found that ANN enabled objective and automated assessment of tumour response in neuro-oncology at high throughput and could ultimately serve as a blueprint for the application of ANN in radiology to improve clinical decision making. Future research should focus on prospective validation within clinical trials and application for automated high-throughput imaging biomarker discovery and extension to other diseases. FUNDING: Medical Faculty Heidelberg Postdoc-Program, Else Kröner-Fresenius Foundation.

Automated brain extraction of multisequence MRI using artificial neural networks.

Brain extraction is a critical preprocessing step in the analysis of neuroimaging studies conducted with magnetic resonance imaging (MRI) and influences the accuracy of downstream analyses. The majority of brain extraction algorithms are, however, optimized for processing healthy brains and thus frequently fail in the presence of pathologically altered brain or when applied to heterogeneous MRI datasets. Here we introduce a new, rigorously validated algorithm (termed HD-BET) relying on artificial neural networks that aim to overcome these limitations. We demonstrate that HD-BET outperforms six popular, publicly available brain extraction algorithms in several large-scale neuroimaging datasets, including one from a prospective multicentric trial in neuro-oncology, yielding state-of-the-art performance with median improvements of +1.16 to +2.50 points for the Dice coefficient and -0.66 to -2.51 mm for the Hausdorff distance. Importantly, the HD-BET algorithm, which shows robust performance in the presence of pathology or treatment-induced tissue alterations, is applicable to a broad range of MRI sequence types and is not influenced by variations in MRI hardware and acquisition parameters encountered in both research and clinical practice. For broader accessibility, the HD-BET prediction algorithm is made freely available (www.neuroAI-HD.org) and may become an essential component for robust, automated, high-throughput processing of MRI neuroimaging data.

Classification of Cancer at Prostate MRI: Deep Learning versus Clinical PI-RADS Assessment.

Background Men suspected of having clinically significant prostate cancer (sPC) increasingly undergo prostate MRI. The potential of deep learning to provide diagnostic support for human interpretation requires further evaluation. Purpose To compare the performance of clinical assessment to a deep learning system optimized for segmentation trained with T2-weighted and diffusion MRI in the task of detection and segmentation of lesions suspicious for sPC. Materials and Methods In this retrospective study, T2-weighted and diffusion prostate MRI sequences from consecutive men examined with a single 3.0-T MRI system between 2015 and 2016 were manually segmented. Ground truth was provided by combined targeted and extended systematic MRI-transrectal US fusion biopsy, with sPC defined as International Society of Urological Pathology Gleason grade group greater than or equal to 2. By using split-sample validation, U-Net was internally validated on the training set (80% of the data) through cross validation and subsequently externally validated on the test set (20% of the data). U-Net-derived sPC probability maps were calibrated by matching sextant-based cross-validation performance to clinical performance of Prostate Imaging Reporting and Data System (PI-RADS). Performance of PI-RADS and U-Net were compared by using sensitivities, specificities, predictive values, and Dice coefficient. Results A total of 312 men (median age, 64 years; interquartile range [IQR], 58-71 years) were evaluated. The training set consisted of 250 men (median age, 64 years; IQR, 58-71 years) and the test set of 62 men (median age, 64 years; IQR, 60-69 years). In the test set, PI-RADS cutoffs greater than or equal to 3 versus cutoffs greater than or equal to 4 on a per-patient basis had sensitivity of 96% (25 of 26) versus 88% (23 of 26) at specificity of 22% (eight of 36) versus 50% (18 of 36). U-Net at probability thresholds of greater than or equal to 0.22 versus greater than or equal to 0.33 had sensitivity of 96% (25 of 26) versus 92% (24 of 26) (both P > .99) with specificity of 31% (11 of 36) versus 47% (17 of 36) (both P > .99), not statistically different from PI-RADS. Dice coefficients were 0.89 for prostate and 0.35 for MRI lesion segmentation. In the test set, coincidence of PI-RADS greater than or equal to 4 with U-Net lesions improved the positive predictive value from 48% (28 of 58) to 67% (24 of 36) for U-Net probability thresholds greater than or equal to 0.33 (P = .01), while the negative predictive value remained unchanged (83% [25 of 30] vs 83% [43 of 52]; P > .99). Conclusion U-Net trained with T2-weighted and diffusion MRI achieves similar performance to clinical Prostate Imaging Reporting and Data System assessment. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Padhani and Turkbey in this issue.

Histopathological to multiparametric MRI spatial mapping of extended systematic sextant and MR/TRUS-fusion-targeted biopsy of the prostate.

PURPOSE: MRI has limited ability to detect multifocal disease or the full extent of prostate involvement with clinically significant prostate cancer (sPC). We compare the spatial co-localization at sextant resolution of MRI lesions and histopathological mapping by combined targeted and extended systematic biopsies. MATERIALS AND METHODS: Sextants were mapped for sPC (ISUP group ≥ 2) by 24-core transperineal systematic biopsies in 316 patients with suspicion for sPC and by MR lesions of PI-RADS score of ≥ 3. The gold standard is combined systematic (median 23 cores) and targeted biopsies. RESULTS: Of 316 men, 121 (38%) harbored sPC. Of these 121 patients, 4 (3%) had a negative MRI. MRI correctly identified 117/121 (97%) patients with sPC. In these patients, mpMRI missed no additional sPC in 96 (82%), while MRI-negative sPC lesions were present in 21 patients (18%). Of 1896 sextants, 379 (20%) harbored sPC. MR-positive sextants contained sPC in 26% (337/1275), compared to 7% (42/621) in MR-negative sextants. On a patient basis, sensitivity was 0.97, specificity 0.22, positive predictive value 0.43, and negative predictive value 0.91. On a sextant basis, sensitivity was 0.73, specificity 0.38, positive predictive value 0.26, and negative predictive value 0.93. CONCLUSION: MpMRI mapping agreed well with histopathology with, at the observed sPC prevalence and on a patient basis, excellent sensitivity and negative predictive value, and acceptable specificity and positive predictive value for sPC. However, 18% of sPC was outside the mpMRI mapped region, quantifying limitations of MRI for complete localization of disease extent. KEY POINTS: • Currently, exclusive MRI mapping of the prostate for focal treatment planning cannot be recommended, as significant prostate cancer may remain untreated in a substantial number of cases. • At the observed sPC prevalence and on a patient basis, mpMRI has excellent sensitivity and NPV, and acceptable specificity and PPV for detection of prostate cancer, supporting its use to detect suspicious lesions before biopsy. • Despite the excellent global performance, 18% of sPC was outside the mpMRI mapped region even when a security margin of 10 mm was considered, indicating that prostate MRI has limited ability to completely map all cancer foci within the prostate.

Twelve-month prostate volume reduction after MRI-guided transurethral ultrasound ablation of the prostate.

PURPOSE: To quantitatively assess 12-month prostate volume (PV) reduction based on T2-weighted MRI and immediate post-treatment contrast-enhanced MRI non-perfused volume (NPV), and to compare measurements with predictions of acute and delayed ablation volumes based on MR-thermometry (MR-t), in a central radiology review of the Phase I clinical trial of MRI-guided transurethral ultrasound ablation (TULSA) in patients with localized prostate cancer. MATERIALS AND METHODS: Treatment day MRI and 12-month follow-up MRI and biopsy were available for central radiology review in 29 of 30 patients from the published institutional review board-approved, prospective, multi-centre, single-arm Phase I clinical trial of TULSA. Viable PV at 12 months was measured as the remaining PV on T2-weighted MRI, less 12-month NPV, scaled by the fraction of fibrosis in 12-month biopsy cores. Reduction of viable PV was compared to predictions based on the fraction of the prostate covered by the MR-t derived acute thermal ablation volume (ATAV, 55°C isotherm), delayed thermal ablation volume (DTAV, 240 cumulative equivalent minutes at 43°C thermal dose isocontour) and treatment-day NPV. We also report linear and volumetric comparisons between metrics. RESULTS: After TULSA, the median 12-month reduction in viable PV was 88%. DTAV predicted a reduction of 90%. Treatment day NPV predicted only 53% volume reduction, and underestimated ATAV and DTAV by 36% and 51%. CONCLUSION: Quantitative volumetry of the TULSA phase I MR and biopsy data identifies DTAV (240 CEM43 thermal dose boundary) as a useful predictor of viable prostate tissue reduction at 12 months. Immediate post-treatment NPV underestimates tissue ablation. KEY POINTS: • MRI-guided transurethral ultrasound ablation (TULSA) achieved an 88% reduction of viable prostate tissue volume at 12 months, in excellent agreement with expectation from thermal dose calculations. • Non-perfused volume on immediate post-treatment contrast-enhanced MRI represents only 64% of the acute thermal ablation volume (ATAV), and reports only 60% (53% instead of 88% achieved) of the reduction in viable prostate tissue volume at 12 months. • MR-thermometry-based predictions of 12-month prostate volume reduction based on 240 cumulative equivalent minute thermal dose volume are in excellent agreement with reduction in viable prostate tissue volume measured on pre- and 12-month post-treatment T2w-MRI.

Radiomics Based on Adapted Diffusion Kurtosis Imaging Helps to Clarify Most Mammographic Findings Suspicious for Cancer.

Purpose To evaluate a radiomics model of Breast Imaging Reporting and Data System (BI-RADS) 4 and 5 breast lesions extracted from breast-tissue-optimized kurtosis magnetic resonance (MR) imaging for lesion characterization by using a sensitivity threshold similar to that of biopsy. Materials and Methods This institutional study included 222 women at two independent study sites (site 1: training set of 95 patients; mean age ± standard deviation, 58.6 years ± 6.6; 61 malignant and 34 benign lesions; site 2: independent test set of 127 patients; mean age, 58.2 years ± 6.8; 61 malignant and 66 benign lesions). All women presented with a finding suspicious for cancer at x-ray mammography (BI-RADS 4 or 5) and an indication for biopsy. Before biopsy, diffusion-weighted MR imaging (b values, 0-1500 sec/mm2) was performed by using 1.5-T imagers from different MR imaging vendors. Lesions were segmented and voxel-based kurtosis fitting adapted to account for fat signal contamination was performed. A radiomics feature model was developed by using a random forest regressor. The fixed model was tested on an independent test set. Conventional interpretations of MR imaging were also assessed for comparison. Results The radiomics feature model reduced false-positive results from 66 to 20 (specificity 70.0% [46 of 66]) at the predefined sensitivity of greater than 98.0% [60 of 61] in the independent test set, with BI-RADS 4a and 4b lesions benefiting from the analysis (specificity 74.0%, [37 of 50]; 60.0% [nine of 15]) and BI-RADS 5 lesions showing no added benefit. The model significantly improved specificity compared with the median apparent diffusion coefficient (P < .001) and apparent kurtosis coefficient (P = .02) alone. Conventional reading of dynamic contrast material-enhanced MR imaging provided sensitivity of 91.8% (56 of 61) and a specificity of 74.2% (49 of 66). Accounting for fat signal intensity during fitting significantly improved the area under the curve of the model (P = .001). Conclusion A radiomics model based on kurtosis diffusion-weighted imaging performed by using MR imaging machines from different vendors allowed for reliable differentiation between malignant and benign breast lesions in both a training and an independent test data set. © RSNA, 2018 Online supplemental material is available for this article.

Radiomic Machine Learning for Characterization of Prostate Lesions with MRI: Comparison to ADC Values.

Purpose To compare biparametric contrast-free radiomic machine learning (RML), mean apparent diffusion coefficient (ADC), and radiologist assessment for characterization of prostate lesions detected during prospective MRI interpretation. Materials and Methods This single-institution study included 316 men (mean age ± standard deviation, 64.0 years ± 7.8) with an indication for MRI-transrectal US fusion biopsy between May 2015 and September 2016 (training cohort, 183 patients; test cohort, 133 patients). Lesions identified by prospective clinical readings were manually segmented for mean ADC and radiomics analysis. Global and zone-specific random forest RML and mean ADC models for classification of clinically significant prostate cancer (Gleason grade group ≥ 2) were developed on the training set and the fixed models tested on an independent test set. Clinical readings, mean ADC, and radiomics were compared by using the McNemar test and receiver operating characteristic (ROC) analysis. Results In the test set, radiologist interpretation had a per-lesion sensitivity of 88% (53 of 60) and specificity of 50% (79 of 159). Quantitative measurement of the mean ADC (cut-off 732 mm2/sec) significantly reduced false-positive (FP) lesions from 80 to 60 (specificity 62% [99 of 159]) and false-negative (FN) lesions from seven to six (sensitivity 90% [54 of 60]) (P = .048). Radiologist interpretation had a per-patient sensitivity of 89% (40 of 45) and specificity of 43% (38 of 88). Quantitative measurement of the mean ADC reduced the number of patients with FP lesions from 50 to 43 (specificity 51% [45 of 88]) and the number of patients with FN lesions from five to three (sensitivity 93% [42 of 45]) (P = .496). Comparison of the area under the ROC curve (AUC) for the mean ADC (AUCglobal = 0.84; AUCzone-specific ≤ 0.87) vs the RML (AUCglobal = 0.88, P = .176; AUCzone-specific ≤ 0.89, P ≥ .493) showed no significantly different performance. Conclusion Quantitative measurement of the mean apparent diffusion coefficient (ADC) improved differentiation of benign versus malignant prostate lesions, compared with clinical assessment. Radiomic machine learning had comparable but not better performance than mean ADC assessment. © RSNA, 2018 Online supplemental material is available for this article.

Simultaneous whole-body 18F-PSMA-1007-PET/MRI with integrated high-resolution multiparametric imaging of the prostatic fossa for comprehensive oncological staging of patients with prostate cancer: a pilot study.

INTRODUCTION: The aim of the present study was to explore the clinical feasibility and reproducibility of a comprehensive whole-body 18F-PSMA-1007-PET/MRI protocol for imaging prostate cancer (PC) patients. METHODS: Eight patients with high-risk biopsy-proven PC underwent a whole-body PET/MRI (3 h p.i.) including a multi-parametric prostate MRI after 18F-PSMA-1007-PET/CT (1 h p.i.) which served as reference. Seven patients presented with non-treated PC, whereas one patient presented with biochemical recurrence. SUVmean-quantification was performed using a 3D-isocontour volume-of-interest. Imaging data was consulted for TNM-staging and compared with histopathology. PC was confirmed in 4/7 patients additionally by histopathology after surgery. PET-artifacts, co-registration of pelvic PET/MRI and MRI-data were assessed (PI-RADS 2.0). RESULTS: The examinations were well accepted by patients and comprised 1 h. SUVmean-values between PET/CT (1 h p.i.) and PET/MRI (3 h p.i.) were significantly correlated (p < 0.0001, respectively) and similar to literature of 18F-PSMA-1007-PET/CT 1 h vs 3 h p.i. The dominant intraprostatic lesion could be detected in all seven patients in both PET and MRI. T2c, T3a, T3b and T4 features were detected complimentarily by PET and MRI in five patients. PET/MRI demonstrated moderate photopenic PET-artifacts surrounding liver and kidneys representing high-contrast areas, no PET-artifacts were observed for PET/CT. Simultaneous PET-readout during prostate MRI achieved optimal co-registration results. CONCLUSIONS: The presented 18F-PSMA-1007-PET/MRI protocol combines efficient whole-body assessment with high-resolution co-registered PET/MRI of the prostatic fossa for comprehensive oncological staging of patients with PC.

Multicentre evaluation of magnetic resonance imaging supported transperineal prostate biopsy in biopsy-naïve men with suspicion of prostate cancer.

OBJECTIVES: To analyse the detection rates of primary magnetic resonance imaging (MRI)-fusion transperineal prostate biopsy using combined targeted and systematic core distribution in three tertiary referral centres. PATIENTS AND METHODS: In this multicentre, prospective outcome study, 807 consecutive biopsy-naïve patients underwent MRI-guided transperineal prostate biopsy, as the first diagnostic intervention, between 10/2012 and 05/2016. MRI was reported following the Prostate Imaging-Reporting and Data System (PI-RADS) criteria. In all, 236 patients had 18-24 systematic transperineal biopsies only, and 571 patients underwent additional targeted biopsies either by MRI-fusion or cognitive targeting if PI-RADS ≥3 lesions were present. Detection rates for any and Gleason score 7-10 cancer in targeted and overall biopsy were calculated and predictive values were calculated for different PI-RADS and PSA density (PSAD) groups. RESULTS: Cancer was detected in 68% of the patients (546/807) and Gleason score 7-10 cancer in 49% (392/807). The negative predictive value of 236 PI-RADS 1-2 MRI in combination with PSAD of <0.1 ng/mL/mL for Gleason score 7-10 was 0.91 (95% confidence interval ± 0.07, 8% of study population). In 418 patients with PI-RADS 4-5 lesions using targeted plus systematic biopsies, the cancer detection rate of Gleason score 7-10 was significantly higher at 71% vs 59% and 61% with either approach alone (P < 0.001). For 153 PI-RADS 3 lesions, the detection rate was 31% with no significant difference to systematic biopsies with 27% (P > 0.05). Limitations include variability of multiparametric MRI (mpMRI) reading and Gleason grading. CONCLUSION: MRI-based transperineal biopsy performed at high-volume tertiary care centres with a significant experience of prostate mpMRI and image-guided targeted biopsies yielded high detection rates of Gleason score 7-10 cancer. Prostate biopsies may not be needed for men with low PSAD and an unsuspicious MRI. In patients with high probability lesions, combined targeted and systematic biopsies are recommended.

Multiparametric MRI fusion-guided biopsy for the diagnosis of prostate cancer.

PURPOSE OF REVIEW: To discuss the timing, benefits, limitations and current controversies of multiparametric magnet resonance imaging (mpMRI) combined with fusion-guided biopsy and consider how additional incorporation of multivariable risk stratification might further improve prostate cancer diagnosis. RECENT FINDINGS: MpMRI has been proven advantageous over standard practice for biopsy-naïve men and men with previous biopsy in large prospective studies providing level 1b evidence. Upfront multivariable risk stratification followed by or combined with mpMRI further improves diagnostic accuracy. Regarding active surveillance, mpMRI in combination with fusion biopsy can support initial candidate selection and may help to monitor disease progression. mpMRI and fusion biopsy, however, do not spare failure and conflicting data exists to what extend (systematic) biopsies can be omitted. SUMMARY: Integration of mpMRI into the diagnostic pathway for prostate cancer is beneficial; yet more prospective and randomized data is needed to establish reliable procedure standards after mpMRI acquisition.

Multiparametric MRI and MRI/TRUS Fusion Guided Biopsy for the Diagnosis of Prostate Cancer.

Purpose of this chapter To demonstrate the timing, benefits, limitations and current controversies of multiparametric magnet resonance imaging (mpMRI) combined with fusion guided biopsy and consider how additional incorporation of multivariable risk stratification might further improve prostate cancer (PC) diagnosis. Recent findings MpMRI has been shown to add important information to the diagnostic pathway for prostate cancer. Fusion biopsy has also shown advantages in comparison to standard practice for biopsy-naïve men and men with previous biopsy in large prospective studies providing level 1b evidence. Adding upfront multivariable risk stratification followed by or combined with mpMRI diagnostic accuracy can further be improved. Regarding active surveillance (AS), mpMRI in combination with fusion biopsy can support initial candidate selection and may help to monitor disease progression. However, mpMRI and fusion biopsy are not without failure and conflicting data exists to what extend (systematic) biopsies can be omitted. Summary The integration of mpMRI into the diagnostic pathway for PC can add important information for further decision making, yet more prospective and randomized data is needed to establish reliable procedure standards after mpMRI acquisition.