Intra- and interreader reproducibility of PI-RADSv2: A multireader study

Evaluation of a deep learning prostate cancer detection system on biparametric MRI against radiological reading

OBJECTIVES

This study aims to evaluate a deep learning pipeline for detecting clinically significant prostate cancer (csPCa), defined as Gleason Grade Group (GGG) ≥ 2, using biparametric MRI (bpMRI) and compare its performance with radiological reading.

MATERIALS AND METHODS

The training dataset included 4381 bpMRI cases (3800 positive and 581 negative) across three continents, with 80% annotated using PI-RADS and 20% with Gleason Scores. The testing set comprised 328 cases from the PROSTATEx dataset, including 34% positive (GGG ≥ 2) and 66% negative cases. A 3D nnU-Net was trained on bpMRI for lesion detection, evaluated using histopathology-based annotations, and assessed with patient- and lesion-level metrics, along with lesion volume, and GGG. The algorithm was compared to non-expert radiologists using multi-parametric MRI (mpMRI).

RESULTS

The model achieved an AUC of 0.83 (95% CI: 0.80, 0.87). Lesion-level sensitivity was 0.85 (95% CI: 0.82, 0.94) at 0.5 False Positives per volume (FP/volume) and 0.88 (95% CI: 0.79, 0.92) at 1 FP/volume. Average Precision was 0.55 (95% CI: 0.46, 0.64). The model showed over 0.90 sensitivity for lesions larger than 650 mm³ and exceeded 0.85 across GGGs. It had higher true positive rates (TPRs) than radiologists equivalent FP rates, achieving TPRs of 0.93 and 0.79 compared to radiologists' 0.87 and 0.68 for PI-RADS ≥ 3 and PI-RADS ≥ 4 lesions (p ≤ 0.05).

CONCLUSION

The DL model showed strong performance in detecting csPCa on an independent test cohort, surpassing radiological interpretation and demonstrating AI's potential to improve diagnostic accuracy for non-expert radiologists. However, detecting small lesions remains challenging.

KEY POINTS

Question Current prostate cancer detection methods often do not involve non-expert radiologists, highlighting the need for more accurate deep learning approaches using biparametric MRI. Findings Our model outperforms radiologists significantly, showing consistent performance across Gleason Grade Groups and for medium to large lesions. Clinical relevance This AI model improves prostate detection accuracy in prostate imaging, serves as a benchmark with reference performance on a public dataset, and offers public PI-RADS annotations, enhancing transparency and facilitating further research and development.

Assessing the Performance of Artificial Intelligence Assistance for Prostate MRI: A Two-Center Study Involving Radiologists With Different Experience Levels

BACKGROUND

Artificial intelligence (AI) assistance may enhance radiologists' performance in detecting clinically significant prostate cancer (csPCa) on MRI. Further validation is needed for radiologists with different experiences.

PURPOSE

To assess the performance of experienced and less-experienced radiologists in detecting csPCa, with and without AI assistance.

STUDY TYPE

Retrospective.

POPULATION

Nine hundred patients who underwent prostate MRI and biopsy (median age 67 years; 356 with csPCa and 544 with non-csPCa).

FIELD STRENGTH/SEQUENCE

3-T and 1.5-T, diffusion-weighted imaging using a single-shot gradient echo-planar sequence, turbo spin echo T2-weighted image.

ASSESSMENT

CsPCa regions based on biopsy results served as the reference standard. Ten less-experienced (<500 prostate MRIs) and six experienced (>1000 prostate MRIs) radiologists reviewed each case twice using Prostate Imaging Reporting and Data System v2.1, with and without AI, separated by 4-week intervals. Cases were equally distributed among less-experienced radiologists, and 90 cases were randomly assigned to each experienced radiologist. Reading time and diagnostic confidence were assessed.

STATISTICAL TESTS

Area under the curve (AUC), sensitivity, specificity, reading time, and diagnostic confidence were compared using the DeLong test, Chi-squared test, Fisher exact test, or Wilcoxon rank-sum test between the two sessions. A P-value <0.05 was considered significant. Adjusting threshold using Bonferroni correction was performed for multiple comparisons.

RESULTS

For less-experienced radiologists, AI assistance significantly improved lesion-level sensitivity (0.78 vs. 0.88), sextant-level AUC (0.84 vs. 0.93), and patient-level AUC (0.84 vs. 0.89). For experienced radiologists, AI assistance only improved sextant-level AUC (0.82 vs. 0.91). AI assistance significantly reduced median reading time (250 s [interquartile range, IQR: 157, 402] vs. 130 s [IQR: 88, 209]) and increased diagnostic confidence (5 [IQR: 4, 5] vs. 5 [IQR: 4, 5]) irrespective of experience and enhanced consistency among experienced radiologists (Fleiss κ: 0.53 vs. 0.61).

DATA CONCLUSION

AI-assisted reading improves the performance of detecting csPCa on MRI, particularly for less-experienced radiologists.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Perspectives on technology: Prostate Imaging-Reporting and Data System (PI-RADS) interobserver variability

OBJECTIVES

To explore the topic of Prostate Imaging-Reporting and Data System (PI-RADS) interobserver variability, including a discussion of major sources, mitigation approaches, and future directions.

METHODS

A narrative review of PI-RADS interobserver variability.

RESULTS

PI-RADS was developed in 2012 to set technical standards for prostate magnetic resonance imaging (MRI), reduce interobserver variability at interpretation, and improve diagnostic accuracy in the MRI-directed diagnostic pathway for detection of clinically significant prostate cancer. While PI-RADS has been validated in selected research cohorts with prostate cancer imaging experts, subsequent prospective studies in routine clinical practice demonstrate wide variability in diagnostic performance. Radiologist and biopsy operator experience are the most important contributing drivers of high-quality care among multiple interrelated factors including variability in MRI hardware and technique, image quality, and population and patient-specific factors such as prostate cancer disease prevalence. Iterative improvements in PI-RADS have helped flatten the curve for novice readers and reduce variability. Innovations in image quality reporting, administrative and organisational workflows, and artificial intelligence hold promise in improving variability even further.

CONCLUSION

Continued research into PI-RADS is needed to facilitate benchmark creation, reader certification, and independent accreditation, which are systems-level interventions needed to uphold and maintain high-quality prostate MRI across entire populations.

AI-derived Tumor Volume from Multiparametric MRI and Outcomes in Localized Prostate Cancer

Background An artificial intelligence (AI)-based method for measuring intraprostatic tumor volume based on data from MRI may provide prognostic information. Purpose To evaluate whether the total volume of intraprostatic tumor from AI-generated segmentations (VAI) provides independent prognostic information in patients with localized prostate cancer treated with radiation therapy (RT) or radical prostatectomy (RP). Materials and Methods For this retrospective, single-center study (January 2021 to August 2023), patients with cT1-3N0M0 prostate cancer who underwent MRI and were treated with RT or RP were identified. Patients who underwent RT were randomly divided into cross-validation and test RT groups. An AI segmentation algorithm was trained to delineate Prostate Imaging Reporting and Data System (PI-RADS) 3-5 lesions in the cross-validation RT group before providing segmentations for the test RT and RP groups. Cox regression models were used to evaluate the association between VAI and time to metastasis and adjusted for clinical and radiologic factors for combined RT (ie, cross-validation RT and test RT) and RP groups. Areas under the receiver operating characteristic curve (AUCs) were calculated for VAI and National Comprehensive Cancer Network (NCCN) risk categorization for prediction of 5-year metastasis (RP group) and 7-year metastasis (combined RT group). Results Overall, 732 patients were included (combined RT group, 438 patients; RP group, 294 patients). Median ages were 68 years (IQR, 62-73 years) and 61 years (IQR, 56-66 years) for the combined RT group and the RP group, respectively. VAI was associated with metastasis in the combined RT group (median follow-up, 6.9 years; adjusted hazard ratio [AHR], 1.09 per milliliter increase; 95% CI: 1.04, 1.15; P = .001) and the RP group (median follow-up, 5.5 years; AHR, 1.22; 95% CI: 1.08, 1.39; P = .001). AUCs for 7-year metastasis for the combined RT group for VAI and NCCN risk category were 0.84 (95% CI: 0.74, 0.94) and 0.74 (95% CI: 0.80, 0.98), respectively (P = .02). Five-year AUCs for the RP group for VAI and NCCN risk category were 0.89 (95% CI: 0.80, 0.98) and 0.79 (95% CI: 0.64, 0.94), respectively (P = .25). Conclusion The volume of AI-segmented lesions was an independent, prognostic factor for localized prostate cancer. © RSNA, 2024 Supplemental material is available for this article.

The impact of integrating PRIMARY score or SUVmax with MRI-based risk models for the detection of clinically significant prostate cancer

PURPOSE

An MRI-based risk calculator (RC) has been recommended for diagnosing clinically significant prostate cancer (csPCa). PSMA PET/CT can detect lesions that are not visible on MRI, and the addition of PSMA PET/CT to MRI may improve diagnostic performance. The aim of this study was to incorporate the PRIMARY score or SUVmax derived from [68Ga]Ga-PSMA-11 PET/CT into the RC and compare these models with MRI-based RC to assess whether this can further reduce unnecessary biopsies.

METHODS

A total of 683 consecutive biopsy-naïve men who underwent both [68Ga]Ga-PSMA-11 PET/CT and MRI before biopsy were temporally divided into a development cohort (n = 552) and a temporal validation cohort (n = 131). Three logistic regression RCs were developed and compared: MRI-RC, MRI-SUVmax-RC and MRI-PRIMARY-RC. Discrimination, calibration, and clinical utility were evaluated. The primary outcome was the clinical utility of the risk calculators for detecting csPCa and reducing the number of negative biopsies.

RESULTS

The prevalence of csPCa was 47.5% (262/552) in the development cohort and 41.9% (55/131) in the temporal validation cohort. In the development cohort, the AUC of MRI-PRIMARY-RC was significantly higher than that of MRI-RC (0.924 vs. 0.868, p < 0.001) and MRI-SUVmax-RC (0.924 vs. 0.904, p = 0.002). In the temporal validation cohort, MRI-PRIMARY-RC also showed the best discriminative ability with an AUC of 0.921 (95% CI: 0.873-0.969). Bootstrapped calibration curves revealed that the model fit was acceptable. MRI-PRIMARY-RC exhibited near-perfect calibration within the range of 0-40%. DCA showed that MRI-PRIMARY-RC had the greatest net benefit for detecting csPCa compared with MRI-RC and MRI-SUVmax-RC at a risk threshold of 5-40% for csPCa in both the development and validation cohorts.

CONCLUSION

The addition of the PRIMARY score to MRI-based multivariable model improved the accuracy of risk stratification prior to biopsy. Our novel MRI-PRIMARY prediction model is a promising approach for reducing unnecessary biopsies and improving the early detection of csPCa.

The Role of Radiomics in the Prediction of Clinically Significant Prostate Cancer in the PI-RADS v2 and v2.1 Era: A Systematic Review

Early detection of clinically significant prostate cancer (csPCa) has substantially improved with the latest PI-RADS versions. However, there is still an overdiagnosis of indolent lesions (iPCa), and radiomics has emerged as a potential solution. The aim of this systematic review is to evaluate the role of handcrafted and deep radiomics in differentiating lesions with csPCa from those with iPCa and benign lesions on prostate MRI assessed with PI-RADS v2 and/or 2.1. The literature search was conducted in PubMed, Cochrane, and Web of Science databases to select relevant studies. Quality assessment was carried out with Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2), Radiomic Quality Score (RQS), and Checklist for Artificial Intelligence in Medical Imaging (CLAIM) tools. A total of 14 studies were deemed as relevant from 411 publications. The results highlighted a good performance of handcrafted and deep radiomics methods for csPCa detection, but without significant differences compared to radiologists (PI-RADS) in the few studies in which it was assessed. Moreover, heterogeneity and restrictions were found in the studies and quality analysis, which might induce bias. Future studies should tackle these problems to encourage clinical applicability. Prospective studies and comparison with radiologists (PI-RADS) are needed to better understand its potential.

Fully Automated Deep Learning Model to Detect Clinically Significant Prostate Cancer at MRI

Background Multiparametric MRI can help identify clinically significant prostate cancer (csPCa) (Gleason score ≥7) but is limited by reader experience and interobserver variability. In contrast, deep learning (DL) produces deterministic outputs. Purpose To develop a DL model to predict the presence of csPCa by using patient-level labels without information about tumor location and to compare its performance with that of radiologists. Materials and Methods Data from patients without known csPCa who underwent MRI from January 2017 to December 2019 at one of multiple sites of a single academic institution were retrospectively reviewed. A convolutional neural network was trained to predict csPCa from T2-weighted images, diffusion-weighted images, apparent diffusion coefficient maps, and T1-weighted contrast-enhanced images. The reference standard was pathologic diagnosis. Radiologist performance was evaluated as follows: Radiology reports were used for the internal test set, and four radiologists' PI-RADS ratings were used for the external (ProstateX) test set. The performance was compared using areas under the receiver operating characteristic curves (AUCs) and the DeLong test. Gradient-weighted class activation maps (Grad-CAMs) were used to show tumor localization. Results Among 5735 examinations in 5215 patients (mean age, 66 years ± 8 [SD]; all male), 1514 examinations (1454 patients) showed csPCa. In the internal test set (400 examinations), the AUC was 0.89 and 0.89 for the DL classifier and radiologists, respectively (P = .88). In the external test set (204 examinations), the AUC was 0.86 and 0.84 for the DL classifier and radiologists, respectively (P = .68). DL classifier plus radiologists had an AUC of 0.89 (P < .001). Grad-CAMs demonstrated activation over the csPCa lesion in 35 of 38 and 56 of 58 true-positive examinations in internal and external test sets, respectively. Conclusion The performance of a DL model was not different from that of radiologists in the detection of csPCa at MRI, and Grad-CAMs localized the tumor. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Johnson and Chandarana in this issue.

Avoiding unnecessary biopsy: the combination of PRIMARY score with prostate-specific antigen density for prostate biopsy decision

BACKGROUND

Avoiding unnecessary biopsies for men with suspected prostate cancer remains a clinical priority. The recently proposed PRIMARY score improves diagnostic accuracy in detecting clinically significant prostate cancer (csPCa). The aim of this study was to determine the best strategy combining PRIMARY score or MRI reporting scores (Prostate Imaging Reporting and Data System [PI-RADS]) with prostate-specific antigen density (PSAD) for prostate biopsy decision making.

METHODS

A retrospective analysis of 343 patients who underwent both 68Ga-PSMA PET/CT and MRI before prostate biopsy was performed. PSA was restricted to <20 ng/ml. Different biopsy strategies were developed and compared based on PRIMARY score or PI-RADS with PSAD thresholds. Decision curve analysis (DCA) was plotted to define the optimal biopsy strategy.

RESULTS

The prevalence of csPCa was 41.1% (141/343). According to DCA, the strategies of PRIMARY score +PSAD (strategy #1, strategy #2, strategy #6) had a higher net benefit than the strategies of PI-RADS + PSAD at the risk threshold of 8-20%. The best diagnostic strategy was strategy #1 (PRIMARY score 4-5 or PSAD ≥ 0.20), which avoided 38.2% biopsy procedures while missed 9.2% of csPCa cases. From a clinical perspective, strategies with a lower risk of missing csPCa were strategy #2 (PRIMARY score ≥4 or PSAD ≥ 0.15), which avoided 28.6% biopsies while missed 5.7% of csPCa cases, or strategy #6 (PRIMARY score≥3 or PSAD ≥ 0.15), which avoided 20.7% biopsies while missed only 3.5% of csPCa cases. The limitations of the study were the retrospective single-center nature.

CONCLUSIONS

The combination of PRIMARY score +PSAD allows individualized decisions to avoid unnecessary biopsy, outperforming the strategies of PI-RADS + PSAD. Further prospective trials are needed to validate these findings.

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

Background Multiparametric MRI (mpMRI) improves prostate cancer (PCa) detection compared with systematic biopsy, but its interpretation is prone to interreader variation, which results in performance inconsistency. Artificial intelligence (AI) models can assist in mpMRI interpretation, but large training data sets and extensive model testing are required. Purpose To evaluate a biparametric MRI AI algorithm for intraprostatic lesion detection and segmentation and to compare its performance with radiologist readings and biopsy results. Materials and Methods This secondary analysis of a prospective registry included consecutive patients with suspected or known PCa who underwent mpMRI, US-guided systematic biopsy, or combined systematic and MRI/US fusion-guided biopsy between April 2019 and September 2022. All lesions were prospectively evaluated using Prostate Imaging Reporting and Data System version 2.1. The lesion- and participant-level performance of a previously developed cascaded deep learning algorithm was compared with histopathologic outcomes and radiologist readings using sensitivity, positive predictive value (PPV), and Dice similarity coefficient (DSC). Results A total of 658 male participants (median age, 67 years [IQR, 61-71 years]) with 1029 MRI-visible lesions were included. At histopathologic analysis, 45% (294 of 658) of participants had lesions of International Society of Urological Pathology (ISUP) grade group (GG) 2 or higher. The algorithm identified 96% (282 of 294; 95% CI: 94%, 98%) of all participants with clinically significant PCa, whereas the radiologist identified 98% (287 of 294; 95% CI: 96%, 99%; P = .23). The algorithm identified 84% (103 of 122), 96% (152 of 159), 96% (47 of 49), 95% (38 of 40), and 98% (45 of 46) of participants with ISUP GG 1, 2, 3, 4, and 5 lesions, respectively. In the lesion-level analysis using radiologist ground truth, the detection sensitivity was 55% (569 of 1029; 95% CI: 52%, 58%), and the PPV was 57% (535 of 934; 95% CI: 54%, 61%). The mean number of false-positive lesions per participant was 0.61 (range, 0-3). The lesion segmentation DSC was 0.29. Conclusion The AI algorithm detected cancer-suspicious lesions on biparametric MRI scans with a performance comparable to that of an experienced radiologist. Moreover, the algorithm reliably predicted clinically significant lesions at histopathologic examination. ClinicalTrials.gov Identifier: NCT03354416 © RSNA, 2024 Supplemental material is available for this article.

Development and validation of a clinical-radiomics model for prediction of prostate cancer: a multicenter study

PURPOSE

To develop an early diagnosis model of prostate cancer based on clinical-radiomics to improve the accuracy of imaging diagnosis of prostate cancer.

METHODS

The multicenter study enrolled a total of 449 patients with prostate cancer from December 2017 to January 2022. We retrospectively collected information from 342 patients who underwent prostate biopsy at Minhang Hospital. We extracted T2WI images through 3D-Slice, and used mask tools to mark the prostate area manually. The radiomics features were extracted by Python using the "Pyradiomics" module. Least Absolute Shrinkage and Selection Operator (LASSO) regression was used for data dimensionality reduction and feature selection, and the radiomics score was calculated according to the correlation coefficients. Multivariate logistic regression analysis was used to develop predictive models. We incorporated the radiomics score, PI-RADS, and clinical features, and this was presented as a nomogram. The model was validated using a cohort of 107 patients from the Xuhui Hospital.

RESULTS

In total, 110 effective radiomics features were extracted. Finally, 9 features were significantly associated with the diagnosis of prostate cancer, from which we calculated the radiomics score. The predictors contained in the individualized prediction nomogram included age, fPSA/tPSA, PI-RADS, and radiomics score. The clinical-radiomics model showed good discrimination in the validation cohort (C-index = 0.88).

CONCLUSION

This study presents a clinical-radiomics model that incorporates age, fPSA/PSA, PI-RADS, and radiomics score, which can be conveniently used to facilitate individualized prediction of prostate cancer before prostate biopsy.

Using deep learning to optimize the prostate MRI protocol by assessing the diagnostic efficacy of MRI sequences

PURPOSE

To explore diagnostic deep learning for optimizing the prostate MRI protocol by assessing the diagnostic efficacy of MRI sequences.

METHOD

This retrospective study included 840 patients with a biparametric prostate MRI scan. The MRI protocol included a T2-weighted image, three DWI sequences (b50, b400, and b800 s/mm2), a calculated ADC map, and a calculated b1400 sequence. Two accelerated MRI protocols were simulated, using only two acquired b-values to calculate the ADC and b1400. Deep learning models were trained to detect prostate cancer lesions on accelerated and full protocols. The diagnostic performances of the protocols were compared on the patient-level with the area under the receiver operating characteristic (AUROC), using DeLong's test, and on the lesion-level with the partial area under the free response operating characteristic (pAUFROC), using a permutation test. Validation of the results was performed among expert radiologists.

RESULTS

No significant differences in diagnostic performance were found between the accelerated protocols and the full bpMRI baseline. Omitting b800 reduced 53% DWI scan time, with a performance difference of + 0.01 AUROC (p = 0.20) and -0.03 pAUFROC (p = 0.45). Omitting b400 reduced 32% DWI scan time, with a performance difference of -0.01 AUROC (p = 0.65) and + 0.01 pAUFROC (p = 0.73). Multiple expert radiologists underlined the findings.

CONCLUSIONS

This study shows that deep learning can assess the diagnostic efficacy of MRI sequences by comparing prostate MRI protocols on diagnostic accuracy. Omitting either the b400 or the b800 DWI sequence can optimize the prostate MRI protocol by reducing scan time without compromising diagnostic quality.

ParaCM-PNet: A CNN-tokenized MLP combined parallel dual pyramid network for prostate and prostate cancer segmentation in MRI

The precise prostate gland and prostate cancer (PCa) segmentations enable the fusion of magnetic resonance imaging (MRI) and ultrasound imaging (US) to guide robotic prostate biopsy systems. This precise segmentation, applied to preoperative MRI images, is crucial for accurate image registration and automatic localization of the biopsy target. Nevertheless, describing local prostate lesions in MRI remains a challenging and time-consuming task, even for experienced physicians. Therefore, this research work develops a parallel dual-pyramid network that combines convolutional neural networks (CNN) and tokenized multi-layer perceptron (MLP) for automatic segmentation of the prostate gland and clinically significant PCa (csPCa) in MRI. The proposed network consists of two stages. The first stage focuses on prostate segmentation, while the second stage uses a prior partition from a previous stage to detect the cancerous regions. Both stages share a similar network architecture, combining CNN and tokenized MLP as the feature extraction backbone to creating a pyramid-structured network for feature encoding and decoding. By employing CNN layers of different scales, the network generates scale-aware local semantic features, which are integrated into feature maps and inputted into an MLP layer from a global perspective. This facilitates the complementarity between local and global information, capturing richer semantic features. Additionally, the network incorporates an interactive hybrid attention module to enhance the perception of the target area. Experimental results demonstrate the superiority of the proposed network over other state-of-the-art image segmentation methods for segmenting the prostate gland and csPCa tissue in MRI images.

The Novel Green Learning Artificial Intelligence for Prostate Cancer Imaging: A Balanced Alternative to Deep Learning and Radiomics

The application of artificial intelligence (AI) on prostate magnetic resonance imaging (MRI) has shown promising results. Several AI systems have been developed to automatically analyze prostate MRI for segmentation, cancer detection, and region of interest characterization, thereby assisting clinicians in their decision-making process. Deep learning, the current trend in imaging AI, has limitations including the lack of transparency "black box", large data processing, and excessive energy consumption. In this narrative review, the authors provide an overview of the recent advances in AI for prostate cancer diagnosis and introduce their next-generation AI model, Green Learning, as a promising solution.

Interactive training workshop to improve prostate mpMRI knowledge: results from the ESOR Nicholas Gourtsoyiannis teaching fellowship

PURPOSE

Prostate MRI is established for the investigation of patients presenting with suspected early prostate cancer. Outcomes are dependent on both image quality and interpretation. This study assessed the impact of an educational intervention on participants' theoretical knowledge of the technique.

METHODS

Eighty-one clinicians from two centers with varying experience in prostate MRI participated. Baseline knowledge was assessed with 10 written and image-based multiple-choice questions (MCQs) prior to a course including didactic lectures and hands-on interactive workshops on prostate MRI interpretation. Post-course, participants completed a second 10-question MCQ test, matched by format, themes, and difficulty, to assess for any improvement in knowledge and performance. Results were assessed using the Wilcoxon rank sum test, and the Wilcoxon signed-rank test for paired data.

RESULTS

Thirty-nine participants, including 25/49 (51.0%) and 14/32 (43.8%) at each center completed both assessments, with their results used for subsequent evaluation. Overall, there was a significant improvement from pre- (4.92 ± 2.41) to post-course scores (6.77 ± 1.46), p < 0.001 and at both Copenhagen (5.92 ± 2.25 to 7.36 ± 1.25) and Toronto (3.14 ± 1.51 to 5.71 ± 1.20); p = 0.005 and p = 0.002, respectively. Participants with no prostate MRI experience showed the greatest improvement (3.77 ± 1.97 to 6.18 ± 1.5, p < 0.001), followed by intermediate level (< 500 MRIs reported) experience (6.18 ± 1.99 to 7.46 ± 1.13, p = 0.058), then advanced (> 500 MRIs reported) experience (6.83 ± 2.48 to 7.67 ± 0.82, p = 0.339).

CONCLUSIONS

A dedicated prostate MRI teaching course combining didactic lectures and hands-on workshops significantly improved short-term theoretical knowledge of the technique for clinicians with differing levels of experience.

CRITICAL RELEVANCE STATEMENT

A dedicated teaching course significantly improved theoretical knowledge of the technique particularly for clinicians with less reporting experience and a lower baseline knowledge. The multiple-choice questions format mapped improved performance and may be considered as part of future MRI certification initiatives.

KEY POINTS

• Prostate MRI knowledge is important for image interpretation and optimizing acquisition sequences. • A dedicated teaching course significantly improved theoretical knowledge of the technique. • Improved performance was more apparent in clinicians with less reporting experience and a lower baseline knowledge.

Using an artificial intelligence model to detect and localize visible clinically significant prostate cancer in prostate magnetic resonance imaging: a multicenter external validation study

Background

An increasing number of patients with suspected clinically significant prostate cancer (csPCa) are undergoing prostate multiparametric magnetic resonance imaging (mpMRI). The role of artificial intelligence (AI) algorithms in interpreting prostate mpMRI needs to be tested with multicenter external data. This study aimed to investigate the diagnostic efficacy of an AI model in detecting and localizing visible csPCa on mpMRI a multicenter external data set.

Methods

The data of 2,105 patients suspected of having prostate cancer from four hospitals were retrospectively collected to develop an AI model to detect and localize suspicious csPCa. The lesions were annotated based on pathology records by two radiologists. Diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) values were used as the input for the three-dimensional U-Net framework. Subsequently, the model was validated using an external data set comprising the data of 557 patients from three hospitals. Sensitivity, specificity, and accuracy were employed to evaluate the diagnostic efficacy of the model.

Results

At the lesion level, the model had a sensitivity of 0.654. At the overall sextant level, the model had a sensitivity, specificity, and accuracy of 0.846, 0.884, and 0.874, respectively. At the patient level, the model had a sensitivity, specificity, and accuracy of 0.943, 0.776, and 0.849, respectively. The AI-predicted accuracy for the csPCa patients (231/245, 0.943) was significantly higher than that for the non-csPCa patients (242/312, 0.776) (P<0.001). The lesion number and tumor volume were greater in the correctly diagnosed patients than the incorrectly diagnosed patients (both P<0.001). Among the positive patients, those with lower average ADC values had a higher rate of correct diagnosis than those with higher average ADC values (P=0.01).

Conclusions

The AI model exhibited acceptable accuracy in detecting and localizing visible csPCa at the patient and sextant levels. However, further improvements need to be made to enhance the sensitivity of the model at the lesion level.

Update on Optimization of Prostate MR Imaging Technique and Image Quality

Prostate MR imaging quality has improved dramatically over recent times, driven by advances in hardware, software, and improved functional imaging techniques. MRI now plays a key role in prostate cancer diagnostic work-up, but outcomes of the MRI-directed pathway are heavily dependent on image quality and optimization. MR sequences can be affected by patient-related degradations relating to motion and susceptibility artifacts which may enable only partial mitigation. In this Review, we explore issues relating to prostate MRI acquisition and interpretation, mitigation strategies at a patient and scanner level, PI-QUAL reporting, and future directions in image quality, including artificial intelligence solutions.

Prostate Cancer Local Staging with Magnetic Resonance Imaging

Accurate determination of the local stage of prostate cancer is crucial for treatment planning and prognosis. The primary objective of local staging is to distinguish between organ-confined and locally advanced disease, with the latter carrying a worse clinical prognosis. The presence of locally advanced disease features of prostate cancer, such as extra-prostatic extension, seminal vesicle invasion, and positive surgical margin, can impact the choice of treatment. Over the past decade, multiparametric MRI (mpMRI) has become the preferred imaging modality for the local staging of prostate cancer and has been shown to provide accurate information on the location and extent of disease. It has demonstrated superior performance compared to staging based on traditional clinical nomograms. Despite being a relatively new technique, mpMRI has garnered considerable attention and ongoing investigations. Therefore, in this review, we will discuss the current use of mpMRI on prostate cancer local staging.

Assessing the Feasibility of Using Artificial Intelligence-Segmented Dominant Intraprostatic Lesion for Focal Intraprostatic Boost With External Beam Radiation Therapy

PURPOSE

The delineation of dominant intraprostatic gross tumor volumes (GTVs) on multiparametric magnetic resonance imaging (mpMRI) can be subject to interobserver variability. We evaluated whether deep learning artificial intelligence (AI)-segmented GTVs can provide a similar degree of intraprostatic boosting with external beam radiation therapy (EBRT) as radiation oncologist (RO)-delineated GTVs.

METHODS AND MATERIALS

We identified 124 patients who underwent mpMRI followed by EBRT between 2010 and 2013. A reference GTV was delineated by an RO and approved by a board-certified radiologist. We trained an AI algorithm for GTV delineation on 89 patients, and tested the algorithm on 35 patients, each with at least 1 PI-RADS (Prostate Imaging Reporting and Data System) 4 or 5 lesion (46 total lesions). We then asked 5 additional ROs to independently delineate GTVs on the test set. We compared lesion detectability and geometric accuracy of the GTVs from AI and 5 ROs against the reference GTV. Then, we generated EBRT plans (77 Gy prostate) that boosted each observer-specific GTV to 95 Gy. We compared reference GTV dose (D98%) across observers using a mixed-effects model.

RESULTS

On a lesion level, AI GTV exhibited a sensitivity of 82.6% and positive predictive value of 86.4%. Respective ranges among the 5 RO GTVs were 84.8% to 95.7% and 95.1% to 100.0%. Among 30 GTVs mutually identified by all observers, no significant differences in Dice coefficient were detected between AI and any of the 5 ROs. Across all patients, only 2 of 5 ROs had a reference GTV D98% that significantly differed from that of AI by 2.56 Gy (P = .02) and 3.20 Gy (P = .003). The presence of false-negative (-5.97 Gy; P < .001) but not false-positive (P = .24) lesions was associated with reference GTV D98%.

CONCLUSIONS

AI-segmented GTVs demonstrate potential for intraprostatic boosting, although the degree of boosting may be adversely affected by false-negative lesions. Prospective review of AI-segmented GTVs remains essential.

Enhancing Prostate Cancer Diagnosis with a Novel Artificial Intelligence-Based Web Application: Synergizing Deep Learning Models, Multimodal Data, and Insights from Usability Study with Pathologists

Prostate cancer remains a significant cause of male cancer mortality in the United States, with an estimated 288,300 new cases in 2023. Accurate grading of prostate cancer is crucial for ascertaining disease severity and shaping treatment strategies. Modern deep learning techniques show promise in grading biopsies, but there is a gap in integrating these advances into clinical practice. Our web platform tackles this challenge by integrating human expertise with AI-driven grading, incorporating diverse data sources. We gathered feedback from four pathologists and one medical practitioner to assess usability and real-world alignment through a survey and the NASA TLX Usability Test. Notably, 60% of users found it easy to navigate, rating it 5.5 out of 7 for ease of understanding. Users appreciated self-explanatory information in popup tabs. For ease of use, all users favored the detailed summary tab, rating it 6.5 out of 7. While 80% felt patient demographics beyond age were unnecessary, high-resolution biopsy images were deemed vital. Acceptability was high, with all users willing to adopt the app, and some believed it could reduce workload. The NASA TLX Usability Test indicated a low-moderate perceived workload, suggesting room for improved explanations and data visualization.

Clinical study of multifactorial diagnosis in prostate biopsy

PURPOSE

To study the feasibility of using an artificial intelligence (AI) algorithm for the diagnosis of clinically significant prostate cancer (csPCa) on multiparametric MRI (mpMRI) in combination with conventional clinical information.

METHODS

A retrospective study cohort with 505 patients was collected, with complete information on age (≤60, 60-80, and >80 years), PSA (≤4, 4-10, and >10 ng/dL), and pathology results. The patients with ISUP group >2 were classified as csPCa, and the patients with ISUP = 1 or no evidence of prostate cancer were classified as non-csPCa. The diagnosis of mpMRI was made by experienced radiologists following the prostate imaging reporting and data system (PIRADS ≤ 2, PIRADS = 3, and PIRADS > 3). The mpMRI images were processed by a homemade AI algorithm, and the AI results were obtained as positive or negative for csPCa. Two logistic regression models were fitted, with pathological findings as the dependent variable, that is, a conventional model and an AI model. The conventional model used age, PSA, and PIRADS as the independent variables. The AI model took the AI result and the abovementioned clinical information as the independent variables. The predicted probability of the patients from the conventional model and the AI model were used to test the prediction efficacy of the models. The DeLong test was performed to compare differences in the area under the receiver operating characteristic (ROC) area under the curve (AUC) between the conventional model and the AI model.

RESULTS

In total, 505 patients were included in the study; 280 were diagnosed with csPCa, and 225 were non-csPCa. The median age was 72.0 (67.0, 76.0) years, with a median PSA value of 13.0 (7.46, 27.5) ng/dL. Statically significant differences were found in age, PSA, PIRADS score and AI results between the csPCa and non-csPCa groups (all p < 0.001). In the multivariable regression models, all the variables were independently associated with csPCa. The conventional model (R2  = 0.361) and the AI model (R2  = 0.474) were compared with analysis of variance (ANOVA) and showed statistically significant differences (χ2  = 63.695, p < 0.001). The AUC of the ROC curve for the conventional model was 0.782 (95% confidence interval [CI]: 0.742-0.823), which was less than the AUC of the AI model with statistical significance (0.849 [95% CI: 0.815-0.883], p < 0.001).

CONCLUSION

In combination with routine clinical information, such as age, PSA, and PIRADS category, adding information from the AI algorithm based on mpMRI could improve the diagnosis of csPCa.

Prostate magnetic resonance imaging and the value of experience: An intrareader variability study

Objective

To measure the intraobserver concordance of an experienced genitourinary radiologist reporting of multiparametric magnetic resonance imaging of the prostate (mpMRIp) scans over time.

Methods

An experienced genitourinary radiologist re-reported his original 100 consecutive mpMRIp scans using Prostate Imaging-Reporting and Data System version 2 (PI-RADS v2) after 5 years of further experience comprising >1000 scans. Intraobserver agreement was measured using Cohen's kappa. Sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and accuracy were calculated, and comparison of sensitivity was performed using McNemar's test.

Results

Ninety-six mpMRIp scans were included in our final analysis. Of the 96 patients, 53 (55.2%) patients underwent subsequent biopsy (n=43) or prostatectomy (n=15), with 73 lesions targeted. Moderate agreement (Cohen's kappa 0.55) was seen in the number of lesions identified at initial reporting and on re-reading (81 vs. 39 total lesions; and 71 vs. 37 number of PI-RADS ≥3 lesions). For clinically significant prostate cancer, re-reading demonstrated an increase in specificity (from 43% to 89%) and PPV (from 62% to 87%), but a decrease in sensitivity (from 94% to 72%, p=0.01) and NPV (from 89% to 77%).

Conclusion

The intraobserver agreement for a novice to experienced radiologist reporting mpMRIp using PI-RADS v2 is moderate. Reduced sensitivity is off-set by improved specificity and PPV, which validate mpMRIp as a gold standard for prebiopsy screening.

Cancer detection and complications of transperineal prostate biopsy with antibiotics when indicated

OBJECTIVES

To describe the prostate cancer (PCa) detection rate, including clinically significant prostate cancer (csPCa), in a large cohort of patients who underwent transperineal ultrasonography-guided systematic prostate biopsy (TPB-US) using a probe-mounted transperineal access system, with magnetic resonance imaging (MRI) cognitive fusion in case of a Prostate Imaging-Reporting and Data System grade 3-5 lesion, under local anaesthesia in an outpatient setting. Additionally, to compare the incidence of procedure-related complications with a cohort of patients undergoing transrectal ultrasonography-guided (TRB-US) and transrectal MRI-guided biopsies (TRB-MRI).

PATIENTS AND METHODS

This was an observational cohort study in men who underwent TPB-US prostate biopsy in a large teaching hospital. For each participant, prostate-specific antigen level, clinical tumour stage, prostate volume, MRI parameters, number of (targeted) prostate biopsies, biopsy International Society of Uropathology (ISUP) grade and procedure-related complications were assessed. csPCa was defined as ISUP grade ≥2. Antibiotic prophylaxis was only given in those with an increased risk of urinary tract infection.

RESULTS

A total of 1288 TPB-US procedures were evaluated. The overall detection rate for PCa in biopsy-naive patients was 73%, and for csPCa it was 63%. The incidence of hospitalization was 1% in TPB-US (13/1288), compared to 4% in TRB-US (8/214) and 3% in TRB-MRI (7/219; P = 0.002).

CONCLUSIONS

Contemporary combined systematic and target TPB-US with MRI cognitive fusion is easy to perform in an outpatient setting, with a high detection rate of csPCa and a low incidence of procedure-related complications.

A Brief Review of Artificial Intelligence in Genitourinary Oncological Imaging

Genitourinary (GU) system is among the most commonly involved malignancy sites in the human body. Imaging plays a crucial role not only in diagnosis of cancer but also in disease management and its prognosis. However, interpretation of conventional imaging methods such as CT or MR imaging (MRI) usually demonstrates variability across different readers and institutions. Artificial intelligence (AI) has emerged as a promising technology that could improve the patient care by providing helpful input to human readers through lesion detection algorithms and lesion classification systems. Moreover, the robustness of these models may be valuable in automating time-consuming tasks such as organ and lesion segmentations. Herein, we review the current state of imaging and existing challenges in GU malignancies, particularly for cancers of prostate, kidney and bladder; and briefly summarize the recent AI-based solutions to these challenges.

Anatomically guided self-adapting deep neural network for clinically significant prostate cancer detection on bi-parametric MRI: a multi-center study

OBJECTIVE

To evaluate the effectiveness of a self-adapting deep network, trained on large-scale bi-parametric MRI data, in detecting clinically significant prostate cancer (csPCa) in external multi-center data from men of diverse demographics; to investigate the advantages of transfer learning.

METHODS

We used two samples: (i) Publicly available multi-center and multi-vendor Prostate Imaging: Cancer AI (PI-CAI) training data, consisting of 1500 bi-parametric MRI scans, along with its unseen validation and testing samples; (ii) In-house multi-center testing and transfer learning data, comprising 1036 and 200 bi-parametric MRI scans. We trained a self-adapting 3D nnU-Net model using probabilistic prostate masks on the PI-CAI data and evaluated its performance on the hidden validation and testing samples and the in-house data with and without transfer learning. We used the area under the receiver operating characteristic (AUROC) curve to evaluate patient-level performance in detecting csPCa.

RESULTS

The PI-CAI training data had 425 scans with csPCa, while the in-house testing and fine-tuning data had 288 and 50 scans with csPCa, respectively. The nnU-Net model achieved an AUROC of 0.888 and 0.889 on the hidden validation and testing data. The model performed with an AUROC of 0.886 on the in-house testing data, with a slight decrease in performance to 0.870 using transfer learning.

CONCLUSIONS

The state-of-the-art deep learning method using prostate masks trained on large-scale bi-parametric MRI data provides high performance in detecting csPCa in internal and external testing data with different characteristics, demonstrating the robustness and generalizability of deep learning within and across datasets.

CLINICAL RELEVANCE STATEMENT

A self-adapting deep network, utilizing prostate masks and trained on large-scale bi-parametric MRI data, is effective in accurately detecting clinically significant prostate cancer across diverse datasets, highlighting the potential of deep learning methods for improving prostate cancer detection in clinical practice.

Research progress on deep learning in magnetic resonance imaging-based diagnosis and treatment of prostate cancer: a review on the current status and perspectives

Multiparametric magnetic resonance imaging (mpMRI) has emerged as a first-line screening and diagnostic tool for prostate cancer, aiding in treatment selection and noninvasive radiotherapy guidance. However, the manual interpretation of MRI data is challenging and time-consuming, which may impact sensitivity and specificity. With recent technological advances, artificial intelligence (AI) in the form of computer-aided diagnosis (CAD) based on MRI data has been applied to prostate cancer diagnosis and treatment. Among AI techniques, deep learning involving convolutional neural networks contributes to detection, segmentation, scoring, grading, and prognostic evaluation of prostate cancer. CAD systems have automatic operation, rapid processing, and accuracy, incorporating multiple sequences of multiparametric MRI data of the prostate gland into the deep learning model. Thus, they have become a research direction of great interest, especially in smart healthcare. This review highlights the current progress of deep learning technology in MRI-based diagnosis and treatment of prostate cancer. The key elements of deep learning-based MRI image processing in CAD systems and radiotherapy of prostate cancer are briefly described, making it understandable not only for radiologists but also for general physicians without specialized imaging interpretation training. Deep learning technology enables lesion identification, detection, and segmentation, grading and scoring of prostate cancer, and prediction of postoperative recurrence and prognostic outcomes. The diagnostic accuracy of deep learning can be improved by optimizing models and algorithms, expanding medical database resources, and combining multi-omics data and comprehensive analysis of various morphological data. Deep learning has the potential to become the key diagnostic method in prostate cancer diagnosis and treatment in the future.

Post-acquisition water-signal removal in 3D water-unsuppressed 1 H-MR spectroscopic imaging of the prostate

PURPOSE

To develop a robust processing procedure of raw signals from water-unsuppressed MRSI of the prostate for the mapping of absolute tissue concentrations of metabolites.

METHODS

Water-unsuppressed 3D MRSI data were acquired from a phantom, from healthy volunteers, and a patient with prostate cancer. Signal processing included sequential computation of the modulus of the FID to remove water sidebands, a Hilbert transformation, and k-space Hamming filtering. For the removal of the water signal, we compared Löwner tensor-based blind source separation (BSS) and Hankel Lanczos singular value decomposition techniques. Absolute metabolite levels were quantified with LCModel and the results were statistically analyzed to compare the water removal methods and conventional water-suppressed MRSI.

RESULTS

The post-processing algorithms successfully removed the water signal and its sidebands without affecting metabolite signals. The best water removal performance was achieved by Löwner tensor-based BSS. Absolute tissue concentrations of citrate in the peripheral zone derived from water-suppressed and unsuppressed ¹ H MRSI were the same and as expected from the known physiology of the healthy prostate. Maps for citrate and choline from water-unsuppressed 3D ¹ H-MRSI of the prostate showed expected spatial variations in metabolite levels.

CONCLUSION

We developed a robust relatively simple post-processing method of water-unsuppressed MRSI of the prostate to remove the water signal. Absolute quantification using the water signal, originating from the same location as the metabolite signals, avoids the acquisition of additional reference data.

A multicenter study of artificial intelligence-aided software for detecting visible clinically significant prostate cancer on mpMRI

BACKGROUND

AI-based software may improve the performance of radiologists when detecting clinically significant prostate cancer (csPCa). This study aims to compare the performance of radiologists in detecting MRI-visible csPCa on MRI with and without AI-based software.

MATERIALS AND METHODS

In total, 480 multiparametric MRI (mpMRI) images were retrospectively collected from eleven different MR devices, with 349 csPCa lesions in 180 (37.5%) cases. The csPCa areas were annotated based on pathology. Sixteen radiologists from four hospitals participated in reading. Each radiologist was randomly assigned to 30 cases and diagnosed twice. Half cases were interpreted without AI, and the other half were interpreted with AI. After four weeks, the cases were read again in switched mode. The mean diagnostic performance was compared using sensitivity and specificity on lesion level and patient level. The median reading time and diagnostic confidence were assessed.

RESULTS

On lesion level, AI-aided improved the sensitivity from 40.1% to 59.0% (18.9% increased; 95% confidence interval (CI) [11.5, 26.1]; p < .001). On patient level, AI-aided improved the specificity from 57.7 to 71.7% (14.0% increase, 95% CI [6.4, 21.4]; p < .001) while preserving the sensitivity (88.3% vs. 93.9%, p = 0.06). AI-aided reduced the median reading time of one case by 56.3% from 423 to 185 s (238-s decrease, 95% CI [219, 260]; p < .001), and the median diagnostic confidence score was increased by 10.3% from 3.9 to 4.3 (0.4-score increase, 95% CI [0.3, 0.5]; p < .001).

CONCLUSIONS

AI software improves the performance of radiologists by reducing false positive detection of prostate cancer patients and also improving reading times and diagnostic confidence.

CLINICAL RELEVANCE STATEMENT

This study involves the process of data collection, randomization and crossover reading procedure.

PI-RADSAI: introducing a new human-in-the-loop AI model for prostate cancer diagnosis based on MRI

BACKGROUND

This study aims to develop and validate an artificial intelligence (AI)-aided Prostate Imaging Reporting and Data System (PI-RADSAI) for prostate cancer (PCa) diagnosis based on MRI.

METHODS

The deidentified MRI data of 1540 biopsy-naïve patients were collected from four centres. PI-RADSAI is a two-stage, human-in-the-loop AI capable of emulating the diagnostic acumen of subspecialists for PCa on MRI. The first stage uses a UNet-Seg model to detect and segment biopsy-candidate prostate lesions, whereas the second stage leverages UNet-Seg segmentation is trained specifically with subspecialist' knowledge-guided 3D-Resnet to achieve an automatic AI-aided diagnosis for PCa.

RESULTS

In the independent test set, UNet-Seg identified 87.2% (628/720) of target lesions, with a Dice score of 44.9% (range, 22.8-60.2%) in segmenting lesion contours. In the ablation experiment, the model trained with the data from three centres was superior (kappa coefficient, 0.716 vs. 0.531) to that trained with single-centre data. In the internal and external tests, the triple-centre PI-RADSAI model achieved an overall agreement of 58.4% (188/322) and 60.1% (92/153) with a referential subspecialist in scoring target lesions; when one-point margin of error was permissible, the agreement rose to 91.3% (294/322) and 97.3% (149/153), respectively. In the paired test, PI-RADSAI outperformed 5/11 (45.5%) and matched the performance of 3/11 (27.3%) general radiologists in achieving a clinically significant PCa diagnosis (area under the curve, internal test, 0.801 vs. 0.770, p < 0.01; external test, 0.833 vs. 0.867, p = 0.309).

CONCLUSIONS

Our closed-loop PI-RADSAI outperforms or matches the performance of more than 70% of general readers in the MRI assessment of PCa. This system might provide an alternative to radiologists and offer diagnostic benefits to clinical practice, especially where subspecialist expertise is unavailable.

Utilizing a Digital Swarm Intelligence Platform to Improve Consensus Among Radiologists and Exploring Its Applications

Radiologists today play a central role in making diagnostic decisions and labeling images for training and benchmarking artificial intelligence (AI) algorithms. A key concern is low inter-reader reliability (IRR) seen between experts when interpreting challenging cases. While team-based decisions are known to outperform individual decisions, inter-personal biases often creep up in group interactions which limit nondominant participants from expressing true opinions. To overcome the dual problems of low consensus and interpersonal bias, we explored a solution modeled on bee swarms. Two separate cohorts, three board-certified radiologists, (cohort 1), and five radiology residents (cohort 2) collaborated on a digital swarm platform in real time and in a blinded fashion, grading meniscal lesions on knee MR exams. These consensus votes were benchmarked against clinical (arthroscopy) and radiological (senior-most radiologist) standards of reference using Cohen's kappa. The IRR of the consensus votes was then compared to the IRR of the majority and most confident votes of the two cohorts. IRR was also calculated for predictions from a meniscal lesion detecting AI algorithm. The attending cohort saw an improvement of 23% in IRR of swarm votes (k = 0.34) over majority vote (k = 0.11). Similar improvement of 23% in IRR (k = 0.25) in 3-resident swarm votes over majority vote (k = 0.02) was observed. The 5-resident swarm had an even higher improvement of 30% in IRR (k = 0.37) over majority vote (k = 0.07). The swarm consensus votes outperformed individual and majority vote decision in both the radiologists and resident cohorts. The attending and resident swarms also outperformed predictions from a state-of-the-art AI algorithm.

Predicting prostate cancer in men with PSA levels of 4-10 ng/mL: MRI-based radiomics can help junior radiologists improve the diagnostic performance

To develop MRI-based radiomics model for predicting prostate cancer (PCa) in men with prostate-specific antigen (PSA) levels of 4-10 ng/mL, to compare the performance of radiomics model and PI-RADS v2.1, and to further verify the predictive ability of radiomics model for lesions with different PI-RADS v2.1 score. 171 patients with PSA levels of 4-10 ng/mL were divided into training (n = 119) and testing (n = 52) groups. PI-RADS v2.1 score was assessed by two radiologists. All volumes of interest were segmented on T₂-weighted imaging, diffusion weighted imaging, and apparent diffusion coefficient sequences, from which quantitative radiomics features were extracted. Multivariate logistic regression analysis was performed to establish radiomics model for predicting PCa. The diagnostic performance was assessed using receiver operating characteristic curve analysis. The radiomics model exhibited the best performance in predicting PCa, which was better than the performance of PI-RADS v2.1 scoring by the junior radiologist in the training group [area under the curve (AUC): 0.932 vs 0.803], testing group (AUC: 0.922 vs 0.797), and the entire cohort (AUC: 0.927 vs 0.801) (P < 0.05). The radiomics model performed well for lesions with PI-RADS v2.1 score of 3 (AUC = 0.854, sensitivity = 84.62%, specificity = 84.34%) and PI-RADS v2.1 score of 4-5 (AUC = 0.967, sensitivity = 98.11%, specificity = 86.36%) assigned by junior radiologist. The radiomics model quantitatively outperformed PI-RADS v2.1 for noninvasive prediction of PCa in men with PSA levels of 4-10 ng/mL. The model can help improve the diagnostic performance of junior radiologists and facilitate better decision-making by urologists for management of lesions with different PI-RADS v2.1 score.

Does deep learning software improve the consistency and performance of radiologists with various levels of experience in assessing bi-parametric prostate MRI?

OBJECTIVE

To investigate whether commercially available deep learning (DL) software improves the Prostate Imaging-Reporting and Data System (PI-RADS) scoring consistency on bi-parametric MRI among radiologists with various levels of experience; to assess whether the DL software improves the performance of the radiologists in identifying clinically significant prostate cancer (csPCa).

METHODS

We retrospectively enrolled consecutive men who underwent bi-parametric prostate MRI at a 3 T scanner due to suspicion of PCa. Four radiologists with 2, 3, 5, and > 20 years of experience evaluated the bi-parametric prostate MRI scans with and without the DL software. Whole-mount pathology or MRI/ultrasound fusion-guided biopsy was the reference. The area under the receiver operating curve (AUROC) was calculated for each radiologist with and without the DL software and compared using De Long's test. In addition, the inter-rater agreement was investigated using kappa statistics.

RESULTS

In all, 153 men with a mean age of 63.59 ± 7.56 years (range 53-80) were enrolled in the study. In the study sample, 45 men (29.80%) had clinically significant PCa. During the reading with the DL software, the radiologists changed their initial scores in 1/153 (0.65%), 2/153 (1.3%), 0/153 (0%), and 3/153 (1.9%) of the patients, yielding no significant increase in the AUROC (p > 0.05). Fleiss' kappa scores among the radiologists were 0.39 and 0.40 with and without the DL software (p = 0.56).

CONCLUSIONS

The commercially available DL software does not increase the consistency of the bi-parametric PI-RADS scoring or csPCa detection performance of radiologists with varying levels of experience.

Characterization of high-grade prostate cancer at multiparametric MRI: assessment of PI-RADS version 2.1 and version 2 descriptors across 21 readers with varying experience (MULTI study)

OBJECTIVE

To assess PI-RADSv2.1 and PI-RADSv2 descriptors across readers with varying experience.

METHODS

Twenty-one radiologists (7 experienced (≥ 5 years) seniors, 7 less experienced seniors and 7 juniors) assessed 240 'predefined' lesions from 159 pre-biopsy multiparametric prostate MRIs. They specified their location (peripheral, transition or central zone) and size, and scored them using PI-RADSv2.1 and PI-RADSv2 descriptors. They also described and scored 'additional' lesions if needed. Per-lesion analysis assessed the 'predefined' lesions, using targeted biopsy as reference; per-lobe analysis included 'predefined' and 'additional' lesions, using combined systematic and targeted biopsy as reference. Areas under the curve (AUCs) quantified the performance in diagnosing clinically significant cancer (csPCa; ISUP ≥ 2 cancer). Kappa coefficients (κ) or concordance correlation coefficients (CCC) assessed inter-reader agreement.

RESULTS

At per-lesion analysis, inter-reader agreement on location and size was moderate-to-good (κ = 0.60-0.73) and excellent (CCC ≥ 0.80), respectively. Agreement on PI-RADSv2.1 scoring was moderate (κ = 0.43-0.47) for seniors and fair (κ = 0.39) for juniors. Using PI-RADSv2.1, juniors obtained a significantly lower AUC (0.74; 95% confidence interval [95%CI]: 0.70-0.79) than experienced seniors (0.80; 95%CI 0.76-0.84; p = 0.008) but not than less experienced seniors (0.74; 95%CI 0.70-0.78; p = 0.75). As compared to PI-RADSv2, PI-RADSv2.1 downgraded 17 lesions/reader (interquartile range [IQR]: 6-29), of which 2 (IQR: 1-3) were csPCa; it upgraded 4 lesions/reader (IQR: 2-7), of which 1 (IQR: 0-2) was csPCa. Per-lobe analysis, which included 60 (IQR: 25-73) 'additional' lesions/reader, yielded similar results.

CONCLUSIONS

Experience significantly impacted lesion characterization using PI-RADSv2.1 descriptors. As compared to PI-RADSv2, PI-RADSv2.1 tended to downgrade non-csPCa lesions, but this effect was small and variable across readers.

Joint Cancer Segmentation and PI-RADS Classification on Multiparametric MRI Using MiniSegCaps Network

MRI is the primary imaging approach for diagnosing prostate cancer. Prostate Imaging Reporting and Data System (PI-RADS) on multiparametric MRI (mpMRI) provides fundamental MRI interpretation guidelines but suffers from inter-reader variability. Deep learning networks show great promise in automatic lesion segmentation and classification, which help to ease the burden on radiologists and reduce inter-reader variability. In this study, we proposed a novel multi-branch network, MiniSegCaps, for prostate cancer segmentation and PI-RADS classification on mpMRI. MiniSeg branch outputted the segmentation in conjunction with PI-RADS prediction, guided by the attention map from the CapsuleNet. CapsuleNet branch exploited the relative spatial information of prostate cancer to anatomical structures, such as the zonal location of the lesion, which also reduced the sample size requirement in training due to its equivariance properties. In addition, a gated recurrent unit (GRU) is adopted to exploit spatial knowledge across slices, improving through-plane consistency. Based on the clinical reports, we established a prostate mpMRI database from 462 patients paired with radiologically estimated annotations. MiniSegCaps was trained and evaluated with fivefold cross-validation. On 93 testing cases, our model achieved a 0.712 dice coefficient on lesion segmentation, 89.18% accuracy, and 92.52% sensitivity on PI-RADS classification (PI-RADS ≥ 4) in patient-level evaluation, significantly outperforming existing methods. In addition, a graphical user interface (GUI) integrated into the clinical workflow can automatically produce diagnosis reports based on the results from MiniSegCaps.

The added value of AI-based computer-aided diagnosis in classification of cancer at prostate MRI

OBJECTIVES

To develop an artificial intelligence (AI) model for prostate segmentation and prostate cancer (PCa) detection, and explore the added value of AI-based computer-aided diagnosis (CAD) compared to conventional PI-RADS assessment.

METHODS

A retrospective study was performed on multi-centers and included patients who underwent prostate biopsies and multiparametric MRI. A convolutional-neural-network-based AI model was trained and validated; the reliability of different CAD methods (concurrent read and AI-first read) were tested in an internal/external cohort. The diagnostic performance, consistency and efficiency of radiologists and AI-based CAD were compared.

RESULTS

The training/validation/internal test sets included 650 (400/100/150) cases from one center; the external test included 100 cases (25/25/50) from three centers. For diagnosis accuracy, AI-based CAD methods showed no significant differences and were equivalent to the radiologists in the internal test (127/150 vs. 130/150 vs. 125/150 for reader 1; 127/150 vs.132/150 vs. 131/150 for reader 2; all p > 0.05), whereas in the external test, concurrent-read methods were superior/equal to AI-first read (87/100 vs. 71/100, p < 0.001, for reader 2; 79/100 vs. 69/100, p = 0.076, for reader 1) and better than/equal to radiologists (79/100 vs. 72/100, p = 0.039, for reader 1; 87/100 vs. 86/100, p = 1.000, for reader 2). Moreover, AI-first read/concurrent read improved consistency in both internal test (κ = 1.000, 0.830) and external test (κ = 0.958, 0.713) compared to radiologists (κ = 0.747, 0.600); AI-first read method (8.54 s/7.66 s) was faster than readers (92.72 s/89.54 s) and concurrent-read method (29.15 s/28.92 s), respectively.

CONCLUSION

AI-based CAD could improve the consistency and efficiency for accurate diagnosis; the concurrent-read method could enhance the diagnostic capabilities of an inexperienced radiologist in unfamiliar situations.

KEY POINTS

• For prostate cancer segmentation, the performance of multi-small Vnet displays optimal compared to small Vnet and Vnet (DSC_msvnet vs. DSC_svnet, p = 0.021; DSC_msvnet vs. DSC_vnet, p < 0.001). • For prostate gland segmentation, the mean/median DSCs for fine and coarse segmentation were 0.91/0.91 and 0.88/0.89, respectively. Fine segmentation displays superior performance compared to coarse (DSC_coarse vs. DSC_fine, p < 0.001). • For PCa diagnosis, AI-based CAD methods improve consistency in internal (κ = 1.000; 0.830) and external (κ = 0.958; 0.713) tests compared to radiologists (κ = 0.747; 0.600); the AI-first read (8.54 s/7.66 s) was faster than the readers (92.72 s/89.54 s) and the concurrent-read method (29.15 s/28.92 s).

Quality checkpoints in the MRI-directed prostate cancer diagnostic pathway

Multiparametric MRI of the prostate is now recommended as the initial diagnostic test for men presenting with suspected prostate cancer, with a negative MRI enabling safe avoidance of biopsy and a positive result enabling MRI-directed sampling of lesions. The diagnostic pathway consists of several steps, from initial patient presentation and preparation to performing and interpreting MRI, communicating the imaging findings, outlining the prostate and intra-prostatic target lesions, performing the biopsy and assessing the cores. Each component of this pathway requires experienced clinicians, optimized equipment, good inter-disciplinary communication between specialists, and standardized workflows in order to achieve the expected outcomes. Assessment of quality and mitigation measures are essential for the success of the MRI-directed prostate cancer diagnostic pathway. Quality assurance processes including Prostate Imaging-Reporting and Data System, template biopsy, and pathology guidelines help to minimize variation and ensure optimization of the diagnostic pathway. Quality control systems including the Prostate Imaging Quality scoring system, patient-level outcomes (such as Prostate Imaging-Reporting and Data System MRI score assignment and cancer detection rates), multidisciplinary meeting review and audits might also be used to provide consistency of outcomes and ensure that all the benefits of the MRI-directed pathway are achieved.

Is Artificial Intelligence Replacing Our Radiology Stars? Not Yet!

Artificial intelligence (AI) is here to stay and will change health care as we know it. The availability of big data and the increasing numbers of AI algorithms approved by the US Food and Drug Administration together will help in improving the quality of care for patients and in overcoming human fatigue barriers. In oncology practice, patients and providers rely on the interpretation of radiologists when making clinical decisions; however, there is considerable variability among readers, and in particular for prostate imaging. AI represents an emerging solution to this problem, for which it can provide a much-needed form of standardization. The diagnostic performance of AI alone in comparison to a combination of an AI framework and radiologist assessment for evaluation of prostate imaging has yet to be explored. Here, we compare the performance of radiologists alone versus a combination of radiologists aided by a modern computer-aided diagnosis (CAD) AI system. We show that the radiologist-CAD combination demonstrates superior sensitivity and specificity in comparison to both radiologists alone and AI alone. Our findings demonstrate that a radiologist + AI combination could perform best for detection of prostate cancer lesions. A hybrid technology-human system could leverage the benefits of AI in improving radiologist performance while also reducing physician workload, minimizing burnout, and enhancing the quality of patient care.

Patient summary

Our report demonstrates the potential of artificial intelligence (AI) for improving the interpretation of prostate scans. A combination of AI and evaluation by a radiologist has the best performance in determining the severity of prostate cancer. A hybrid system that uses both AI and radiologists could maximize the quality of care for patients while reducing physician workload and burnout.

Machine learning-based analysis of a semi-automated PI-RADS v2.1 scoring for prostate cancer

Background

Prostate Imaging-Reporting and Data System version 2.1 (PI-RADS v2.1) was developed to standardize the interpretation of multiparametric MRI (mpMRI) for prostate cancer (PCa) detection. However, a significant inter-reader variability among radiologists has been found in the PI-RADS assessment. The purpose of this study was to evaluate the diagnostic performance of an in-house developed semi-automated model for PI-RADS v2.1 scoring using machine learning methods.

Methods

The study cohort included an MRI dataset of 59 patients (PI-RADS v2.1 score 2 = 18, score 3 = 10, score 4 = 16, and score 5 = 15). The proposed semi-automated model involved prostate gland and zonal segmentation, 3D co-registration, lesion region of interest marking, and lesion measurement. PI-RADS v2.1 scores were assessed based on lesion measurements and compared with the radiologist PI-RADS assessment. Machine learning methods were used to evaluate the diagnostic accuracy of the proposed model by classification of PI-RADS v2.1 scores.

Results

The semi-automated PI-RADS assessment based on the proposed model correctly classified 50 out of 59 patients and showed a significant correlation (r = 0.94, p < 0.05) with the radiologist assessment. The proposed model achieved an accuracy of 88.00% ± 0.98% and an area under the receiver-operating characteristic curve (AUC) of 0.94 for score 2 vs. score 3 vs. score 4 vs. score 5 classification and accuracy of 93.20 ± 2.10% and AUC of 0.99 for low score vs. high score classification using fivefold cross-validation.

Conclusion

The proposed semi-automated PI-RADS v2.1 assessment system could minimize the inter-reader variability among radiologists and improve the objectivity of scoring.

Prostate MRI Qualification: AJR Expert Panel Narrative Review

Prostate MRI is now established as a first-line investigation for individuals presenting with suspected localized or locally advanced prostate cancer. Successful delivery of the MRI-directed pathway for prostate cancer diagnosis relies on high-quality imaging as well as the interpreting radiologist's experience and expertise. Radiologist certification in prostate MRI may help limit interreader variability, optimize outcomes, and provide individual radiologists with documentation of meeting predefined standards. This AJR Expert Panel Narrative Review summarizes existing certification proposals, recognizing variable progress across regions in establishing prostate MRI certification programs. To our knowledge, Germany is the only country with a prostate MRI certification process that is currently available for radiologists. However, prostate MRI certification programs have also recently been proposed in the United States and United Kingdom and by European professional society consensus panels. Recommended qualification processes entail a multifaceted approach, incorporating components such as minimum case numbers, peer learning, course participation, continuing medical education credits, and feedback from pathology results. Given the diversity in health care systems, including in the provision and availability of MRI services, national organizations will likely need to take independent approaches to certification and accreditation. The relevant professional organizations should begin developing these programs or continue existing plans for implementation.

Three-dimensional ultrasound integrating nomogram and the blood flow image for prostate cancer diagnosis and biopsy: A retrospective study

Backgrounds

Prostate cancer (PCa) is the second most common male cancer in the world and based on its high prevalence and overwhelming effect on patients, more precise diagnostic and therapeutic methods are essential research topics. As such, this study aims to evaluate the value of three-dimensional transrectal ultrasound (3D-TRUS) in the detection, diagnosis and biopsy of PCa, and to provide a basis for clinical practice of PCa.

Methods

Retrospective analysis and comparison of a total of 401 male patients who underwent prostate TRUS in our hospital from 2019 to 2020 were conducted, with all patients having prostate biopsy. Nomogram was used to estimate the probability of different ultrasound signs in diagnosing prostate cancer. The ROC curve was used to estimate the screening and diagnosis rates of 3D-TRUS, MRI and TRUS for prostate cancer.

Results

A total of 401 patients were randomly divided into two groups according to different methods of prostate ultrasonography, namely the TRUS group (251 patients) and the 3D-TRUS group (150 patients). Of these cases, 111 patients in 3D-TRUS group underwent MRI scan. The nomogram further determined the value of 3D-TRUS for prostate cancer. The ROC AUC of prostate cancer detected by TRUS, MRI and 3D-TRUS was 0.5580, 0.6216 and 0.6267 respectively. Biopsy complications were lower in 3D-TRUS group than TRUS group, which was statistically significant (P&lt;0.005).

Conclusions

The accuracy of 3D-TRUS was higher in diagnosis and biopsy of prostate cancer. Meanwhile, the positive rate of biopsy could be improved under direct visualization of 3D-TRUS, and the complications could be decreased markedly. Therefore, 3D-TRUS was of high clinical value in diagnosis and biopsy of prostate cancer.

Comparative Performance of Deep Learning and Radiologists for the Diagnosis and Localization of Clinically Significant Prostate Cancer at MRI: A Systematic Review

Background: Deep learning (DL)-based models have demonstrated an ability to automatically diagnose clinically significant prostate cancer (PCa) on MRI scans and are regularly reported to approach expert performance. The aim of this work was to systematically review the literature comparing deep learning (DL) systems to radiologists in order to evaluate the comparative performance of current state-of-the-art deep learning models and radiologists. Methods: This systematic review was conducted in accordance with the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist. Studies investigating DL models for diagnosing clinically significant (cs) PCa on MRI were included. The quality and risk of bias of each study were assessed using the checklist for AI in medical imaging (CLAIM) and QUADAS-2, respectively. Patient level and lesion-based diagnostic performance were separately evaluated by comparing the sensitivity achieved by DL and radiologists at an identical specificity and the false positives per patient, respectively. Results: The final selection consisted of eight studies with a combined 7337 patients. The median study quality with CLAIM was 74.1% (IQR: 70.6–77.6). DL achieved an identical patient-level performance to the radiologists for PI-RADS ≥ 3 (both 97.7%, SD = 2.1%). DL had a lower sensitivity for PI-RADS ≥ 4 (84.2% vs. 88.8%, p = 0.43). The sensitivity of DL for lesion localization was also between 2% and 12.5% lower than that of the radiologists. Conclusions: DL models for the diagnosis of csPCa on MRI appear to approach the performance of experts but currently have a lower sensitivity compared to experienced radiologists. There is a need for studies with larger datasets and for validation on external data.

Artificial Intelligence for Automated Cancer Detection on Prostate MRI: Opportunities and Ongoing Challenges, From the AJR Special Series on AI Applications

Use of prostate MRI has increased greatly in the past decade, primarily in directing targeted prostate biopsy. However, prostate MRI interpretation remains prone to interreader variation. Artificial intelligence (AI) has the potential to standardize detection of lesions on MRI that are suspicious for prostate cancer (PCa). The purpose of this review is to explore the current status of AI for the automated detection of PCa on MRI. Recent literature describing promising results regarding AI models for PCa detection on MRI is highlighted. Numerous limitations of the existing literature are also described, including biases in model validation, heterogeneity in reporting of performance metrics, and lack of sufficient evidence of clinical translation. Challenges related to AI ethics and data governance are also discussed. An outlook is provided for AI in lesion detection on prostate MRI in the coming years, emphasizing current research needs. Future investigations, incorporating large-scale diverse multiinstitutional training and testing datasets, are anticipated to enable the development of more robust AI models for PCa detection on MRI, though prospective clinical trials will ultimately be required to establish benefit of AI in patient management.

Developments in proton MR spectroscopic imaging of prostate cancer

In this paper, we review the developments of 1H-MR spectroscopic imaging (MRSI) methods designed to investigate prostate cancer, covering key aspects such as specific hardware, dedicated pulse sequences for data acquisition and data processing and quantification techniques. Emphasis is given to recent advancements in MRSI methodologies, as well as future developments, which can lead to overcome difficulties associated with commonly employed MRSI approaches applied in clinical routine. This includes the replacement of standard PRESS sequences for volume selection, which we identified as inadequate for clinical applications, by sLASER sequences and implementation of 1H MRSI without water signal suppression. These may enable a new evaluation of the complementary role and significance of MRSI in prostate cancer management.

Combining and analyzing novel multi-parametric magnetic resonance imaging metrics for predicting Gleason score

BACKGROUND

Radiologists currently subjectively examine multi-parametric magnetic resonance imaging (MP-MRI) to determine prostate tumor aggressiveness using the Prostate Imaging Reporting and Data System scoring system (PI-RADS). Recent studies showed that modified signal to clutter ratio (SCR), tumor volume, and eccentricity (elongation or roundness) of prostate tumors correlated with Gleason score (GS). No previous studies have combined the prostate tumor's shape, SCR, tumor volume, in order to predict potential tumor aggressiveness and GS.

METHODS

MP-MRI (T1, T2, diffusion, dynamic contrast-enhanced images) were obtained, resized, translated, and stitched to form spatially registered multi-parametric cubes. Multi-parametric signatures that characterize prostate tumors were inserted into a target detection algorithm [adaptive cosine estimator (ACE)]. Pixel-based blobbing, and labeling were applied to the threshold ACE images. Eccentricity calculation used moments of inertia from the blobs. Tumor volume was computed by counting pixels within multi parametric MRI blobs and tumor outlines based on pathologist assessment of whole mount histology. Pathology assessment of GS was performed on whole mount prostatectomy. The covariance matrix and mean of normal tissue background was computed from normal prostate. Using signatures and normal tissue statistics, the z-score, noise corrected SCR [principal component (PC), modified regularization] from each patient was computed. Eccentricity, tumor volume, and SCR were fitted to GS. Analysis of variance assesses the relationship among the variables.

RESULTS

A multivariate analysis generated correlation coefficient (0.60 to 0.784) and P value (0.00741 to <0.0001) from fitting two sets of independent variates, namely, tumor eccentricity (the eccentricity for the largest blob, weighted average for the eccentricity) and SCR (removing 3 PCs, removing 4 PCs, modified regularization, and z-score) to GS. The eccentricity t-statistic exceeded the SCR t-statistic. The three-variable fit to GS using tumor volume (histology, MRI) yielded correlation coefficients ranging from 0.724 to 0.819 (P value <<0.05). Tumor volumes generated from histology yielded higher correlation coefficients than MRI volumes. Adding volume to eccentricity and SCR adds little improvement for fitting GS due to higher correlation coefficients among independent variables and little additional, independent information.

CONCLUSIONS

Combining prostate tumors eccentricity with SCR relatively highly correlates with GS.

Multiclass Classification of Prostate Tumors Following an MR Image Analysis-Based Radiomics Approach

Prostate cancer is one of the most common cancers in men, with symptoms that may be confused with those caused by benign prostatic hyperplasia. One of the key aspects of treating prostate cancer is its early detection, increasing life expectancy and improving the quality of life of those patients. However, the tests performed are often invasive, resulting in a biopsy. A non-invasive alternative is the magnetic resonance imaging (MRI)-based PI-RADS v2 classification. The aim of this work was to find objective biomarkers that allow the PI-RADS classification of prostate lesions using a radiomics approach on Multiparametric MRI. A total of 90 subjects were analyzed. From each segmented lesion, 609 different texture features were extracted using five different statistical methods. Two feature selection methods and eight multiclass predictive models were evaluated. This was a multiclass study in which the best AUC result was 0.7442 ± 0.0880, achieved with the Naïve Bayes model using a subset of 120 features. Valuable results were also obtained using the Random Forests model, obtaining an AUC of 0.7394 ± 0.0965 with a lower number of features (52). Clinical Relevance- The current study establishes a methodology for classifying prostate cancer and supporting clinical decision-making in a fast and efficient manner and avoiding additional invasive procedures using MRI.

Utility of a Rule-Based Algorithm in the Assessment of Standardized Reporting in PI-RADS

RATIONALE AND OBJECTIVES

Adoption of the Prostate Imaging Reporting & Data System (PI-RADS) has been shown to increase detection of clinically significant prostate cancer on prostate mpMRI. We propose that a rule-based algorithm based on Regular Expression (RegEx) matching can be used to automatically categorize prostate mpMRI reports into categories as a means by which to assess for opportunities for quality improvement.

MATERIALS AND METHODS

All prostate mpMRIs performed in the Duke University Health System from January 2, 2015, to January 29, 2021, were analyzed. Exclusion criteria were applied, for a total of 5343 male patients and 6264 prostate mpMRI reports. These reports were then analyzed by our RegEx algorithm to be categorized as PI-RADS 1 through PI-RADS 5, Recurrent Disease, or "No Information Available." A stratified, random sample of 502 mpMRI reports was reviewed by a blinded clinical team to assess performance of the RegEx algorithm.

RESULTS

Compared to manual review, the RegEx algorithm achieved overall accuracy of 92.6%, average precision of 88.8%, average recall of 85.6%, and F1 score of 0.871. The clinical team also reviewed 344 cases that were classified as "No Information Available," and found that in 150 instances, no numerical PI-RADS score for any lesion was included in the impression section of the mpMRI report.

CONCLUSION

Rule-based processing is an accurate method for the large-scale, automated extraction of PI-RADS scores from the text of radiology reports. These natural language processing approaches can be used for future initiatives in quality improvement in prostate mpMRI reporting with PI-RADS.

Ultra-high-field MR in Prostate cancer: Feasibility and Potential

Multiparametric MRI of the prostate at clinical magnetic field strengths (1.5/3 Tesla) has emerged as a reliable noninvasive imaging modality for identifying clinically significant cancer, enabling selective sampling of high-risk regions with MRI-targeted biopsies, and enabling minimally invasive focal treatment options. With increased sensitivity and spectral resolution, ultra-high-field (UHF) MRI (≥ 7 Tesla) holds the promise of imaging and spectroscopy of the prostate with unprecedented detail. However, exploiting the advantages of ultra-high magnetic field is challenging due to inhomogeneity of the radiofrequency field and high local specific absorption rates, raising local heating in the body as a safety concern. In this work, we review various coil designs and acquisition strategies to overcome these challenges and demonstrate the potential of UHF MRI in anatomical, functional and metabolic imaging of the prostate and pelvic lymph nodes. When difficulties with power deposition of many refocusing pulses are overcome and the full potential of metabolic spectroscopic imaging is used, UHF MR(S)I may aid in a better understanding of the development and progression of local prostate cancer. Together with large field-of-view and low-flip-angle anatomical 3D imaging, 7 T MRI can be used in its full strength to characterize different tumor stages and help explain the onset and spatial distribution of metastatic spread.