Biparametric MRI Research Articles

An MRI radiomics model for predicting a prostate-specific antigen response following abiraterone treatment in patients with metastatic castration-resistant prostate cancer

ObjectiveTo establish a combined radiomics-clinical model for the early prediction of a prostate-specific antigen(PSA) response in patients with metastatic castration-resistant prostate cancer(mCRPC) after treatment with abiraterone acetate(AA).MethodsThe data of a total of 60 mCRPC patients from two hospitals were retrospectively analyzed and randomized into a training group(n=48) or a validation group(n=12). By extracting features from biparametric MRI, including T2-weighted imaging(T2WI), diffusion-weighted imaging(DWI), and apparent diffusion coefficient(ADC) maps, radiomics features from the training dataset were selected using least absolute shrinkage and selection operator(LASSO) regression. Four predictive models were developed to assess the efficacy of abiraterone in treating patients with mCRPC. The primary outcome variable was the PSA response following AA treatment. The performance of each model was evaluated using the area under the receiver operating characteristic curve(AUC). Univariate and multivariate analyses were performed using Cox regression to identify significant predictors of the efficacy of abiraterone treatment in patients with mCRPC.ResultsThe integrated model was constructed from seven radiomics features extracted from the T2WI, DWI, and ADC sequence images of the training data. This model demonstrated the highest AUC in both the training and validation cohorts, with values of 0.889 (95% CI, 0.764-0.961) and 0.875 (95% CI, 0.564-0.991). The Rad-score served as an independent predictor of the response to abiraterone treatment in patients with mCRPC (HR: 2.21, 95% CI: 1.01-4.44).ConclusionThe biparametric MRI-based radiomics model has the potential to predict the PSA response in patients with mCRPC following abiraterone treatment.Clinical relevance statementThe MRI-based radiomics model could be used to noninvasively identify the AA response in mCRPC patients, which is helpful for early clinical decision-making.

Impact of Scanner Manufacturer, Endorectal Coil Use, and Clinical Variables on Deep Learning-assisted Prostate Cancer Classification Using Multiparametric MRI.

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Purpose To assess the impact of scanner manufacturer and scan protocol on the performance of deep learning models to classify prostate cancer (PCa) aggressiveness on biparametric MRI (bpMRI). Materials and Methods In this retrospective study, 5,478 cases from ProstateNet, a PCa bpMRI dataset with examinations from 13 centers, were used to develop five deep learning (DL) models to predict PCa aggressiveness with minimal lesion information and test how using data from different subgroups-scanner manufacturers and endorectal coil (ERC) use (Siemens, Philips, GE with and without ERC and the full dataset)-impacts model performance. Performance was assessed using the area under the receiver operating characteristic curve (AUC). The impact of clinical features (age, prostate-specific antigen level, Prostate Imaging Reporting and Data System [PI-RADS] score) on model performance was also evaluated. Results DL models were trained on 4,328 bpMRI cases, and the best model achieved AUC = 0.73 when trained and tested using data from all manufacturers. Hold-out test set performance was higher when models trained with data from a manufacturer were tested on the same manufacturer (within-and between-manufacturer AUC differences of 0.05 on average, P < .001). The addition of clinical features did not improve performance (P = .24). Learning curve analyses showed that performance remained stable as training data increased. Analysis of DL features showed that scanner manufacturer and scan protocol heavily influenced feature distributions. Conclusion In automated classification of PCa aggressiveness using bpMRI data, scanner manufacturer and endorectal coil use had a major impact on DL model performance and features. Published under a CC BY 4.0 license.

Evaluating the feasibility of AI-predicted bpMRI image features for predicting prostate cancer aggressiveness: a multi-center study

ObjectiveTo evaluate the feasibility of utilizing artificial intelligence (AI)-predicted biparametric MRI (bpMRI) image features for predicting the aggressiveness of prostate cancer (PCa).Materials and methodsA total of 878 PCa patients from 4 hospitals were retrospectively collected, all of whom had pathological results after radical prostatectomy (RP). A pre-trained AI algorithm was used to select suspected PCa lesions and extract lesion features for model development. The study evaluated five prediction methods, including (1) A clinical-imaging model of clinical features and image features of suspected PCa lesions selected by AI algorithm, (2) the PIRADS category, (3) a conventional radiomics model, (4) a deep-learning bases radiomics model, and (5) biopsy pathology.ResultsIn the externally validated dataset, the deep learning-based radiomics model showed the highest area under the curve (AUC 0.700 to 0.791). It exceeded the clinical-imaging model (AUC 0.597 to 0.718), conventional radiomic model (AUC 0.566 to 0.632), PIRADS score (AUC 0.554 to 0.613), and biopsy pathology (AUC 0.537 to 0.578). The AUC predicted by the model did not show a statistically significant difference among the three externally verified hospitals (p > 0.05).ConclusionDeep-learning radiomics models utilizing AI-extracted image features from bpMRI images can potentially be used to predict PCa aggressiveness, demonstrating a generalized ability for external validation.Critical relevance statementPredicting the aggressiveness of prostate cancer (PCa) is important for formulating the best treatment plan for patients. The radiomic model based on deep learning is expected to provide an objective and non-invasive method for evaluating the aggressiveness of PCa.Key PointsPredicting the aggressiveness of PCa is important for patients to obtain the best treatment options.The deep learning-based radiomics model can predict the aggressiveness of PCa with high accuracy.The model has good universality when tested on multiple external datasets.Graphical

Multi-site, multi-vendor development and validation of a deep learning model for liver stiffness prediction using abdominal biparametric MRI.

Chronic liver disease (CLD) is a substantial cause of morbidity and mortality worldwide. Liver stiffness, as measured by MR elastography (MRE), is well-accepted as a surrogate marker of liver fibrosis. To develop and validate deep learning (DL) models for predicting MRE-derived liver stiffness using routine clinical non-contrast abdominal T1-weighted (T1w) and T2-weighted (T2w) data from multiple institutions/system manufacturers in pediatric and adult patients. We identified pediatric and adult patients with known or suspected CLD from four institutions, who underwent clinical MRI with MRE from 2011 to 2022. We used T1w and T2w data to train DL models for liver stiffness classification. Patients were categorized into two groups for binary classification using liver stiffness thresholds (≥ 2.5 kPa, ≥ 3.0 kPa, ≥ 3.5 kPa, ≥ 4 kPa, or ≥ 5 kPa), reflecting various degrees of liver stiffening. We identified 4695 MRI examinations from 4295 patients (mean ± SD age, 47.6 ± 18.7 years; 428 (10.0%) pediatric; 2159 males [50.2%]). With a primary liver stiffness threshold of 3.0 kPa, our model correctly classified patients into no/minimal (< 3.0 kPa) vs moderate/severe (≥ 3.0 kPa) liver stiffness with AUROCs of 0.83 (95% CI: 0.82, 0.84) in our internal multi-site cross-validation (CV) experiment, 0.82 (95% CI: 0.80, 0.84) in our temporal hold-out validation experiment, and 0.79 (95% CI: 0.75, 0.81) in our external leave-one-site-out CV experiment. The developed model is publicly available ( https://github.com/almahdir1/Multi-channel-DeepLiverNet2.0.git ). Our DL models exhibited reasonable diagnostic performance for categorical classification of liver stiffness on a large diverse dataset using T1w and T2w MRI data. Question Can DL models accurately predict liver stiffness using routine clinical biparametric MRI in pediatric and adult patients with CLD? Findings DeepLiverNet2.0 used biparametric MRI data to classify liver stiffness, achieving AUROCs of 0.83, 0.82, and 0.79 for multi-site CV, hold-out validation, and external CV. Clinical relevance Our DeepLiverNet2.0 AI model can categorically classify the severity of liver stiffening using anatomic biparametric MR images in children and young adults. Model refinements and incorporation of clinical features may decrease the need for MRE.

Deep Learning to Simulate Contrast-Enhanced MRI for Evaluating Suspected Prostate Cancer.

Background Multiparametric MRI, including contrast-enhanced sequences, is recommended for evaluating suspected prostate cancer, but concerns have been raised regarding potential contrast agent accumulation and toxicity. Purpose To evaluate the feasibility of generating simulated contrast-enhanced MRI from noncontrast MRI sequences using deep learning and to explore their potential value for assessing clinically significant prostate cancer using Prostate Imaging Reporting and Data System (PI-RADS) version 2.1. Materials and Methods Male patients with suspected prostate cancer who underwent multiparametric MRI were retrospectively included from three centers from April 2020 to April 2023. A deep learning model (pix2pix algorithm) was trained to synthesize contrast-enhanced MRI scans from four noncontrast MRI sequences (T1-weighted imaging, T2-weighted imaging, diffusion-weighted imaging, and apparent diffusion coefficient maps) and then tested on an internal and two external datasets. The reference standard for model training was the second postcontrast phase of the dynamic contrast-enhanced sequence. Similarity between simulated and acquired contrast-enhanced images was evaluated using the multiscale structural similarity index. Three radiologists independently scored T2-weighted and diffusion-weighted MRI with either simulated or acquired contrast-enhanced images using PI-RADS, version 2.1; agreement was assessed with Cohen κ. Results A total of 567 male patients (mean age, 66 years ± 11 [SD]) were divided into a training test set (n = 244), internal test set (n = 104), external test set 1 (n = 143), and external test set 2 (n = 76). Simulated and acquired contrast-enhanced images demonstrated high similarity (multiscale structural similarity index: 0.82, 0.71, and 0.69 for internal test set, external test set 1, and external test set 2, respectively) with excellent reader agreement of PI-RADS scores (Cohen κ, 0.96; 95% CI: 0.94, 0.98). When simulated contrast-enhanced imaging was added to biparametric MRI, 34 of 323 (10.5%) patients were upgraded to PI-RADS 4 from PI-RADS 3. Conclusion It was feasible to generate simulated contrast-enhanced prostate MRI using deep learning. The simulated and acquired contrast-enhanced MRI scans exhibited high similarity and demonstrated excellent agreement in assessing clinically significant prostate cancer based on PI-RADS, version 2.1. © RSNA, 2025 Supplemental material is available for this article. See also the editorial by Neji and Goh in this issue.

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

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. 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). 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). 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. 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.

Evaluating deep learning and radiologist performance in volumetric prostate cancer analysis with biparametric MRI and histopathologically mapped slides.

Evaluating deep learning and radiologist performance in volumetric prostate cancer analysis with biparametric MRI and histopathologically mapped slides.

Ultra-high gradient performance 3-Tesla MRI for super-fast and high-quality prostate imaging: initial experience

ObjectivesTo implement and evaluate a super-fast and high-quality biparametric MRI (bpMRI) protocol for prostate imaging acquired at a new ultra-high gradient 3.0-T MRI system.MethodsParticipants with clinically suspected prostate cancer prospectively underwent a multiparametric MRI (mpMRI) on a new 3.0-T MRI scanner (maximum gradient strength: 200 mT/m, maximum slew rate: 200 T/m/s). The bpMRI protocol was extracted from the full mpMRI protocol, including axial T2-weighted and diffusion-weighted (DWI) sequences (b0/800, b1500). Overall image quality was rated by two readers on a five-point Likert scale from (1) non-diagnostic to (5) excellent. PI-RADS 2.1 scores were assessed by three readers separately for the bpMRI and mpMRI protocols. Cohen’s and Fleiss’ κ were calculated for PI-RADS agreement between protocols and interrater reliability between readers, respectively.ResultsSeventy-seven male participants (mean age, 66 ± 8 years) were included. Acquisition time of the bpMRI protocol was reduced by 62% (bpMRI: 5 min, 33 ± 21 s; mpMRI: 14 min, 50 ± 42 s). The bpMRI protocol showed excellent overall image quality for both the T2-weighted (median score both readers: 5 [IQR: 4–5]) and DWI (b1500) sequence (median score reader 1: 4 [IQR: 4–5]; reader 2: 4 [IQR: 4–4]). PI-RADS score agreement between protocols was excellent (Cohen’s κ range: 0.91–0.95 [95% CI: 0.89, 0.99]) with an overall good interrater reliability (Fleiss’ κ, 0.86 [95% CI: 0.80, 0.92]).ConclusionUltra-high gradient MRI allows the establishment of a high-quality and rapidly acquired bpMRI with high PI-RADS agreement to a full mpMRI protocol.Trials registrationClinicaltrials.gov, NCT06244680, Registered 06 February 2024, retrospectively registered, https://classic.clinicaltrials.gov/ct2/show/NCT06244680.Critical relevance statementA novel 3.0-Tesla MRI system with an ultra-high gradient performance enabled high-quality biparametric prostate MRI in 5.5 min while achieving excellent PI-RADS agreement with a standard multiparametric protocol.Key PointsMulti- and biparametric prostate MRIs were prospectively acquired utilizing a maximum gradient of 200 mT/m.Super-fast biparametric MRIs showed excellent image quality and had high PI-RADS agreement with multiparametric MRIs.Implementation of high gradient MRI in clinical routine allows accelerated and high-quality biparametric prostate examinations.Graphical

PI-QUAL version 2 image quality categorisation and inter-reader agreement compared to version 1.

Prostate imaging quality (PI-QUAL) was developed to standardise the evaluation of prostate MRI quality and has recently been updated to version 2. This study aims to assess inter-reader agreement for PI-QUAL v1 and v2 scores and investigates changes in MRI quality score categories. The study retrospectively analysed 350 multiparametric MRI (mpMRI) scans. Two expert uroradiologists independently assessed mpMRI quality using PI-QUAL v1 and v2 guidelines. Biparametric MRI (bpMRI) categorisation based on PI-QUAL v2 included only T2WI and diffusion-weighted imaging (DWI) results. Inter-reader agreement was determined using percentage agreement and kappa, and categorisation comparisons were made using the chi-square test. Substantial inter-reader agreement was observed for the overall PI-QUAL v1 score (κ = 0.64) and moderate agreement for v2 mpMRI (κ = 0.54) and v2 bpMRI scores (κ = 0.57). Inter-reader agreements on individual sequences were similar between v1 and v2 (kappa for individual sequences: T2WI, 0.46 and 0.49; DWI, 0.66 and 0.70; DCE, 0.71 and 0.61). Quality levels shifted from predominantly "optimal" in v1 (65%) down to "acceptable" using v2 (55%); p < 0.001. The addition of DCE increased the proportion of cases with at least "adequate" quality at mpMRI (64%) compared to bpMRI (30%); p < 0.001. This study shows consistent inter-reader agreement between PI-QUAL v1 and v2, encompassing overall and individual sequence categorisation. A notable shift from "optimal" to "acceptable" quality was demonstrated when moving from v1 to v2, with DCE tending improving quality from "inadequate" (bpMRI) to "acceptable" (mpMRI). Question What are the agreement levels of image quality of prostate MRI by using PI-QUAL v1 and v2? Findings Inter-reader agreement based on PI-QUAL v1 and v2 is comparable. Dynamic contrast enhancement (DCE) enables an overall shift from inadequate quality (at bpMRI) to acceptable quality (mpMRI). Clinical relevance The inter-reader agreement on PI-QUAL v1 and v2 is equivalent. PI-QUAL v2 assesses prostate bpMRI as well as mpMRI quality. Transitioning from inadequate to acceptable between v2-bpMRI and v2-mpMRI highlights the role of DCE as an "image quality safety net."

Long-Term Risk of Clinically Significant Prostate Cancer in Biopsy-Negative Patients With Baseline Biparametric Prostate MRI.

The long-term prevalence of clinically significant prostate cancer (csPCa) in patients with initial negative prostate biopsy is unknown. To investigate the rate of csPCa of men with initial negative biopsy. Retrospective analysis of prospectively collected data. A total of 197 men (mean age 63 years [SD ±6.98, range 29-79]) without csPCa on initial biopsy and available baseline biparametric prostate MRI (bpMRI). 3.0 T, turbo spin-echo T2-weighted (axial and sagittal) and three sets of diffusion-weighted imaging using single-shot spin-echo planar imaging (5 b-values 0-500 seconds/mm2; 2 b-values 0 and 1500 seconds/mm2, and 2 b-values 0 and 2000 seconds/mm2). BpMRI was read using Prostate Imaging Reporting Data System (PI-RADS) v2.1. Systematic or targeted biopsy results served as reference standard. Continuous variables were compared using Kruskal-Wallis rank sum test. Categorical variables were compared using either Fisher's exact test or Pearson's chi-square test. Uni- and multivariate regression odds ratios (95% confidence interval) were used to study factors affecting csPCa being diagnosed during follow-up. Time to diagnosis of csPCa is calculated using the Kaplan-Meier method. Of 197 men, 74 (38%), 57 (29%), and 66 (34%) presented with PI-RADS 1-2, 3, and 4-5 findings in the baseline bpMRI. During the median follow-up of 52 months, 8.1%, 5.3%, and 18.2% of these men were diagnosed with csPCa, respectively. Baseline PI-RADS finding was the only factor that associated with csPCa found during the follow-up. Baseline bpMRI with PI-RADS scores 1-3 and initial biopsies negative of csPCa had low rate of csPCa during follow-up, which supports more conservative follow-up for them but further research with longer follow-up is warranted. 3 TECHNICAL EFFICACY: Stage 2.

Cross-shaped windows transformer with self-supervised pretraining for clinically significant prostate cancer detection in bi-parametric MRI.

Bi-parametric magnetic resonance imaging (bpMRI) has demonstrated promising results in prostate cancer (PCa) detection. Vision transformers have achieved competitive performance compared to convolutional neural network (CNN) in deep learning, but they need abundant annotated data for training. Self-supervised learning can effectively leverage unlabeled data to extract useful semantic representations without annotation and its associated costs. This study proposes a novel self-supervised learning framework and a transformer model to enhance PCa detection using prostate bpMRI. We introduce a novel end-to-end Cross-Shaped windows (CSwin) transformer UNet model, CSwin UNet, to detect clinically significant prostate cancer (csPCa) in prostate bpMRI. We also propose a multitask self-supervised learning framework to leverage unlabeled data and improve network generalizability. Using a large prostate bpMRI dataset (PI-CAI) with 1476 patients, we first pretrain CSwin transformer using multitask self-supervised learning to improve data-efficiency and network generalizability. We then finetune using lesion annotations to perform csPCa detection. We also test the network generalization using a separate bpMRI dataset with 158 patients (Prostate158). Five-fold cross validation shows that self-supervised CSwin UNet achieves 0.888±0.010 aread under receiver operating characterstics curve (AUC) and 0.545±0.060 Average Precision (AP) on PI-CAI dataset, significantly outperforming four comparable models (nnFormer, Swin UNETR, DynUNet, Attention UNet, UNet). On model generalizability, self-supervised CSwin UNet achieves 0.79 AUC and 0.45 AP, still outperforming all other comparable methods and demonstrating good generalization to external data. This study proposes CSwin UNet, a new transformer-based model for end-to-end detection of csPCa, enhanced by self-supervised pretraining to enhance network generalizability. We employ an automatic weighted loss (AWL) to unify pretext tasks, improving representation learning. Evaluated on two multi-institutional public datasets, our method surpasses existing methods in detection metrics and demonstrates good generalization to external data.

Use of Multiparametric and Biparametric Magnetic Resonance Imaging in Bladder Cancer Staging: Prospective Observational Study and Analysis of Radiologist Learning Curve.

Background: Nowadays, thanks to the introduction of the VI-RADS scoring system, mpMRI has shown promising results in pre-TURBT assessment of muscular invasiveness of BCa, even if its application in everyday practice is still limited. This might be due to a lack in the literature about the learning curve of radiologists and about the characteristics of the exam. With the aim to reduce scan time and patient discomfort while maintaining diagnostic accuracy, bpMRI has been introduced as a possible alternative to mpMRI in this group of patients. This study reports a single-center experience using mpMRI and the VI-RADS scoring system to differentiate NMIBC from MIBC. The primary aim of the study is to assess diagnostic accuracy of mpMRI using the VI-RADS scoring system. The secondary aim is to evaluate the learning curve of an experienced mpMRI radiologist. Additionally, we perform a retrospective assessment of the same group of patients evaluating only DWIs and T2-weighted images, as they underwent bpMRI, to compare the performance of mpMRI and bpMRI. Materials and Methods: From 11/2021 to 11/2023, patients with suspected newly diagnosed BCa were enrolled in this prospective study. All patients underwent mpMRI prior to TURBT in a highly specialized radiology center for MRI. According to VI-RADS, a cutoff of ≥3 was assumed to define MIBC. Histological TURBT reports were compared with preoperative VI-RADS scores to assess the accuracy of mpMRI in discriminating between NMIBC and MIBC. Furthermore, to assess the learning curve of the reading radiologist we analyzed the rate of patients correctly classified as MIBC at MRI. Finally, we evaluated the performance of a hypothetic biparametric MRI in classifying our cohort according to VI-RADS score and compared it with mpMRI performance by using DeLong's test. Data analysis was performed using Jamovi software v.2.3 and R software v.4.2.1. Results: A total of 133 patients were enrolled. mpMRI showed sensitivity and specificity of 86% (95% confidence interval [CI]: 64-97) and 95% (95% CI: 89-98), respectively. The learning curve analysis of the reading radiologist showed that the rate of patients correctly classified as MIBC rapidly increases reaching its plateau after 40 cases. The hypothetic bpMRI showed a sensitivity of 76% (95% CI: 53-92) and a specificity of 93% (95% CI: 86-97), with no significant difference with mpMRI performance (p = 0.10). Conclusions: Our study confirms the effectiveness of MRI, particularly with the VI-RADS scoring system, in differentiating NMIBC from MIBC. The learning curve analysis underscores the importance of radiologist training in optimizing diagnostic accuracy. Future research should focus on enhancing the sensitivity of bpMRI and further validating these findings in larger and multicentric studies.

The role of MRI in muscle-invasive bladder cancer: an update from the last two years.

Muscle invasive bladder cancer (MIBC) is aggressive and requires radical cystectomy and neoadjuvant therapy, yet over 40% of patients face recurrence. The loss of the bladder also significantly reduces quality of life. Accurate staging, crucial for treatment decisions, is typically done through transurethral resection (TURBT), but inconsistencies in pathology affect diagnosis in 25% of cases. MRI is the most precise imaging method for evaluating local tumor invasiveness. This review discusses recent advances in MRI for staging MIBC and predicting responses to neoadjuvant therapy. Vesical imaging - reporting and data system (VI-RADS) accuracy may improve if combined with ADC maps and tumor contact length, while a bi-parametric MRI approach without contrast could reduce side effects without losing diagnostic precision, though evidence is mixed. VI-RADS shows promise in predicting neoadjuvant therapy responses, and the new nacVI-RADS score is in development. Non-Gaussian diffusion-weighted imaging techniques and machine learning could enhance accuracy but need more integration with mpMRI. VI-RADS may assist in evaluating responses in bladder-sparing regimens. Urodrill, an MRI-guided biopsy, aims to replace diagnostic TURBT but needs more accuracy data. MRI in MIBC is evolving, offering potential for accurate local staging and reduced side effects by avoiding TURBT. Predicting neoadjuvant treatment response could guide personalized treatment and bladder preservation. Larger trials are needed to validate these findings.

Detecting Clinically Significant Prostate Cancer in PI-RADS 3 Lesions Using T2w-Derived Radiomics Feature Maps in 3T Prostate MRI.

Prostate Imaging Reporting and Data System version 2.1 (PI-RADS) category 3 lesions are a challenge in the clinical workflow. A better detection of the infrequently occurring clinically significant prostate cancer (csPCa) in PI-RADS 3 lesions is an important objective. The purpose of this study was to evaluate if feature maps calculated from T2-weighted (T2w) 3 Tesla (3T) MRI can help detect csPCa in PI-RADS category 3 lesions. In-house biparametric 3T prostate MRI examinations acquired between January 2019 and June 2023 because of elevated prostate-specific antigen (PSA) levels were retrospectively screened. Inclusion criteria were a PI-RADS 3 lesion and available results of an ultrasound-guided targeted and systematic biopsy. Exclusion criteria were a simultaneous PI-RADS category 4 or 5 lesion and hip replacement. Target lesions with the International Society of Urological Pathology (ISUP) grade group 1 were rated clinically insignificant PCa (ciPCa) and ≥2 csPCa. This resulted in 52 patients being included in the final analysis, of whom 11 (21.1%), 8 (15.4%), and 33 (63.5%) patients had csPCa, ciPCa, and no PCa, respectively, with the latter two groups being combined as non-csPCa. Eight of the csPCas were located in the peripheral zone (PZ) and three in the transition zone (TZ). In the non-csPCa group, 29 were located in the PZ and 12 in the TZ. Target lesions were marked with volumes of interest (VOIs) on axial T2w images. Axial T2w images were then converted to 93 feature maps. VOIs were copied into the maps, and feature quantity was retrieved directly. Features were tested for significant differences with the Mann-Whitney U-test. Univariate models for single feature performance and bivariate models implementing PSA density (PSAD) were calculated. Ten map-derived features differed significantly between the csPCa and non-csPCa groups (AUCs: 0.70-0.84). The diagnostic performance for TZ lesions (AUC: 0.83-1.00) was superior to PZ lesions (AUC: 0.74-0.85). In the bivariate models, performance in the PZ improved with AUCs >0.90 throughout. Parametric feature maps alone and as bivariate models with PSAD can (?) noninvasively identify csPCa in PI-RADS 3 lesions and could serve as a quantitative tool reducing ambiguity in PI-RADS 3 lesions.

External Validation of a Previously Developed Deep Learning-based Prostate Lesion Detection Algorithm on Paired External and In-House Biparametric MRI Scans.

Purpose To evaluate the performance of an artificial intelligence (AI) model in detecting overall and clinically significant prostate cancer (csPCa)-positive lesions on paired external and in-house biparametric MRI (bpMRI) scans and assess performance differences between each dataset. Materials and Methods This single-center retrospective study included patients who underwent prostate MRI at an external institution and were rescanned at the authors' institution between May 2015 and May 2022. A genitourinary radiologist performed prospective readouts on in-house MRI scans following the Prostate Imaging Reporting and Data System (PI-RADS) version 2.0 or 2.1 and retrospective image quality assessments for all scans. A subgroup of patients underwent an MRI/US fusion-guided biopsy. A bpMRI-based lesion detection AI model previously developed using a completely separate dataset was tested on both MRI datasets. Detection rates were compared between external and in-house datasets with use of the paired comparison permutation tests. Factors associated with AI detection performance were assessed using multivariable generalized mixed-effects models, incorporating features selected through forward stepwise regression based on the Akaike information criterion. Results The study included 201 male patients (median age, 66 years [IQR, 62-70 years]; prostate-specific antigen density, 0.14 ng/mL2 [IQR, 0.10-0.22 ng/mL2]) with a median interval between external and in-house MRI scans of 182 days (IQR, 97-383 days). For intraprostatic lesions, AI detected 39.7% (149 of 375) on external and 56.0% (210 of 375) on in-house MRI scans (P < .001). For csPCa-positive lesions, AI detected 61% (54 of 89) on external and 79% (70 of 89) on in-house MRI scans (P < .001). On external MRI scans, better overall lesion detection was associated with a higher PI-RADS score (odds ratio [OR] = 1.57; P = .005), larger lesion diameter (OR = 3.96; P < .001), better diffusion-weighted MRI quality (OR = 1.53; P = .02), and fewer lesions at MRI (OR = 0.78; P = .045). Better csPCa detection was associated with a shorter MRI interval between external and in-house scans (OR = 0.58; P = .03) and larger lesion size (OR = 10.19; P < .001). Conclusion The AI model exhibited modest performance in identifying both overall and csPCa-positive lesions on external bpMRI scans. Keywords: MR Imaging, Urinary, Prostate Supplemental material is available for this article. © RSNA, 2024.

Prospective evaluation of PI-RADSv2.1 using multiparametric and biparametric MRI for detecting clinically significant prostate cancer based on MRI/US fusion-guided biopsy.

To evaluate the cancer detection rates for each category of Prostate Imaging-Reporting and Data System version 2.1 (PI-RADSv2.1) using multiparametric magnetic resonance imaging (mpMRI) and biparametric MRI (bpMRI) based on MRI/ultrasound (US)-fusion biopsy. This prospective study included participants who underwent mpMRI or bpMRI with a PI-RADSv2.1 interpretation and subsequently received MRI/US-fusion biopsy between August 2022 and December 2023. The lesion-based detection rates of clinically significant prostate cancer (csPCa) in each PI-RADSv2.1 category and the correlation between PI-RADSv2.1 categories and International Society of Urological Pathology (ISUP) grade groups were analyzed. The diagnostic performance of PI-RADSv2.1 in predicting csPCa was evaluated, and diagnostic performance of mpMRI and bpMRI was compared using cut-offs, with PI-RADSv2.1 categories ≥ 3 or ≥ 4 defined as positive. A total of 247 lesions from 216 participants were included in this study. A total of 157 patients underwent mpMRI and the remaining 59 underwent bpMRI. The csPCa detection rates for each PI-RADSv2.1 category of mpMRI and bpMRI were as follows: category 1, 0% (0/11); 2, 13% (3/23); 3, 16% (5/31); 4, 60% (43/72); 5, 65% (26/40), in mpMRI; category 1, 0% (0/4); 2, 33% (1/3); 3, 25% (3/12); 4, 61% (19/31); 5, 75% (15/20) in bpMRI. PI-RADSv2.1 categories were significantly positively associated with csPCa detection rates in both mpMRI and bpMRI (p < 0.0001 and p = 0.00048, respectively). PI-RADSv2.1 categories correlated with ISUP grade groups for mpMRI and bpMRI (p < 0.0001 for both). There were no significant differences in the detection rates between mpMRI and bpMRI for PI-RADS v2.1 positive and negative lesions. PI-RADSv2.1 using mpMRI and bpMRI could stratify the risk of csPCa, and the csPCa detection rate of bpMRI was compatible with that of mpMRI using cut-offs of PI-RADSv2.1 categories ≥ 3 or ≥ 4.

Concordance Between the Expert Reading of Biparametric-MRI and the Nonexpert Multiparametric-MRI for the Detection of Clinically Significant Prostate Cancer: Clinical Implications

PurposeProstate-magnetic resonance imaging (MRI) interpretation is challenging, with expertise playing a crucial role. Biparametric MRI (bpMRI) is gaining popularity in experienced centers due to its time and cost advantages over multiparametric MRI (mpMRI). We aim to analyze concordance between nonexpert radiologist PI-RADS from mpMRI and expert radiologist PI-RADS from bpMRI, and its clinical implications. Material and Methods222 men suspected of having prostate cancer (PCa) and mpMRI reported by nonexpert radiologists were referred to a reference center for transperineal MRI-TRUS fusion biopsy where an expert radiologist reported bpMRI PI-RADS 2.1 and segmentation, blinded to external mpMRI. Mapping targeted suspected lesions and 12-core systematic biopsies were performed. Clinically significant PCa (csPCa) was diagnosed when ISUP-grade group was ≥2. ResultsConcordance between both PI-RADS existed in 49.1% of cases (Kappa index 0.288). In 102 cases (45.9%), expert reclassification to lower PI-RADS existed, while an increase existed in 11 cases (5.0%), P < .001. Agreement existed in 30.8% of nonexpert PI-RADS 3, 43.6% of PI-RADS 4, and 83.7% of PI-RADS 5, P < .001. Potential clinical implications included 27% reduction in prostate biopsies when using expert bpMRI readings compared to nonexpert mpMRI readings (P < 0.001), while undetected csPCa were 4.2% and 3.4%, respectively, P = .669. Over-detection reduction of insignificant PCa was 29.4% and 0%, respectively, P = .034. ConclusionsConcordance between nonexpert PI-RADS mpMRI and expert PI-RADS bpMRI was low, increasing with nonexpert PI-RADS. Expert reclassification would reduce prostate biopsies by more than one quarter and over-detection of iPCa, while csPCa detection remained similar.

Role of contrast agent in evaluation of periprostatic invasion in prostate cancer.

Our study aims to demonstrate the detection of invasion by biparametric prostate MRI (bpMRI). The cases whose histopathological diagnosis was prostate cancer (PCa) and whose mpMRI report was reported as PIRADS 4 and 5 were evaluated retrospectively by two radiologists with different prostate imaging experiences. The images were grouped into two data sets. Dataset-1 was bpMRI, and dataset-2 was mpMRI. Two radiologists first evaluated dataset-1 independently of each other, and 1 month later, dataset-2. They recorded whether there was an invasion and where it was seen in the patients. Then, the results were compared. A total of 75 patients were included in the study. Periprostatic invasion was detected in 33 of the patients. Both the 1st reader and the 2nd reader image detected all the cases with invasion (100%) separately between dataset-1 and set-2. Compatibility for image dataset-1 and dataset-2 between both readers was observed to be excellent. There is no need to use contrast agent to evaluate periprostatic invasion and to have an idea about local staging in PCa patients.

External validation of AI for detecting clinically significant prostate cancer using biparametric MRI.

External validation of AI for detecting clinically significant prostate cancer using biparametric MRI.

Early economic evaluation of magnetic resonance imaging for prostate cancer detection in primary care.

To explore the potential impacts of incorporating prebiopsy magnetic resonance imaging into primary care as a triage test within the prostate cancer diagnostic pathway. Decision analytic modelling with decision trees was utilised for this early economic evaluation. A conceptual model was developed reflecting the common primary care routes to diagnosis for prostate cancer: opportunistic, asymptomatic prostate-specific antigen (PSA) screening or symptomatic presentation. The use of multiparametric MRI (mpMRI) or biparametric MRI (bpMRI) as a primary care triage test following an elevated PSA result was evaluated. A health system perspective was adopted with a time horizon of 12 months. Health effects were expressed in terms of utilities drawn from the literature. The primary outcome was prostate cancer diagnosis. Evidence used to inform the model was drawn from published primary studies, systematic reviews, and secondary analyses of primary and secondary care datasets. Base case analysis showed that the PSA pathway was dominated by both mpMRI- and bpMRI-based pathways for patients undergoing opportunistic screening and symptomatic assessment. bpMRI pathways had greater improvement in cost and utility than mpMRI pathways in both clinical scenarios. Significantly more MRI scans would be performed using the modelled approach (66 626 scans vs. 37 456 scans per 100 000 patients per annum), with fewer subsequent urgent suspected cancer referrals for both mpMRI (38% reduction for screening and symptomatic patients) and bpMRI (72% reduction for screening; 71% for symptomatic) pathways, and a small increase in number of missed cancer diagnoses. Deterministic sensitivity analyses, varying each parameter to its upper and lower 95% confidence intervals, showed no significant change in the dominance of the MRI-based prostate cancer diagnostic pathways. Using prostate MRI as a second-level triage test for suspected prostate cancer in primary care could reduce health service costs without a detrimental effect on patient utility.