Journal of Magnetic Resonance Imaging

BackgroundHigh b-value diffusion-weighted images (DWI) are used for detection of clinically significant prostate cancer (csPCa). To decrease scan time and improve signal-to-noise ratio, high b-value (>1000 s/mm2) images are often synthesized instead of acquired. PurposeQualitatively and quantitatively compare synthesized DWI (sDWI) to acquired (aDWI) for detection of csPCa. Study TypeRetrospective Subjects151 consecutive patients who underwent prostate MRI and biopsy. SequenceAxial DWI with b=0, 500, 1000, and 2000 s/mm2 using a 3T clinical scanner using a 32-channel phased-array body coil AssessmentWe synthesized DWI for b=2000 s/mm2 via extrapolation based on monoexponential decay, using b=0 and b=500 s/mm2 (sDWI500) and b=0, b=500, and b=1000 s/mm2 (sDWI1000). Differences between sDWI and aDWI were evaluated within regions of interest (ROIs). The maximum DWI value within each ROI was evaluated for prediction of csPCa. Classification accuracy was also compared to Restriction Spectrum Imaging restriction score (RSIrs), a previously validated biomarker based on multi-exponential DWI. Statistical TestsDiscrimination of csPCa was evaluated via area under the receiver operating characteristic curve (AUC). Statistical significance was assessed using bootstrap difference (two-sided =0.05). ResultsWithin the prostate, mean {+/-} standard deviation of percent mean differences between sDWI and aDWI signal were -46{+/-}35% for sDWI1000 and -67{+/-}24% for sDWI500. AUC for aDWI, sDWI500, sDWI1000, and RSIrs within the prostate 0.62[95% confidence interval: 0.53, 0.71], 0.63[0.54, 0.72], 0.65[0.56, 0.73] and 0.78[0.71, 0.86], respectively. When considering the whole field of view, classification accuracy and qualitative image quality decreased notably for sDWI compared to aDWI and RSIrs. Data ConclusionsDWI is qualitatively comparable to aDWI within the prostate. However, hyperintense artifacts are introduced with sDWI in the surrounding pelvic tissue that interfere with quantitative cancer detection and might mask metastases. In the prostate, RSIrs yields superior quantitative csPCa detection than sDWI or aDWI.

BackgroundThe Restriction Spectrum Imaging restriction score (RSIrs) has demonstrated higher diagnostic accuracy for clinically significant prostate cancer (csPCa) than conventional DWI. Both diffusion and T2 properties of prostate tissue inform the RSI signal, and studies have shown that each may be valuable for csPCa discrimination. PurposeTo determine whether prostate T2 varies across RSI compartments and in the presence of csPCa, and to evaluate whether consideration of compartmental T2 (cT2) improves csPCa detection over RSIrs alone. Study TypeRetrospective. PopulationTwo cohorts (46 and 195 patients) scanned for csPCa. Field Strength/Sequence3T multi-b-value DWI acquired at multiple TEs. AssessmentcT2 values were computed from multi-TE RSI data and compared between RSI model compartments. csPCa detection was compared between RSIrs and a logistic regression model (LRM) for predicting the probability of csPCa using cT2 in combination with RSI measurements. Statistical TestsTwo-sample t-tests (=0.05) were used to compare cT2 values between compartments and between patients with and without csPCa. Area under the receiver operating characteristic curve (AUC) was used to evaluate csPCa detection performance. ResultsIn both cohorts, T2 differed (p<0.05) across all RSI compartments (C1, C2, C3, C4). Voxel-level data from cohort 1 showed that T2 differed between normal and cancerous tissue in C1, C2, C3 (p<0.001). Whole-prostate T2 differed between patients with and without csPCa in C3 (p=0.02). In cohort 2, whole-prostate T2 differed in C1 (p=0.01), C3 (p=0.01), and C4 (p<0.01). Consideration of cT2 improved csPCa discrimination compared to diffusion alone, but not compared to RSIrs [cohort 1: 0.80 vs 0.70 (diffusion) and 0.80 (RSIrs), cohort 2: 0.72 vs 0.65 (diffusion) and 0.72 (RSIrs)]. Data ConclusionSignificant differences in cT2 were observed between normal and cancerous prostatic tissue. With our data, however, consideration of cT2 did not significantly improve cancer detection performance over RSIrs alone.

IntroductionProstate multiparametric magnetic resonance imaging (mpMRI) has greatly improved the detection of clinically significant prostate cancer (csPCa). However, the limited number of expert sub-specialist radiologists capable of interpreting conventional prostate mpMRI is a bottleneck for universal access to this healthcare advance. A reliable and reproducible quantitative imaging biomarker could facilitate implementation of accurate prostate MRI at clinical sites with limited experience, thus ensuring more equitable patient care. Restriction Spectrum Imaging restriction score (RSIrs) is an MRI biomarker that has shown the ability to enhance the qualitative and quantitative interpretation of prostate MRI. However, patient-level factors (age, race, ethnicity, prostate volume, and 5-alpha-reductase inhibitor (5-ARI) use) and acquisition-level factors (scanner manufacturer/model and protocol parameters) can affect prostate mpMRI, and their impact on quantitative RSIrs is unknown. MethodsRSI data from patients with known or suspected csPCa were collected from seven centers. We estimated effects of patient and acquisition factors on prostate voxels overall (Method 1: benign patients only) and on only the maximum RSIrs within each prostate (RSIrsmax; Method 2: benign and csPCa patients) using linear models. We then tested whether adjusting for any estimated systematic biases would improve performance of RSIrs for patient-level detection of csPCa, as measured by area under the ROC curve (AUC). ResultsUsing both Method 1 and Method 2, we observed statistically significant effects on RSIrs of age and acquisition group (p < 0.05). Prostate volume had significant effects using only Method 2. All of these effects were small, and adjusting for them did not improve csPCa detection performance (p [&ge;] 0.05). AUC of RSIrsmax for patient-level csPCa detection was 0.77 (95% CI: 0.75, 0.79) unadjusted, compared to 0.77 (0.76, 0.79) and 0.74 (0.72, 0.76) after adjustment using Method 1 and 2 respectively. ConclusionAge, prostate volume, and imaging acquisition factors may lead to systematic differences in RSIrs, but these effects are small and have minimal impact on performance of RSIrs for detection of csPCa. RSIrs can be used as a reliable biomarker across a wide range of patients, centers, scanners, and acquisition factors.

BackgroundConventional distortion correction techniques include the Reversed Polarity Gradient (RPG) method and FSL-topup, which estimate tissue displacement from EPI images of opposite phase-encoding polarity, and scale image intensity by the Jacobian of the estimated displacement. PurposeTo demonstrate that Jacobian intensity correction (JIC) can cause misleading improvement of EPI image distortion. We propose an alternative distortion correction approach (multi-b RPG; mRPG) that eliminates the JIC factor by normalizing opposite-polarity EPI images across multiple b-values. Study typeRetrospective. Population163 prostate cancer patients without metallic implants. Fieldstrength/Sequence3T diffusion-weighted sequence with EPI readout, using multiple b-values. AssessmentMaps of spatial shift (distortion) were estimated from opposite-polarity EPI volumes using RPG, topup, and mRPG. The estimated spatial shifts from each method were then applied to correct the b=0s/mm2 images (both with and without JIC) and ADC maps (for which JIC is meaningless). Distortion was quantified by the Pearson correlation between opposite-polarity volumes. The distribution of correlation coefficients across all patients was examined for b=0s/mm2 images and ADC maps, before and after distortion correction by each method. The mean, median, and 10th percentile were reported for each distribution. Statistical testsWilcoxon signed-rank tests (=0.05) were used to assess whether correlation increased significantly after distortion correction by each method, and whether mRPG yielded a larger increase versus RPG or topup. ResultsMedian improvement in the correlation between b=0s/mm2 volumes was significantly smaller without JIC (p<0.001): 0.04 vs 0.16 (RPG), 0.06 vs 0.18 (topup). mRPG yielded significantly larger improvements compared to RPG or topup (p<0.001). b=0s/mm2: 0.09 vs 0.04 (RPG) and 0.06 (topup). ADC: 0.09 vs 0.02 (RPG) and 0.03 (topup). Data conclusionDisparity in the distortion-correction performance of conventional methods with and without JIC suggests underestimation of tissue displacement. mRPG shows improved correction of distortion artifacts compared to conventional methods.

BackgroundT2-weighted MRI is standard for detecting clinically significant prostate cancer (csPCa) by identifying visible lesions that stand out from the background prostate. PurposeTo determine whether patients with csPCa have abnormal T2-weighted signal in non-lesion, background prostate tissue (BP). MethodsThis retrospective study included two patient cohorts who underwent 3T MRI examination for suspected csPCa. Median (urine-normalized) T2-weighted signal was computed for BP and compared between patients with and without csPCa. csPCa discrimination performance of T2-weighted BP signal was evaluated using area under receiver operating characteristic curves (AUC). T2 and S0 (a proxy for proton density) were computed and compared between patients with and without csPCa. T2 was also recomputed using larger buffers around csPCa lesions. csPCa discrimination performance was compared between two predictors: Restriction Spectrum Imaging (RSI) C1 and RSI C1 normalized by global prostate median T2-weighted signal. ResultsCohort 1: 46 patients (age: 64{+/-}10 years). Cohort 2: 151 patients (65{+/-}8 years). Urine-normalized T2-weighted signal was systematically lower in BP of subjects with csPCa (p[&le;]0.034) and indicated the presence of cancer (cohort 1: AUC=0.80; cohort 2: AUC=0.68). BP T2 was significantly lower in csPCa patients (p[&le;]0.011), while S0 was not (p[&ge;]0.30). BP T2 measurements were stable to within 5% with buffers from 0 to 30 mm around visible lesions. csPCa discrimination improved with incorporation of BP T2-weighted signal (cohort 1: AUC=0.72 for RSI C1 alone, versus 0.81 with BP T2-weighted signal; cohort 2: AUC=0.63 versus 0.76). ConclusionLower T2-weighted signal in BP suggests the presence of csPCa.

BackgroundMultiparametric MRI is useful for early detection of clinically significant prostate cancer (csPCa), but its standard Apparent Diffusion Coefficient (ADC) has limited utility as a quantitative metric for automated, patient-level detection of csPCa. Restriction Spectrum Imaging (RSI), an advanced diffusion technique, yields a quantitative biomarker (RSIrs) that improves csPCa detection. RSIrs is typically calculated from a dedicated multi-b-value acquisition. RSIrs estimated from conventional MRI has not been studied. PurposeTo evaluate the accuracy and validity of RSI metrics estimated post-hoc from conventional diffusion-weighted imaging (DWI) to serve as a viable surrogate for a dedicated RSI acquisition. Materials and MethodsWe conducted a retrospective, multi-center study of patients with both a dedicated RSI acquisition and conventional DWI. We compared three different RSI restriction score (RSIrs) calculation methods: from the dedicated acquisition (RSIrsdedicated), from conventional DWI alone (RSIrspost-hoc), and from a combination of conventional DWI with only the high b-values from the RSI acquisition (RSIrscombo). We compared these methods for quantitative agreement and csPCa detection performance (Area under the Receiver Operating Characteristic [AUC, 95% Confidence Interval]) of maximum RSIrs (RSIrsmax) in the prostate compared to that of minimum ADC (ADC). ResultsData from n=1095 patients (16 centers) were analyzed. Post-hoc RSIrsmax differed systematically from RSIrsdedicated by a median of +156 (RSIrspost-hoc) and -59 (RSIrscombo), respectively. AUCs for csPCa detection were 0.51 [0.47,0.54], 0.60 [0.57,0.64], 0.70 [0.67,0.74], and 0.77 [0.74,0.80] for ADC, RSIrspost-hoc, RSIrscombo, and RSIrsdedicated, respectively. ConclusionEven when estimated using conventional DWI, RSIrs is a superior quantitative biomarker to ADC for automated, patient-level detection of csPCa. A dedicated RSI acquisition gives the best performance. A compromise would be to acquire high b-values (1500 s/mm2, 2500 s/mm2) to complement low b-values (<1000 s/mm2) from conventional DWI.

BackgroundProstate cancer (PCa) is the most prevalent male cancer in the U.S., accounting for 29% of new cancer diagnoses. Multiparametric MRI (MP-MRI), including T2-weighted imaging (T2WI) and apparent diffusion coefficient (ADC) maps, is an effective tool for detecting PCa; however, accuracy varies, and false-positives may lead to unnecessary biopsies or overtreatment. Radio-pathomic maps (RPMs), derived from MP-MRI and machine learning, have been advantageous in differentiating clinically significant PCa. This study tested whether RPMs of tissue density and histo-morphometric features could better predict cancer presence than conventional MR imaging. Materials and MethodsMP-MRI from 236 patients prospectively recruited between 2014 and 2023 with confirmed PCa were analyzed. Whole-mount prostate sections sliced to match the MRI were processed, digitized, and Gleason-pattern annotated by a GU pathologist. Automated algorithms identified glands and calculated quantitative histo-morphometric features, which were mapped across whole slide images. Slides were nonlinearly aligned to each patients T2WI using in-house software, enabling direct comparison of slides, features, and annotations in MR-space. A multi-step prediction model was trained using a 2/3 - 1/3 train/test split to predict histo-morphometric features using 5x5 voxel tiles from T2WI and ADC. These feature maps were then used generate tumor probability maps. ResultsHistological feature models produced RMSE values approximately within one standard deviation of the ground truths variability, indicating acceptable performance. The best RPM, using histological density features, achieved an accuracy of [~]80%. Visual inspection of RPMs showed good concordance to high-grade cancer annotations. ConclusionThis study demonstrates that the use of MRI intensities can predict complex histo-morphometric features and delineate regions of PCa non-invasively. Future research is warranted to determine the clinical benefit of using RPMs in treatment guidance.

Rationale and ObjectivesTo compare the a commercially available automatic and manually adjusted segmentation software program (DynaCAD (R)) to two ellipsoid volume methods using T2-weighted magnetic resonance imaging (MRI). Material and MethodsThis is a retrospective IRB-approved study of 146 patients randomly selected from 1600 consecutive men referred for T2-weighted MRI. All measurements were performed by a single expert senior radiologist. Total prostate volume was calculated using automatic DynaCAD (R) software (RCAD), manually adjusted DynaCAD (R) (ACAD), traditional ellipsoid method (TE) and a new alternative biproximate ellipsoid method (BE). Results were assessed with ANOVA and linear regression. ResultsMean volumes for RCAD, ACAD, BE and TE were 61.5, 58.4, 56, and 53.2 respectively. ANOVA showed no difference of the means (p> 0.05.) Linear regression showed a coefficient of determination (r 2) between ACAD and TE of 0.92 and between ACAD and BE of 0.90. Using the planigraphic-based segmented ACAD as the "gold standard, RCAD overestimated volume by 5%. TE and BE underestimated prostatic volume by 4% and 9% respectively. ACAD processing time was 4.5 to 9.5 minutes (mean=6.6 min.) compared to 1.5 to 3.0 minutes (mean=2.3 min.) for prolate ellipsoid methods. ConclusionManually adjusted MRI T2-weighted segmentation is likely the most accurate measure of total prostate volume. DynaCAD appears to fulfill that function, but manual adjustment of automatic misregistration of boundaries is necessary. ACAD and RCAD are best applied to research use. Ellipsoid methods are faster, more convenient, nearly as accurate and more practical for clinical use.

BackgroundMultiparametric MRI (mpMRI) is strongly recommended by current clinical guidelines for improved detection of clinically significant prostate cancer (csPCa). However, major limitations of mpMRI are the need for intravenous (IV) contrast and dependence on reader expertise. Efforts to address these issues include use of biparametric MRI (bpMRI) and advanced, quantitative MRI techniques. One such advanced technique is the Restriction Spectrum Imaging restriction score (RSIrs), an imaging biomarker that has been shown to improve quantitative accuracy of patient-level csPCa detection. PurposeTo evaluate whether IV contrast can be avoided in the setting of standardized, state-of-the-art image acquisition, with or without addition of RSIrs, and to evaluate characteristics of RSIrs as a stand-alone, quantitative biomarker. Design, setting, and participantsART-Pro is a multisite, multinational trial that will be conducted in two stages, evaluating bpMRI, mpMRI, and RSIrs on accuracy of expert (ART-Pro-1) and non-expert (ART-Pro-2) radiologists detection of csPCa. Additionally, RSIrs will be evaluated as a stand-alone, quantitative, objective biomarker (ART-Pro-1). This study will include a total of 500 patients referred for a multiparametric prostate MRI with a clinical suspicion of prostate cancer at any of the five participating sites (100 patients per site). InterventionIn ART-Pro-1, patients receive standard of care mpMRI, with addition of the RSI sequence, and subsets of the patients images are read separately by two expert radiologists, one of whom is the standard of care radiologist (Reader 1). Three research reports are generated using: bpMRI only (Reader 1), mpMRI (Reader 1), and bpMRI + RSIrs (Reader 2). The clinical report is submitted by Reader 1. Patients future prostate cancer management will be recorded and used to evaluate the performance of the MRI techniques being tested. In ART-Pro-2, the dataset created in ART-Pro-1 will be retrospectively reviewed by radiologists of varying experience level (novice, basic, and expert). Radiologists will be assigned to read cases and record research reports while viewing subsets of either mpMRI only or RSIrs + mpMRI. Patient cases will be read by two readers from each experience level (6 reads total), and findings will be evaluated against the expertly created dataset from ART-Pro-1. Outcome measurements and statistical analysisThe primary endpoint is to evaluate if bpMRI is non-inferior to mpMRI among expert radiologists (ART-Pro-1) and non-expert radiologists (ART-Pro-2) for detection of grade group (GG) [&ge;]2 csPCa. We will conduct one-sided non-inferiority tests of correlated proportions (ART-Pro-1) and use McNemars test and AUC to test the null hypothesis of non-inferiority (ART-Pro-1 and ART-Pro-2). ConclusionsThis trial is registered in the US National Library of Medicine Trial Registry (NCT number: NCT06579417) at ClinicalTrials.gov. Patient accrual at the first site (UC San Diego) began in December 2023. The expected trial timeline is three years to complete accrual with a six-month endpoint.

PurposeTo investigate the effect of scanner and prostate MRI acquisition characteristics when compared to PI-RADSv2.1 technical standards in the performance of a deep learning prostate segmentation model trained with data from one center (INST1), longitudinally evaluated at the same institution and when transferred to other institutions. Materials and MethodsIn this retrospective study, a nn-UNet for prostate MRI segmentation was trained with data from 204 patients from one institution (INST1) (0.50mm2 in-plane, 3.6mm thickness and 16cm field of view [FOV]). Post-deployment performance at INST1 was tested with 30 patients acquired with a different protocol and in a different period of time (0.60mm2 in-plane, 4.0mm thickness and 19cm FOV). Transferability was tested on 248 patient sequences from five institutions (INST2, INST3, INST4, INST5 and INST6) acquired with different scanners and with heterogeneous degrees of PI-RADS v2.1 technical adherence. Performance was assessed using Dice Score Coefficient, Hausdorff Distance, Absolute Boundary Distance and Relative Volume Difference. ResultsThe model presented a significant degradation for the whole gland (WG) in the presence of a change of acquisition protocol at INST1 (DSC:99.46{+/-}0.12% and 91.24{+/-}3.32%,P<.001; RVD:-0.006{+/-}0.127% and 8.10{+/-}8.16%, P<.001). The model had a significantly higher performance in centers adhering to PI-RADS v2.1 when compared to those that did not (DSC: 86.24{+/-}9.67% and 74.83{+/-}15.45%, P <.001; RVD: -6.50{+/-}18.46% and 1.64{+/-}29.12%, P=.003). ConclusionsAdherence to PI-RADSv2.1 technical standards benefits inter-institutional transferability of a deep learning prostate segmentation model. Post-deployment evaluations are critical to ensure model performance is maintained over time in the presence of protocol acquisition modifications.

PurposeMultiparametric MRI (mpMRI) improves detection of clinically significant prostate cancer (csPCa), but the qualitative PI-RADS system and quantitative apparent diffusion coefficient (ADC) yield inconsistent results. An advanced Restrictrion Spectrum Imaging (RSI) model may yield a better quantitative marker for csPCa, the RSI restriction score (RSIrs). We evaluated RSIrs for patient-level detection of csPCa. Materials and MethodsRetrospective analysis of men who underwent mpMRI with RSI and prostate biopsy for suspected prostate cancer from 2017-2019. Maximum RSIrs within the prostate was assessed by area under the receiver operating characteristic curve (AUC) for discriminating csPCa (grade group [&ge;]2) from benign or grade group 1 biopsies. Performance of RSIrs was compared to minimum ADC and PI-RADS v2-2.1via bootstrap confidence intervals and bootstrap difference (two-tailed =0.05). We also tested whether the combination of PI-RADS and RSIrs (PI-RADS+RSIrs) was superior to PI-RADS, alone. Results151 patients met criteria for inclusion. AUC values for ADC, RSIrs, and PI-RADS were 0.50 [95% confidence interval: 0.41, 0.60], 0.76 [0.68, 0.84], and 0.78 [0.71, 0.85], respectively. RSIrs (p=0.0002) and PI-RADS (p<0.0001) were superior to ADC for patient-level detection of csPCa. The performance of RSIrs was comparable to that of PI-RADS (p=0.6). AUC for PI-RADS+RSIrs was 0.84 [0.77, 0.90], superior to PI-RADS or RSIrs, alone (p=0.008, p=0.009). ConclusionsRSIrs was superior to conventional ADC and comparable to (routine, clinical) PI-RADS for patient-level detection of csPCa. The combination of PI-RADS and RSIrs was superior to either alone. RSIrs is a promising quantitative marker worthy of prospective study in the setting of csPCa detection. DisclosuresMEH reports honoraria from Multimodal Imaging Services Corporation and research funding from General Electric Healthcare. AMD is a Founder of and holds equity in CorTechs Labs, Inc, and serves on its Scientific Advisory Board. He is a member of the Scientific Advisory Board of Human Longevity, Inc. and receives funding through research agreements with General Electric Healthcare. The terms of these arrangements have been reviewed and approved by the University of California San Diego in accordance with its conflict-of-interest policies. TMS reports honoraria from Multimodal Imaging Services Corporation, Varian Medical Systems, and WebMD; he has an equity interest in CorTechs Labs, Inc. and also serves on its Scientific Advisory Board. These companies might potentially benefit from the research results. The terms of this arrangement have been reviewed and approved by the University of California San Diego in accordance with its conflict-of-interest policies.

Rationale and ObjectivesTo evaluate short and long-term changes in T2 relaxation as a response to radiotherapy in patients with low and intermediate risk localized prostate cancer. Materials and MethodsA total of 24 patients were selected for this retrospective study. Each participant underwent 1.5T magnetic resonance imaging on seven separate occasions: initially after the implantation of gold fiducials, the required step for Cyberknife therapy guidance, followed by MRI scans two weeks post-therapy and monthly thereafter. As part of each MRI scan, the prostate region was manually delineated, and the T2 relaxation times were calculated for quantitative analysis. The T2 relaxation times between individual follow-ups were analyzed using Repeated Measures Analysis of Variance. ResultsRepeated Measures Analysis of Variance (RM-ANOVA) revealed a significant difference across all measurements (F (6, 120) = 0.611, p << 0.001). A Bonferroni post hoc test revealed significant differences in median T2 values between the baseline and subsequent measurements, particularly between pre-therapy (M0) and two weeks post-therapy (M1), as well as during the monthly interval checks (M2 - M6). Some cases showed a delayed decrease in relaxation times, indicating the prolonged effects of therapy. ConclusionThe changes in T2 values during the course of radiotherapy can help in monitoring radiotherapy response in unconfirmed patients, quantifying the scarring process, and recognizing the therapy failure.

PurposePrecise delineation of genitourinary structures during prostate cancer (PCa) care is critical to optimize treatment delivery while minimizing toxicity and injury. The Prostate and UREthra on MRI (PURE-MRI) study was an international, prospective study to assess physicians accuracy segmenting prostate and urethra on MRI. MethodsPhysicians who diagnose or treat PCa were invited to contour prostate and urethra on patient cases using standard T2-weighted MRI (all planes). We compared these contours to reference consensus segmentations produced by a multidisciplinary panel of experts. We also evaluated performance of a validated prostate auto- segmentation AI tool. Accuracy was assessed with spatial and volumetric analyses. Mixed effects model was used to evaluate potential factors influencing contour performance. Results62 specialists from 11 countries created 114 prostate and 110 urethra contours. Prostate median (min, max) Dice score was 0.92 (0.62, 0.95) for physicians. There was no clear effect of clinical experience or focus. Maximum deviation inside (under-segmentation), maximum deviation beyond expert contour, and mean deviation (per case) from the reference prostate were 3.4 mm (1.0, 12.4), 5.3 mm (2.4, 17.3), and 1.6 mm (0.9, 3.9), respectively. In comparison, prostate auto-segmentation tool results were 0.95 (0.94, 0.96), 3.0 mm, 3.9 mm (3.1, 4.9), and 1.2 mm (1.1, 1.6), respectively. Physician performance was considerably worse for urethra, with Dice score of 0.33 (0.03, 0.69). No urethra AI tool was tested. ConclusionPhysicians contour the prostate on MRI with overall Dice score >0.9, though contours typically had errors >5 mm and sometimes >10 mm. These patterns were observed regardless of clinical experience, specialty, or clinical focus. AI tool performs well enough for clinical use, given comparable accuracy to practicing physicians. In contrast, urethra segmentation on MRI is challenging. More training, better imaging, and/or AI tools may be necessary to achieve consistent, accurate results for the urethra.

BackgroundProstate cancer is a malignancy that originates in the prostate gland,and can vary in aggressiveness, often requiring a combination of diagnostic methods and imaging for accurate detection and management. In an attempt to ensure timely diagnosis and prevent complications, the choice of the right diagnostic modality becomes crucial. MRI has gained prominence in the diagnosis of prostate cancer. MRI aids prostate cancer diagnosis by pinpointing suspicious areas for in-depth investigation and guiding precise biopsies, enhancing accuracy. It also informs treatment plans by visualizing tumour extent and assists in monitoring disease progression during active surveillance. The purpose of this meta-analysis is to ascertain the accuracy of MRI in diagnosing prostate cancer. MethodologyMedical literature was comprehensively searched and reviewed without restrictions to particular study designs, or publication dates using PubMed, Cochrane Library, and Google Scholar databases for all relevant literature. The extraction of necessary data proceeded after specific inclusion and exclusion criteria were applied. In this Meta-Analysis, A total of 47 RCTs with 13,211 subjects were selected for comparing Multiparametric MRI vs. Gold Standard and A total of 23 RCTs with 3440 subjects were selected for comparing Biparametric MRI vs Gold Standard. Two writers independently assessed the calibre of each study as well as the use of the Cochrane tool for bias risk apprehension. The statistical software packages RevMan (Review Manager, version 5.3), SPSS (Statistical Package for the Social Sciences, version 20), and Excel in Stata 14 were used to perform the statistical analyses. ResultsWe calculated the sensitivity and specificity of multiparametric as well as biparametric MRI. The Multiparametric MRI demonstrates a sensitivity of 0.84 (95% confidence interval: 0.83 - 0.85) and specificity of 0.69 (95% confidence interval: 0.68 - 0.70). Meanwhile, Biparametric MRI shows a sensitivity of 0.85 (95% confidence interval: 0.84 - 0.86) and specificity of 0.71 (95% confidence interval: 0.69 - 0.73). ConclusionIn conclusion, both Multiparametric MRI and Biparametric MRI exhibit high sensitivity values of 0.84 and 0.85, respectively, indicating their ability to accurately detect prostate cancer. MRI is a robust diagnostic tool for prostate cancer due to its high-resolution imaging and multiparametric approach. It enables targeted biopsies, informs treatment plans, and aids in active surveillance.

BackgroundThe Prostate Imaging Reporting & Data System (PI-RADS), based on multiparametric MRI (mpMRI), is widely used for the detection of clinically significant prostate cancer (csPCa, Gleason Grade Group (GG[&ge;]2)). However, its diagnostic accuracy can be impacted by variability in interpretation. Restriction Spectrum Imaging (RSI), an advanced diffusion-weighted technique, offers a standardized, quantitative approach for detecting csPCa, potentially enhancing diagnostic consistency and performing comparably to expert-level assessments. PurposeTo evaluate whether combining maximum RSI-derived restriction scores (RSIrs-max) with deep learning (DL) models can enhance patient-level detection of csPCa compared to using PI-RADS or RSIrs-max alone. Materials and MethodsData from 1,892 patients across seven institutions were analyzed, selected based on MRI results and biopsy-confirmed diagnoses. Two deep learning architectures, 3D-DenseNet and 3D-DenseNet+RSI (incorporating RSIrs-max), were developed and trained using biparametric MRI (bpMRI) and RSI data across two data splits. Model performance was compared using the area under the receiver operating characteristic curve (AUC) for patient-level csPCa detection, using PI-RADS performance for clinical reference. ResultsNeither RSIrs-max nor the best DL model combined with RSIrs-max significantly outperformed PI-RADS interpretation by expert radiologists. However, when combined with PI-RADS, both approaches significantly improved patient-level csPCa detection, with AUCs of 0.79 (95% CI: 0.74-0.83; P=.005) for combination of RSIrs-max with PI-RADS and 0.81 (95% CI: 0.76-0.85; P<.001) for combination of best DL model with PI-RADS, compared to 0.73 (95% CI: 0.68-0.78) for PI-RADS alone. ConclusionBoth RSIrs-max and DL models demonstrate comparable performance to PI-RADS alone. Integrating either model with PI-RADS significantly enhances patient-level detection of csPCa compared to using PI-RADS alone. Summary StatementRSIrs-max and deep learning models match the performance of expert PI-RADS in patient-level csPCa detection and combining either with PI-RADS yields a significant improvement over PI-RADS alone. Key PointsO_LIIn a study of 1,892 patients from seven institutions undergoing MRI and biopsy for prostate cancer, RSIrs-max and the DL model (AUC, 0.75 (P=.59) and 0.78 (P=.09)) performed comparably to expert-level PI-RADS scores (AUC, 0.73). C_LIO_LIIncluding prostate auto-segmentation improved the DL model (AUC, 0.68 (P=.01) vs 0.72 (P=.60)). C_LIO_LICombining RSIrs-max or the DL model (AUC, 0.79 (P=.005) and 0.81 (P <.001)) with PI-RADS statistically significantly outperformed PI-RADS alone (AUC, 0.73). C_LI

BackgroundTo determine the accuracy of a novel BIA test endorectal probe. MethodsOne hundred-forty consecutive patient candidates to prostate biopsy and 40 healthy volunteers were selected (NCT03428087). Total PSA and PSA density (PSAD) determinations, digital rectal examination (DRE), and the BIA test were analysed in patients and controls. A 16 cores trans rectal prostate biopsy was performed on all patients with clinical suspicion of PCa after a multiparametric MRI (mMRI) test. The study endpoints were to determine accuracy of BIA test in comparison to PSA, PSAD levels, and mMRI and obtain PCa prediction in candidates to prostate biopsy by BIA test. The Mann-Withney U test, the Wilkoxon rank test, and Holm-Bonferronis method were adopted for statistical analyses, and a computational approach was also applied to differentiate patients with PCa from those with benign disease (BPH). ResultsCombined DRE, TRUS, PSA, and PSAD alone failed to satisfactorily discern patients with PCa from those with BPH (62.86% of discrimination accuracy) and mMRI PIRADS [&ge;]3 showed a sensitivity of 83% and a specificity of 59%. The accuracy in discerning PCa and BPH increased up to 75% by BIA test (sensitivity 63.33% and specificity 83.75%). ConclusionsThe BIA test is a simple, promising, cheap, and reliable test for PCa non-invasive diagnosis. The novel finger probe may improve PCa detection also in patients with low-risk PCa, thus reducing the need of useless biopsies.

PurposeThe purpose of this study was to determine the quantitative variability of diffusion weighted imaging and apparent diffusion coefficient values across a large fleet of MR systems. Using a NIST traceable magnetic resonance imaging diffusion phantom, imaging was reproducible and the measurements were quantitatively compared to known values. MethodsA fleet of 23 clinical MRI scanners was investigated in this study. A NIST/QIBA DWI phantom was imaged with protocols provided with the phantom. The resulting images were analyzed and ADC maps were generated. User-directed region-of-interests on each of the different vials provided ADC measurements among a wide range of known ADC values. ResultsThree diffusion phantoms were used in this study and compared to one another. From the one-way analysis of the variance, the mean and standard deviation of the percent errors from each phantom were not significantly different from one another. The low ADC vials showed larger errors and variation and appear directly related to SNR. Across all the MR systems and data, the coefficient of variation was calculated and Bland-Altman analysis was performed. ADC measurements were similar to one another except for the vials with the lower ADC values, which had a higher coefficient of variation. ConclusionADC values among the three phantoms showed good agreement and were not significantly different from one another. The large percent errors seen primarily at the low ADC values were shown to be a consequence of the SNR dependence and very little bias was observed between magnetic strengths and manufacturers. ADC values between diffusion phantoms were not statistically significant. Future investigations will be performed to study differences in magnetic field strength, vendor, MR system models, gradients, and bore size. More data across different MR platforms would facilitate quantitative measurements for multi-platform and multi-site imaging studies. With the increasing usage of diffusion weighted imaging in the clinic, the characterization of ADC variability for MR systems provides an improved quality control over the MR systems.

We investigated the effectiveness of the most commonly used registration methods (deformable and rigid-body registrations) with different reference images on pharmacokinetic parameters estimated from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) of esophageal cancer patients. We obtained DCE-MRI images from 10 patients with esophageal cancer. Both rigid-body and deformable registrations of the images were performed on DCE-MRI images at different time points as reference images before the pharmacokinetic parameters were estimated. The deformable registration used non-rigid B-spline transforms in a multi-resolution scheme, and Euler transform were used for the rigid body registration. A nonparametric statistical test and the intra-class correlation coefficient assessed the consistency and reproducibility of the pharmacokinetic parameters estimated with both registration methods and using images acquired at different time points. Kruskal-Wallis testing demonstrated significant differences (p < 0.05) in all the estimated parameters for deformable registration but no significant differences (p > 0.78) for rigid-body registration. The intra-class correlation coefficient for rigid-body registration was higher than that for deformable registration for each pharmacokinetic parameter, indicating that, for rigid-body registration, the parameter values from different reference images of one patient tended to be similar to each other. In contrast, the values for deformable registration were more variable. In conclusion, the choice of the reference image of deformable registration significantly affected the estimates of pharmacokinetic parameters, and rigid-body registration showed small variations in pharmacokinetic parameters over the choice of the reference images for small motion artifacts of small distal esophageal cancer on DCE-MRI.

BackgroundThere is increasing interest in high-contrast cerebrospinal fluid (CSF) MRI for imaging perivascular spaces (PVSs). Dilated PVSs, associated with aging, dementia, and various other conditions, are readily detected within the white matter (WM), basal ganglia, and midbrain. While 7T MRI enables detection of smaller PVSs, cortical PVS burden has received limited attention despite its potential value for understanding neurological conditions. PurposeTo investigate the detectability of cortical PVS segments in healthy participants using heavily T2-weighted MRI at 7T. Materials and MethodsA T2-weighted 3D-TSE sequence was optimized at 7T to detect CSF with high resolution and contrast-to-noise ratio (CNR) while minimizing signal from surrounding tissues. A semi-automated pipeline was developed to extract PVSs and quantify their density in the whole brain, including the cortex. ResultsSeventeen healthy volunteers (40{+/-}14 years) were scanned at 7T. Optimized TSE achieved a CSF-to-tissue CNR of [~]180:1, enabling detection of small PVSs throughout the brain and leukocortical segments. About 20% of WM PVSs contain a leukocortical segment. WM PVSs with a leukocortical segment represented 70% of the total PVS volume. PVS density in the cortex was [~]0.7% ([~]6-fold lower than WM), with highest in the insula and lowest in the auditory cortex. ConclusionHigh-resolution CSF imaging using optimized 3D-TSE MRI at 7T allows detection and quantification of leukocortical PVS segments at the gray-white matter interface in healthy individuals. This study lays the groundwork for exploring regional PVS changes related to the cortex and their potential use in diagnosis or prognosis of neurological diseases.

Background and PurposeThe magnetic resonance imaging (MRI) access for patients with active and passive implants is limited by radiofrequency (RF) safety. The time-averaged root-mean-square RF field (B1+rms) and specific absorption rate (SAR) are being evaluated to monitor and control RF-induced heating near conductive metallic implants, such as deep brain stimulation (DBS) leads, during MRI. However, experimental methods to assess the relationship between RF power, B1+rms, and SAR are lacking for RF coils, metallic implants, and ionic solutions. Materials and MethodsA method is developed to evaluate the variation of RF power, B1+rms, and SAR with RF coils, metallic implants, and ionic solutions using phantoms consisting of water (H2O) and sodium chloride (NaCl) with four ionic concentrations (0, 1, 2, 3 %), four metallic wavelengths (0,{lambda} /2,{lambda} , 2{lambda}), two RF coils (body, head) transmit/receive (Tx/Rx) combinations, and five RF pulse flip angles (30{degrees}, 45{degrees}, 60{degrees}, 75{degrees}, 90{degrees}) in two B0 fields (1.5T and 3T). ResultsThe scanner-reported RF power and SAR varied with RF pulse sequences, RF coils, Tx/Rx, metallic implants, and ionic solutions, whereas B1+rms varied only with RF pulse sequences. The RF power, B1+rms, and SAR relationship depends on RF pulse sequences, RF coils, Tx/Rx, implant wavelengths, and ionic concentrations. SAR (whole-body, head) scaled with RF power by absorption ratios () variable with experimental conditions. ConclusionsB1+rms is insensitive to the presence and absence of conductive metallic implants and ionic solutions, implant wavelengths, ionic concentrations, RF coils, and Tx/Rx combinations. RF power must be monitored because scanner-reported SAR may vary unpredictably with experiments.