Search

BackgroundUnderstanding the characteristics of multiparametric MRI (mpMRI) in patients from different racial/ethnic backgrounds is important for reducing the observed gaps in clinical outcomes.PurposeTo investigate the diagnostic performance of mpMRI and quantitative MRI parameters of prostate cancer (PCa) in African American (AA) and matched White (W) men.Study typeRetrospective.SubjectsOne hundred twenty-nine patients (43 AA, 86 W) with histologically proven PCa who underwent mpMRI before radical prostatectomy.Field strength/sequence3.0 T, T2-weighted turbo spin echo imaging, a single-shot spin-echo EPI sequence diffusion-weighted imaging, and a gradient echo sequence dynamic contrast-enhanced MRI with an ultrafast 3D spoiled gradient-echo sequence.AssessmentThe diagnostic performance of mpMRI in AA and W men was assessed using detection rates (DRs) and positive predictive values (PPVs) in zones defined by the PI-RADS v2.1 prostate sector map. Quantitative MRI parameters, including Ktrans and ve of clinically significant (cs) PCa (Gleason score ≥ 7) tumors were compared between AA and W sub-cohorts after matching age, prostate-specific antigen (PSA), and prostate volume.Statistical testsWeighted Pearson's chi-square and Mann-Whitney U tests with a statistically significant level of 0.05 were used to examine differences in DR and PPV and to compare parameters between AA and matched W men, respectively.ResultsA total number of 264 PCa lesions were identified in the study cohort. The PPVs in the peripheral zone (PZ) and posterior prostate of mpMRI for csPCa lesions were significantly higher in AA men than in matched W men (87.8% vs. 68.1% in PZ, and 89.3% vs. 69.6% in posterior prostate). The Ktrans of index csPCa lesions in AA men was significantly higher than in W men (0.25 ± 0.12 vs. 0.20 ± 0.08 min-1; P

BackgroundMultiparametric MRI (mpMRI) has shown a substantial impact on prostate cancer (PCa) diagnosis. However, the understanding of the spatial correlation between mpMRI performance and PCa location is still limited.PurposeTo investigate the association between mpMRI performance and tumor spatial location within the prostate using a prostate sector map, described by Prostate Imaging Reporting and Data System (PI-RADS) v2.1.Study typeRetrospective.SubjectsOne thousand one hundred forty-three men who underwent mpMRI before radical prostatectomy between 2010 and 2022.Field strength/sequence3.0 T. T2-weighted turbo spin-echo, a single-shot spin-echo EPI sequence for diffusion-weighted imaging, and a gradient echo sequence for dynamic contrast-enhanced MRI sequences.AssessmentIntegrated relative cancer prevalence (rCP), detection rate (DR), and positive predictive value (PPV) maps corresponding to the prostate sector map for PCa lesions were created. The relationship between tumor location and its detection/missing by radiologists on mpMRI compared to WMHP as a reference standard was investigated.Statistical testsA weighted chi-square test was performed to examine the statistical differences for rCP, DR, and PPV of the aggregated sectors within the zone, anterior/posterior, left/right prostate, and different levels of the prostate with a statistically significant level of 0.05.ResultsA total of 1665 PCa lesions were identified in 1143 patients, and from those 1060 lesions were clinically significant (cs)PCa tumors (any Gleason score [GS] ≥7). Our sector-based analysis utilizing weighted chi-square tests suggested that the left posterior part of PZ had a high likelihood of missing csPCa lesions at a DR of 67.0%. Aggregated sector analysis indicated that the anterior or apex locations in PZ had the significantly lowest csPCa detection at 67.3% and 71.5%, respectively.Data conclusionSpatial characteristics of the per-lesion-based mpMRI performance for diagnosis of PCa were studied. Our results demonstrated that there is a spatial correlation between mpMRI performance and locations of PCa on the prostate.Evidence level4 TECHNICAL EFFICACY: Stage 2.

Opinion statementPSMA-PET has been a practice-changing imaging biomarker for the management of men with PCa. Research suggests improved accuracy over conventional imaging and other PET radiotracers in many contexts. With multiple approved PSMA-targeting radiotracers, PSMA PET will become even more available in clinical practice. Its increased use requires an understanding of the prospective data available and caution when extrapolating from prior trial data that utilized other imaging modalities. Future trials leveraging PSMA PET for treatment optimization and management decision-making will ultimately drive its clinical utility.

Plasma cell-free DNA (cfDNA) is a noninvasive biomarker for cell death of all organs. Deciphering the tissue origin of cfDNA can reveal abnormal cell death because of diseases, which has great clinical potential in disease detection and monitoring. Despite the great promise, the sensitive and accurate quantification of tissue-derived cfDNA remains challenging to existing methods due to the limited characterization of tissue methylation and the reliance on unsupervised methods. To fully exploit the clinical potential of tissue-derived cfDNA, here we present one of the largest comprehensive and high-resolution methylation atlas based on 521 noncancer tissue samples spanning 29 major types of human tissues. We systematically identified fragment-level tissue-specific methylation patterns and extensively validated them in orthogonal datasets. Based on the rich tissue methylation atlas, we develop the first supervised tissue deconvolution approach, a deep-learning-powered model, cfSort, for sensitive and accurate tissue deconvolution in cfDNA. On the benchmarking data, cfSort showed superior sensitivity and accuracy compared to the existing methods. We further demonstrated the clinical utilities of cfSort with two potential applications: aiding disease diagnosis and monitoring treatment side effects. The tissue-derived cfDNA fraction estimated from cfSort reflected the clinical outcomes of the patients. In summary, the tissue methylation atlas and cfSort enhanced the performance of tissue deconvolution in cfDNA, thus facilitating cfDNA-based disease detection and longitudinal treatment monitoring.

Conditional image generation plays a vital role in medical image analysis as it is effective in tasks such as super-resolution, denoising, and inpainting, among others. Diffusion models have been shown to perform at a state-of-the-art level in natural image generation, but they have not been thoroughly studied in medical image generation with specific conditions. Moreover, current medical image generation models have their own problems, limiting their usage in various medical image generation tasks. In this paper, we introduce the use of conditional Denoising Diffusion Probabilistic Models (cDDPMs) for medical image generation, which achieve state-of-the-art performance on several medical image generation tasks.

The prevalence of nonalcoholic fatty liver disease (NAFLD) is increasing worldwide and is projected to become a major etiology of cirrhosis and hepatocellular carcinoma (HCC). HCC occurs more commonly in NAFLD patients who develop cirrhosis, though HCC is known to occur in the setting of noncirrhotic NAFLD as well. This is of particular importance given that the American College of Radiology (ACR) CT/MRI Liver Reporting and Data System (LI-RADS) algorithm may only be applied to a certain population of patients, and this population does not include those with noncirrhotic NAFLD. Conventional ultrasound (US) has long been in use for HCC surveillance, but contrast-enhanced US (CEUS) is a relatively newer modality, growing in use for assessment of liver lesions, and its use in HCC diagnosis has been formalized with CEUS LI-RADS. The use of US and CEUS in the assessment of liver lesions in NAFLD patients involves the consideration of certain particular nuances, and familiarity with these considerations will continue increasing in importance as the disease becomes more common.

High-resolution magnetic resonance imaging (MRI) sequences, such as 3D turbo or fast spin-echo (TSE/FSE) imaging, are clinically desirable but suffer from long scanning time-related blurring when reformatted into preferred orientations. Instead, multi-slice two-dimensional (2D) TSE imaging is commonly used because of its high in-plane resolution but is limited clinically by poor through-plane resolution due to elongated voxels and the inability to generate multi-planar reformations due to staircase artifacts. Therefore, multiple 2D TSE scans are acquired in various orthogonal imaging planes, increasing the overall MRI scan time. In this study, we propose a novel slice-profile transformation super-resolution (SPTSR) framework with deep generative learning for through-plane super-resolution (SR) of multi-slice 2D TSE imaging. The deep generative networks were trained by synthesized low-resolution training input via slice-profile downsampling (SP-DS), and the trained networks inferred on the slice profile convolved (SP-conv) testing input for 5.5x through-plane SR. The network output was further slice-profile deconvolved (SP-deconv) to achieve an isotropic super-resolution. Compared to SMORE SR method and the networks trained by conventional downsampling, our SPTSR framework demonstrated the best overall image quality from 50 testing cases, evaluated by two abdominal radiologists. The quantitative analysis cross-validated the expert reader study results. 3D simulation experiments confirmed the quantitative improvement of the proposed SPTSR and the effectiveness of the SP-deconv step, compared to 3D ground-truths. Ablation studies were conducted on the individual contributions of SP-DS and SP-conv, networks structure, training dataset size, and different slice profiles.

Prostate cancer is the second leading cause of cancer death among men in the United States. The diagnosis of prostate MRI often relies on accurate prostate zonal segmentation. However, state-of-the-art automatic segmentation methods often fail to produce well-contained volumetric segmentation of the prostate zones since certain slices of prostate MRI, such as base and apex slices, are harder to segment than other slices. This difficulty can be overcome by leveraging important multi-scale image-based information from adjacent slices, but current methods do not fully learn and exploit such cross-slice information. In this paper, we propose a novel cross-slice attention mechanism, which we use in a Transformer module to systematically learn cross-slice information at multiple scales. The module can be utilized in any existing deep-learning-based segmentation framework with skip connections. Experiments show that our cross-slice attention is able to capture cross-slice information significant for prostate zonal segmentation in order to improve the performance of current state-of-the-art methods. Cross-slice attention improves segmentation accuracy in the peripheral zones, such that segmentation results are consistent across all the prostate slices (apex, mid-gland, and base). The code for the proposed model is available at https://bit.ly/CAT-Net.

Multiparametric magnetic resonance imaging (mpMRI) is an emerging standard for diagnosing and prognosing prostate cancer, but ~ 20% of clinically significant tumors are invisible to mpMRI, as defined by the Prostate Imaging Reporting and Data System version 2 (PI-RADSv2) score of one or two. To understand the biological underpinnings of tumor visibility on mpMRI, we examined the proteomes of forty clinically significant tumors (i.e., International Society of Urological Pathology (ISUP) Grade Group 2)-twenty mpMRI-visible and twenty mpMRI-invisible, with matched histologically normal prostate. Normal prostate tissue was indistinguishable between patients with visible and invisible tumors, and invisible tumors closely resembled the normal prostate. These data indicate that mpMRI-visibility arises when tumor evolution leads to large-magnitude proteomic divergences from histologically normal prostate.

ObjectivesTo investigate the usefulness of minimal ablative margin (MAM) control by intra-procedural contrast-enhanced CT (CECT) in microwave ablation (MWA) of liver tumors.MethodsA total of 334 consecutive liver tumors (240 hepatocellular carcinomas [HCCs] and 94 colorectal liver metastases [CRLMs]) in 172 patients treated with percutaneous MWA were retrospectively included. MAM of each tumor was assessed after expected ablation completion using intra-procedural CECT, allowing within-session additional ablation to any potentially insufficient margin. On immediate post-MWA MRI, complete ablation coverage of tumor and final MAM status were determined. The cumulative local tumor progression (LTP) rate was estimated by using the Kaplan-Meier method. To identify predictors of LTP, Cox regression analysis with a shared frailty model was performed.ResultsIntra-procedural CECT findings prompted additional ablation in 18.9% (63/334) of tumors. Final complete ablation coverage of tumor and sufficient MAM were determined by MRI to be achieved in 99.4% (332/334) and 77.5% (259/334), and their estimated 6-month, 1-year, and 2-year LTP rates were 3.2%, 7.5%, and 12.9%; and 1.0%, 2.1%, and 6.9%, respectively. Insufficient MAM on post-MWA MRI, perivascular tumor location, and tumor size (cm) were independent risk factors for LTP (hazard ratio = 14.4, 6.0, and 1.1, p < 0.001, p = 0.003, and p = 0.011, respectively), while subcapsular location and histology (HCC vs CRLM) were not.ConclusionsIn MWA of liver tumors, intra-procedural CECT monitoring of minimal ablative margin facilitates identification of potentially suboptimal margins and guides immediate additional intra-session ablation to maximize rates of margin-sufficient ablations, the latter being a highly predictive marker for excellent long-term local tumor control.Key points• In MWA of liver tumors, intra-procedural CECT can identify potentially suboptimal minimal ablative margin, leading to immediate additional ablation in a single treatment session. • Achieving a finally sufficient ablative margin through the MWA with intra-procedural CECT monitoring of minimal ablative margin results in excellent local tumor control.

ObjectivesThis study aimed at developing dictionary learning (DL) based compressed sensing (CS) reconstruction for randomly undersampled five-dimensional (5D) MR Spectroscopic Imaging (3D spatial + 2D spectral) data acquired in prostate cancer patients and healthy controls, and test its feasibility at 8x and 12x undersampling factors.Materials and methodsProspectively undersampled 5D echo-planar J-resolved spectroscopic imaging (EP-JRESI) data were acquired in nine prostate cancer (PCa) patients and three healthy males. The 5D EP-JRESI data were reconstructed using DL and compared with gradient sparsity-based Total Variation (TV) and Perona-Malik (PM) methods. A hybrid reconstruction technique, Dictionary Learning-Total Variation (DLTV), was also designed to further improve the quality of reconstructed spectra.ResultsThe CS reconstruction of prospectively undersampled (8x and 12x) 5D EP-JRESI data acquired in prostate cancer and healthy subjects were performed using DL, DLTV, TV and PM. It is evident that the hybrid DLTV method can unambiguously resolve 2D J-resolved peaks including myo-inositol, citrate, creatine, spermine and choline.ConclusionImproved reconstruction of the accelerated 5D EP-JRESI data was observed using the hybrid DLTV. Accelerated acquisition of in vivo 5D data with as low as 8.33% samples (12x) corresponds to a total scan time of 14 min as opposed to a fully sampled scan that needs a total duration of 2.4 h (TR = 1.2 s, 32 [Formula: see text]×16 [Formula: see text]×8 [Formula: see text], 512 [Formula: see text] and 64 [Formula: see text]).

BackgroundMen with grade group 2 or 3 prostate cancer are often considered ineligible for active surveillance; some patients with grade group 2 prostate cancer who are managed with active surveillance will have early disease progression requiring radical therapy. This study aimed to investigate whether MRI-guided focused ultrasound focal therapy can safely reduce treatment burden for patients with localised grade group 2 or 3 intermediate-risk prostate cancer.MethodsIn this single-arm, multicentre, phase 2b study conducted at eight health-care centres in the USA, we recruited men aged 50 years and older with unilateral, MRI-visible, primary, intermediate-risk, previously untreated prostate adenocarcinoma (prostate-specific antigen ≤20 ng/mL, grade group 2 or 3; tumour classification ≤T2) confirmed on combined biopsy (combining MRI-targeted and systematic biopsies). MRI-guided focused ultrasound energy, sequentially titrated to temperatures sufficient for tissue ablation (about 60-70°C), was delivered to the index lesion and a planned margin of 5 mm or more of normal tissue, using real-time magnetic resonance thermometry for intraoperative monitoring. Co-primary outcomes were oncological outcomes (absence of grade group 2 and higher cancer in the treated area at 6-month and 24-month combined biopsy; when 24-month biopsy data were not available and grade group 2 or higher cancer had occurred in the treated area at 6 months, the 6-month biopsy results were included in the final analysis) and safety (adverse events up to 24 months) in all patients enrolled in the study. This study is registered with ClinicalTrials.gov, NCT01657942, and is no longer recruiting.FindingsBetween May 4, 2017, and Dec 21, 2018, we assessed 194 patients for eligibility and treated 101 patients with MRI-guided focused ultrasound. Median age was 63 years (IQR 58-67) and median concentration of prostate-specific antigen was 5·7 ng/mL (IQR 4·2-7·5). Most cancers were grade group 2 (79 [78%] of 101). At 24 months, 78 (88% [95% CI 79-94]) of 89 men had no evidence of grade group 2 or higher prostate cancer in the treated area. No grade 4 or grade 5 treatment-related adverse events were reported, and only one grade 3 adverse event (urinary tract infection) was reported. There were no treatment-related deaths.Interpretation24-month biopsy outcomes show that MRI-guided focused ultrasound focal therapy is safe and effectively treats grade group 2 or 3 prostate cancer. These results support focal therapy for select patients and its use in comparative trials to determine if a tissue-preserving approach is effective in delaying or eliminating the need for radical whole-gland treatment in the long term.FundingInsightec and the National Cancer Institute.

The role of prostate-specific membrane antigen (PSMA)-targeted PET in comparison to multiparametric MRI (mpMRI) in the evaluation of intraprostatic cancer foci is not well defined. The aim of our study was to compare the diagnostic performance of 68Ga-PSMA-11 PET/CT (PSMA PET/CT), mpMRI, and PSMA PET/CT + mpMRI using 3 independent masked readers for each modality and with histopathology as the gold standard in the detection, intraprostatic localization, and determination of local extension of primary prostate cancer. Methods: Patients with intermediate- or high-risk prostate cancer who underwent PSMA PET/CT as part of a prospective trial (NCT03368547) and mpMRI before radical prostatectomy were included. Each imaging modality was interpreted by 3 independent readers who were unaware of the other modality result. A central majority rule was applied (2:1). Pathologic examination of whole-mount slices was used as the gold standard. Imaging scans and whole-mount slices were interpreted using the same standardized approach on a segment level and a lesion level. A "neighboring" approach was used to define imaging-pathology correlation for the detection of individual prostate cancer foci. Accuracy in determining the location, extraprostatic extension (EPE), and seminal vesicle invasion (SVI) of prostate cancer foci was assessed using receiver-operating-characteristic curve analysis. Interreader agreement was calculated using intraclass correlation coefficient analysis. Results: The final analysis included 74 patients (14 [19%] with intermediate risk and 60 [81%] with high risk). The cancer detection rate (lesion-based analysis) was 85%, 83%, and 87% for PSMA PET/CT, mpMRI, and PSMA PET/CT + mpMRI, respectively. The change in AUC was statistically significant between PSMA PET/CT + mpMRI and the 2 imaging modalities alone for delineation of tumor localization (segment-based analysis) (P < 0.001) but not between PSMA PET/CT and mpMRI (P = 0.093). mpMRI outperformed PSMA PET/CT in detecting EPE (P = 0.002) and SVI (P = 0.001). In the segment-level analysis, intraclass correlation coefficient analysis showed moderate reliability among PSMA PET/CT and mpMRI readers using a 5-point Likert scale (range, 0.53-0.64). In the evaluation of T staging, poor reliability was found among PSMA PET/CT readers and poor to moderate reliability was found for mpMRI readers. Conclusion: PSMA PET/CT and mpMRI have similar accuracy in the detection and intraprostatic localization of prostate cancer foci. mpMRI performs better in identifying EPE and SVI. For the T-staging evaluation of intermediate to high-risk prostate cancer, mpMRI should still be considered the imaging modality of reference. Whenever available, PSMA PET/MRI or the coregistration or fusion of PSMA PET/CT and mpMRI (PSMA PET/CT + mpMRI) should be used as it improves tumor extent delineation.

BackgroundMultiparametric MRI (mpMRI) is commonly recommended as a triage test prior to any prostate biopsy. However, there exists limited consensus on which patients with a negative prostate mpMRI could avoid prostate biopsy.PurposeTo identify which patient could safely avoid prostate biopsy when the prostate mpMRI is negative, via a radiomics-based machine learning approach.Study typeRetrospective.SubjectsThree hundred thirty patients with negative prostate 3T mpMRI between January 2016 and December 2018 were included.Field strength/sequenceA 3.0 T/T2-weighted turbo spin echo (TSE) imaging (T2 WI) and diffusion-weighted imaging (DWI).AssessmentThe integrative machine learning (iML) model was trained to predict negative prostate biopsy results, utilizing both radiomics and clinical features. The final study cohort comprised 330 consecutive patients with negative mpMRI (PI-RADS

IntroductionThe off-label use of ferumoxytol as a vascular MR imaging agent is growing rapidly. However, the properties of ferumoxytol suggest that it may play an important role in the detection and characterization of abdominal mass lesions.MethodsThirty-six patients with benign abdominal mass lesions who underwent MR angiography with ferumoxytol also had T2-weighted HASTE imaging and fat-suppressed 3D T1-weighted imaging. The T1 and T2 enhancement characteristics of the lesions were analyzed and correlated with other imaging modalities and/or surgical findings and/or clinical follow-up.ResultsIn all patients with benign masses in the liver (n = 22 patients), spleen (n = 6 patients), kidneys (n = 33 patients), adrenal (n = 2 patients) and pancreas (n = 4 patients), based on the enhancement characteristics with ferumoxytol, readers were confident of the benign nature of the lesions and their conclusions were consistent with correlative imaging, tissue sampling and follow-up. One patient with a suspicious enhancing 2F Bosniak renal cyst had renal cell carcinoma confirmed on biopsy.ConclusionFerumoxytol-enhanced MRI can increase diagnostic confidence for benign abdominal masses and can increase the conspicuity of mass lesions, relative to unenhanced MRI.

The current standardized scheme for interpreting MRI requires a high level of expertise and exhibits a significant degree of inter-reader and intra-reader variability. An automated prostate cancer (PCa) classification can improve the ability of MRI to assess the spectrum of PCa. The purpose of the study was to evaluate the performance of a texture-based deep learning model (Textured-DL) for differentiating between clinically significant PCa (csPCa) and non-csPCa and to compare the Textured-DL with Prostate Imaging Reporting and Data System (PI-RADS)-based classification (PI-RADS-CLA), where a threshold of PI-RADS ≥ 4, representing highly suspicious lesions for csPCa, was applied. The study cohort included 402 patients (60% (n = 239) of patients for training, 10% (n = 42) for validation, and 30% (n = 121) for testing) with 3T multiparametric MRI matched with whole-mount histopathology after radical prostatectomy. For a given suspicious prostate lesion, the volumetric patches of T2-Weighted MRI and apparent diffusion coefficient images were cropped and used as the input to Textured-DL, consisting of a 3D gray-level co-occurrence matrix extractor and a CNN. PI-RADS-CLA by an expert reader served as a baseline to compare classification performance with Textured-DL in differentiating csPCa from non-csPCa. Sensitivity and specificity comparisons were performed using Mcnemar's test. Bootstrapping with 1000 samples was performed to estimate the 95% confidence interval (CI) for AUC. CIs of sensitivity and specificity were calculated by the Wald method. The Textured-DL model achieved an AUC of 0.85 (CI [0.79, 0.91]), which was significantly higher than the PI-RADS-CLA (AUC of 0.73 (CI [0.65, 0.80]); p < 0.05) for PCa classification, and the specificity was significantly different between Textured-DL and PI-RADS-CLA (0.70 (CI [0.59, 0.82]) vs. 0.47 (CI [0.35, 0.59]); p < 0.05). In sub-analyses, Textured-DL demonstrated significantly higher specificities in the peripheral zone (PZ) and solitary tumor lesions compared to the PI-RADS-CLA (0.78 (CI [0.66, 0.90]) vs. 0.42 (CI [0.28, 0.57]); 0.75 (CI [0.54, 0.96]) vs. 0.38 [0.14, 0.61]; all p values < 0.05). Moreover, Textured-DL demonstrated a high negative predictive value of 92% while maintaining a high positive predictive value of 58% among the lesions with a PI-RADS score of 3. In conclusion, the Textured-DL model was superior to the PI-RADS-CLA in the classification of PCa. In addition, Textured-DL demonstrated superior performance in the specificities for the peripheral zone and solitary tumors compared with PI-RADS-based risk assessment.

PurposeThe appropriate number of systematic biopsy cores to retrieve during magnetic resonance imaging (MRI)-targeted prostate biopsy is not well defined. We aimed to demonstrate a biopsy sampling approach that reduces required core count while maintaining diagnostic performance.Materials and methodsWe collected data from a cohort of 971 men who underwent MRI-ultrasound fusion targeted biopsy for suspected prostate cancer. A regional targeted biopsy (RTB) was evaluated retrospectively; only cores within 2 cm of the margin of a radiologist-defined region of interest were considered part of the RTB. We compared detection rates for clinically significant prostate cancer (csPCa) and cancer upgrading rate on final whole mount pathology after prostatectomy between RTB, combined, MRI-targeted, and systematic biopsy.ResultsA total of 16,459 total cores from 971 men were included in the study data sets, of which 1,535 (9%) contained csPCa. The csPCa detection rates for systematic, MRI-targeted, combined, and RTB were 27.0% (262/971), 38.3% (372/971), 44.8% (435/971), and 44.0% (427/971), respectively. Combined biopsy detected significantly more csPCa than systematic and MRI-targeted biopsy (p

ObjectiveTo demonstrate enabling multi-institutional training without centralizing or sharing the underlying physical data via federated learning (FL).Materials and methodsDeep learning models were trained at each participating institution using local clinical data, and an additional model was trained using FL across all of the institutions.ResultsWe found that the FL model exhibited superior performance and generalizability to the models trained at single institutions, with an overall performance level that was significantly better than that of any of the institutional models alone when evaluated on held-out test sets from each institution and an outside challenge dataset.DiscussionThe power of FL was successfully demonstrated across 3 academic institutions while avoiding the privacy risk associated with the transfer and pooling of patient data.ConclusionFederated learning is an effective methodology that merits further study to enable accelerated development of models across institutions, enabling greater generalizability in clinical use.

Multi-parametric MRI (mpMRI) is a powerful non-invasive tool for diagnosing prostate cancer (PCa) and is widely recommended to be performed before prostate biopsies. Prostate Imaging Reporting and Data System version (PI-RADS) is used to interpret mpMRI. However, when the pre-biopsy mpMRI is negative, PI-RADS 1 or 2, there exists no consensus on which patients should undergo prostate biopsies. Recently, radiomics has shown great abilities in quantitative imaging analysis with outstanding performance on computer-aid diagnosis tasks. We proposed an integrative radiomics-based approach to predict the prostate biopsy results when pre-biopsy mpMRI is negative. Specifically, the proposed approach combined radiomics features and clinical features with machine learning to stratify positive and negative biopsy groups among negative mpMRI patients. We retrospectively reviewed all clinical prostate MRIs and identified 330 negative mpMRI scans, followed by biopsy results. Our proposed model was trained and validated with 10-fold cross-validation and reached the negative predicted value (NPV) of 0.99, the sensitivity of 0.88, and the specificity of 0.63 in receiver operating characteristic (ROC) analysis. Compared with results from existing methods, ours achieved 11.2% higher NPV and 87.2% higher sensitivity with a cost of 23.2% less specificity.

PurposeDeveloping large-scale datasets with research-quality annotations is challenging due to the high cost of refining clinically generated markup into high precision annotations. We evaluated the direct use of a large dataset with only clinically generated annotations in development of high-performance segmentation models for small research-quality challenge datasets.Materials and methodsWe used a large retrospective dataset from our institution comprised of 1,620 clinically generated segmentations, and two challenge datasets (PROMISE12: 50 patients, ProstateX-2: 99 patients). We trained a 3D U-Net convolutional neural network (CNN) segmentation model using our entire dataset, and used that model as a template to train models on the challenge datasets. We also trained versions of the template model using ablated proportions of our dataset, and evaluated the relative benefit of those templates for the final models. Finally, we trained a version of the template model using an out-of-domain brain cancer dataset, and evaluated the relevant benefit of that template for the final models. We used five-fold cross-validation (CV) for all training and evaluation across our entire dataset.ResultsOur model achieves state-of-the-art performance on our large dataset (mean overall Dice 0.916, average Hausdorff distance 0.135 across CV folds). Using this model as a pre-trained template for refining on two external datasets significantly enhanced performance (30% and 49% enhancement in Dice scores respectively). Mean overall Dice and mean average Hausdorff distance were 0.912 and 0.15 for the ProstateX-2 dataset, and 0.852 and 0.581 for the PROMISE12 dataset. Using even small quantities of data to train the template enhanced performance, with significant improvements using 5% or more of the data.ConclusionWe trained a state-of-the-art model using unrefined clinical prostate annotations and found that its use as a template model significantly improved performance in other prostate segmentation tasks, even when trained with only 5% of the original dataset.

Nonalcoholic fatty liver disease (NAFLD)-associated hepatocellular carcinoma (HCC) is projected to become the leading indication for liver transplantation. Previous studies indicate that tumor growth rates (TGR) may predict survival and were helpful in determining HCC surveillance intervals. Therefore, we aimed to determine its usefulness in predicting clinical outcomes and treatments. We conducted a retrospective study of hepatitis B, C and NAFLD-HCC cases. TGR was measured using 2-consecutive pre-treatment contrast-enhanced imaging studies ≥ 25 days apart. A multivariate regression model was used to determine predictors of TGR. In addition, the Cox regression model was used to evaluate the relationship between TGR and overall survival. From 2000-2019, the study cohort comprised 38, 60, and 47 HBV, HCV, and NAFLD patients, respectively, with TGRs. NAFLD-HCC tumor size was inversely correlated to the extent of liver disease as measured by Child-Pugh score (7.2 cm in non-cirrhosis; 3.7 cm, 2.6 cm, and 2.1 cm in Child A, B, and C, respectively; P < 0.001). After adjusting for baseline characteristics, the TGR per month was fastest in HBV (9.4%, 95%CI: 6.3%-12.5%) compared to HCV (4.9%, 95%CI: 2.8%-7%) and NAFLD patients (3.6%, 95%CI: 1.6%-6.7%). Predictors of TGR included elevated AFP, low albumin, and smaller tumor size. Fast TGR in viral etiologies had higher mortality [adj. hazard ratio (HR) = 2.6, 95%CI: 1.2-5.7, P = 0.02] than slow TGRs, independent of treatments. Fast TGR in NAFLD had a trend towards higher mortality (HR = 3.6, 95%CI: 0.95-13.3, P = 0.059). NAFLD-HCC patients have more indolent growths than viral-related HCC TGRs. The addition of TGR as a biomarker may assist in stratifying treatment options.

Purpose: Prostate cancer (PCa) is the most common solid organ cancer and second leading cause of death in men. Multiparametric magnetic resonance imaging (mpMRI) enables detection of the most aggressive, clinically significant PCa (csPCa) tumors that require further treatment. A suspicious region of interest (ROI) detected on mpMRI is now assigned a Prostate Imaging-Reporting and Data System (PIRADS) score to standardize interpretation of mpMRI for PCa detection. However, there is significant inter-reader variability among radiologists in PIRADS score assignment and a minimal input semi-automated artificial intelligence (AI) system is proposed to harmonize PIRADS scores with mpMRI data. Approach: The proposed deep learning model (the seed point model) uses a simulated single-click seed point as input to annotate the lesion on mpMRI. This approach is in contrast to typical medical AI-based approaches that require annotation of the complete lesion. The mpMRI data from 617 patients used in this study were prospectively collected at a major tertiary U.S. medical center. The model was trained and validated to classify whether an mpMRI image had a lesion with a PIRADS score greater than or equal to PIRADS 4. Results: The model yielded an average receiver-operator characteristic (ROC) area under the curve (ROC-AUC) of 0.704 over a 10-fold cross-validation, which is significantly higher than the previously published benchmark. Conclusions: The proposed model could aid in PIRADS scoring of mpMRI, providing second reads to promote quality as well as offering expertise in environments that lack a radiologist with training in prostate mpMRI interpretation. The model could help identify tumors with a higher PIRADS for better clinical management and treatment of PCa patients at an early stage.

PurposeTo characterize the effect of hepatic vessel flow using 4-dimensional (4D) flow magnetic resonance (MR) imaging and correlate their effect on microwave ablation volumes in an in vivo non-cirrhotic porcine liver model.Materials and methodsMicrowave ablation antennas were placed under ultrasound guidance in each liver lobe of swine (n = 3 in each animal) for a total of 9 ablations. Pre- and post-ablation 4D flow MR imaging was acquired to quantify flow changes in the hepatic vasculature. Flow measurements, along with encompassed vessel size and vessel-antenna spacing, were then correlated with final ablation volume from segmented MR images.ResultsThe linear regression model demonstrated that the preablation measurement of encompassed hepatic vein size (β = -0.80 ± 0.25, 95% confidence interval [CI] -1.15 to -0.22; P = .02) was significantly correlated to final ablation zone volume. The addition of hepatic vein flow rate found via 4D flow MRI (β = -0.83 ± 0.65, 95% CI -2.50 to 0.84; P = .26), and distance from antenna to hepatic vein (β = 0.26 ± 0.26, 95% CI -0.40 to 0.92; P = .36) improved the model accuracy but not significantly so (multivariate adjusted R2 = 0.70 vs univariate (vessel size) adjusted R2 = 0.63, P = .24).ConclusionsHepatic vein size in an encompassed ablation zone was found to be significantly correlated with final ablation zone volume. Although the univariate 4D flow MR imaging-acquired measurements alone were not found to be statistically significant, its addition to hepatic vein size improved the accuracy of the ablation volume regression model. Pre-ablation 4D flow MR imaging of the liver may assist in prospectively optimizing thermal ablation treatment.

PurposeTo investigate the role of qualitative and quantitative DCE-MRI parameters in prostate cancer (PCa) stratified by whole-mount histopathology (WMHP) Gleason score (GS) and PI-RADSv2.MethodsThis retrospective study included 323 PCa tumors in 254 men, who underwent 3T MRI prior to prostatectomy, 7/2009-12/2016. Qualitative DCE curve types included type 1 (progressive), type 2 (plateau) and type 3 (washout). Quantitative DCE-MRI pharmacokinetic (PK) parameters included Ktrans (influx volume transfer coefficient), Kep (efflux reflux rate constant) and iAUC (initial area under the curve). DCE-MRI features of true positive lesions were evaluated for overall, index, transition zone (TZ) and peripheral zone (PZ), based on GS grade (low = 6, high > 6) and PI-RADSv2 score using SPSSv24.ResultsThere were 57 (17.6%) low-grade and 266 (82.4%) high-grade PCa lesions. PI-RADSv2 3, 4 and 5 included 106, 120 and 97 lesions, respectively. 251 (77.7%) and 72 (22.3%) lesions were located in PZ and TZ, respectively. High-grade lesions had significantly higher proportion of Type 3 curves compared to low-grade lesions in overall (70.3% vs. 54.4%) and TZ (73.5% vs. 43.5%). As PI-RADSv2 increased, the proportion of type 3 curve significantly increased for overall (80.4-51.9%), index (80.4-54.7%) and PZ (78.7-52.1%) lesions. Among PK parameters, Ktrans (0.43 vs 0.32) and iAUC (8.99 vs 6.9) for overall PCa, Ktrans (0.43 vs 0.31) and iAUC (9 vs 6.67) for PZ PCa, and iAUC (8.94 vs 7.42) for index PCa were significantly higher for high-grade versus low-grade lesions. Also, Ktrans (0.51-0.34), Kep (1.75-1.29) and iAUC (9.79-7.6) for overall PCa, Ktrans (0.53-0.32), Kep (1.81-1.26) and iAUC (9.83-7.34) for PZ PCa; and Kep (1.79-1.17) and iAUC (11.3-8.45) for index PCa increased significantly with a higher PI-RADSv2 score.ConclusionsThe results of study show the possible utility of qualitative and quantitative DCE-MRI parameters for assessment of PCa GS and PI-RADSv2 categorization.

Arteriovenous malformation (AVM) is a rare lesion in the uterus, which can lead to abnormal uterine bleeding. While AVM has been described in other organs in the literature, there is a paucity of pathology reports of the AVM in uterus. On gross examination, the uterus was markedly enlarged and partly distorted with a pedunculated solid mass, which on the cut surface showed multiple well-circumscribed hemorrhagic cysts ranging from 0.1 to 4.0 cm in size. Microscopically, they were malformed dilated vascular structures containing organized thrombi. We present this case of uterine AVM with gross and microscopic findings, which can serve as a crucial reminder for pathologists to keep in the differential diagnoses as a potential cause of abnormal uterine bleeding.

Automatic segmentation of prostatic zones on multiparametric MRI (mpMRI) can improve the diagnostic workflow of prostate cancer. We designed a spatial attentive Bayesian deep learning network for the automatic segmentation of the peripheral zone (PZ) and transition zone (TZ) of the prostate with uncertainty estimation. The proposed method was evaluated by using internal and external independent testing datasets, and overall uncertainties of the proposed model were calculated at different prostate locations (apex, middle, and base). The study cohort included 351 MRI scans, of which 304 scans were retrieved from a de-identified publicly available datasets (PROSTATEX) and 47 scans were extracted from a large U.S. tertiary referral center (external testing dataset; ETD)). All the PZ and TZ contours were drawn by research fellows under the supervision of expert genitourinary radiologists. Within the PROSTATEX dataset, 259 and 45 patients (internal testing dataset; ITD) were used to develop and validate the model. Then, the model was tested independently using the ETD only. The segmentation performance was evaluated using the Dice Similarity Coefficient (DSC). For PZ and TZ segmentation, the proposed method achieved mean DSCs of 0.80±0.05 and 0.89±0.04 on ITD, as well as 0.79±0.06 and 0.87±0.07 on ETD. For both PZ and TZ, there was no significant difference between ITD and ETD for the proposed method. This DL-based method enabled the accuracy of the PZ and TZ segmentation, which outperformed the state-of-art methods (Deeplab V3+, Attention U-Net, R2U-Net, USE-Net and U-Net). We observed that segmentation uncertainty peaked at the junction between PZ, TZ and AFS. Also, the overall uncertainties were highly consistent with the actual model performance between PZ and TZ at three clinically relevant locations of the prostate.

OBJECTIVE. The purpose of this article is to evaluate restriction spectrum imaging (RSI) in men undergoing MRI-ultrasound fusion biopsy for suspected prostate cancer (PCa) and to compare the performance of RSI with that of conventional DWI. MATERIALS AND METHODS. One hundred ninety-eight biopsy-naïve men enrolled in a concurrent prospective clinical trial evaluating MRI-targeted prostate biopsy underwent multiparametric MRI with RSI. Clinical and imaging features were compared between men with and without clinically significant (CS) PCa (MRI-ultrasound fusion biopsy Gleason score ≥ 3 + 4). RSI z score and apparent diffusion coefficient (ADC) were correlated, and their diagnostic performances were compared. RESULTS. CS PCa was detected in 109 of 198 men (55%). Using predefined thresholds of ADC less than or equal to 1000 μm2/s and RSI z score greater than or equal to 3, sensitivity and specificity for CS PCa were 86% and 38%, respectively, for ADC and 61% and 70%, respectively, for RSI. In the transition zone (n = 69), the sensitivity and specificity were 94% and 17%, respectively, for ADC and 59% and 69%, respectively, for RSI. Among lesions with CS PCa, RSI z score and ADC were significantly inversely correlated in the peripheral zone (ρ = -0.4852; p < 0.01) but not the transition zone (ρ = -0.2412; p = 0.17). Overall diagnostic accuracies of RSI and DWI were 0.70 and 0.68, respectively (p = 0.74). CONCLUSION. RSI and DWI achieved equivalent diagnostic performance for PCa detection in a large population of men undergoing first-time prostate biopsy for suspected PCa, but RSI had superior specificity for transition zone lesions.

Cystic renal cell carcinoma (RCC) is almost certainly overdiagnosed and overtreated. Efforts to diagnose and treat RCC at a curable stage result in many benign neoplasms and indolent cancers being resected without clear benefit. This is especially true for cystic masses, which compared with solid masses are more likely to be benign and, when malignant, less aggressive. For more than 30 years, the Bosniak classification has been used to stratify the risk of malignancy in cystic renal masses. Although it is widely used and still effective, the classification does not formally incorporate masses identified at MRI or US or masses that are incompletely characterized but are highly likely to be benign, and it is affected by interreader variability and variable reported malignancy rates. The Bosniak classification system cannot fully differentiate aggressive from indolent cancers and results in many benign masses being resected. This proposed update to the Bosniak classification addresses some of these shortcomings. The primary modifications incorporate MRI, establish definitions for previously vague imaging terms, and enable a greater proportion of masses to enter lower-risk classes. Although the update will require validation, it aims to expand the number of cystic masses to which the Bosniak classification can be applied while improving its precision and accuracy for the likelihood of cancer in each class.

Multiparametric magnetic resonance imaging (mpMRI) has transformed the management of localized prostate cancer by improving identification of clinically significant disease at diagnosis. Approximately 20% of primary prostate tumors are invisible to mpMRI, and we hypothesize that this invisibility reflects fundamental molecular properties of the tumor. We therefore profiled the genomes and transcriptomes of 40 International Society of Urological Pathology grade 2 tumors: 20 mpMRI-invisible (Prostate Imaging-Reporting and Data System [PI-RADS] v2

PurposeTo evaluate 3T mpMRI characteristics of transition zone and peripheral zone index prostate cancer lesions stratified by Gleason Score and PI-RADSv2 with whole mount histopathology correlation.MethodsAn institution review board-approved, HIPAA-compliant single-arm observational study of 425 consecutive men with 3T mpMRI prior to radical prostatectomy from December 2009 to October 2016 was performed. A genitourinary radiologist and a genitourinary pathologist matched all lesions detected on whole mount histopathology with lesions concordant for size and location on 3T mpMRI. Differences in clinical, MRI parameters, and histopathology between transition zone and peripheral zone were determined and analyzed with χ2 and Mann-Whitney U test. AUC was measured.Results3T mpMRI detected 248/323 (76.7%) index lesions in peripheral zone and 75/323 (23.2%) in transition zone. Transition zone prostate cancer had higher median prostate-specific antigen (p = 0.001), larger tumor on 3T mpMRI (p = 0.001), lower proportions of PI-RADSv2 category 4 and 5 (p

PurposeWe present a method for generating a T2 MR-based probabilistic model of tumor occurrence in the prostate to guide the selection of anatomical sites for targeted biopsies and serve as a diagnostic tool to aid radiological evaluation of prostate cancer.Materials and methodsIn our study, the prostate and any radiological findings within were segmented retrospectively on 3D T2-weighted MR images of 266 subjects who underwent radical prostatectomy. Subsequent histopathological analysis determined both the ground truth and the Gleason grade of the tumors. A randomly chosen subset of 19 subjects was used to generate a multi-subject-derived prostate template. Subsequently, a cascading registration algorithm involving both affine and non-rigid B-spline transforms was used to register the prostate of every subject to the template. Corresponding transformation of radiological findings yielded a population-based probabilistic model of tumor occurrence. The quality of our probabilistic model building approach was statistically evaluated by measuring the proportion of correct placements of tumors in the prostate template, i.e., the number of tumors that maintained their anatomical location within the prostate after their transformation into the prostate template space.ResultsProbabilistic model built with tumors deemed clinically significant demonstrated a heterogeneous distribution of tumors, with higher likelihood of tumor occurrence at the mid-gland anterior transition zone and the base-to-mid-gland posterior peripheral zones. Of 250 MR lesions analyzed, 248 maintained their original anatomical location with respect to the prostate zones after transformation to the prostate.ConclusionWe present a robust method for generating a probabilistic model of tumor occurrence in the prostate that could aid clinical decision making, such as selection of anatomical sites for MR-guided prostate biopsies.