Prostate Magnetic Resonance Imaging Research Articles

Association of quantifiable prostate MRI parameters with any and large cribriform pattern in prostate cancer patients undergoing radical prostatectomy

PurposeCribriform pattern has recently been recognized as an important independent risk factor for prostate cancer (PCa) outcome. This study aimed to identify the association of quantifiable prostate magnetic resonance imaging (MRI) parameters with any and large cribriform pattern at radical prostatectomy (RP) specimens. MethodsPreoperative prostate MRI’s from 188 men undergoing RP between 2010 and 2018 were retrospectively acquired. RP specimens of the patients were revised for Gleason score (GS), and presence of any and large cribriform pattern. MRI parameters such as MRI visibility, PI-RADS score, lowest apparent diffusion coefficient (ADC) value, lesion size, and radiologic extra-prostatic extension (EPE) were reviewed. The association of prostate MRI parameters for presence of any and large cribriform pattern at RP was analysed using logistic regression. Results116/188 (61.7%) PCa patients had any cribriform and 36/188 (19.1%) large cribriform pattern at RP. 171/188 (91.0%) men had MRI-visible lesions; 111/116 (95.7%) tumours with any and 36/36 (100%) with large cribriform pattern were visible at MRI. PCa with any and large cribriform pattern both had lower ADC values than those without (p < 0.001). In adjusted analysis, lowest ADC value was as an independent predictor for any cribriform (Odds Ratio (OR) 0.2, 95% Confidence Interval (CI) 0.1–0.8; p = 0.01) and large cribriform pattern (OR 0.2, 95% CI 0.1–0.7; p = 0.01), while other parameters were not. ConclusionsThe majority of PCa with cribriform pattern at RP were visible at MRI, and lowest ADC value was an independent predictor for both any and large cribriform pattern.

Reliable Visualization of the Treatment Effect of Transperineal Focal Laser Ablation in Prostate Cancer Patients by Magnetic Resonance Imaging and Contrast-enhanced Ultrasound Imaging

BackgroundTransperineal focal laser ablation (TPLA) treatment for prostate cancer (PCa) is an experimental focal ablative therapy modality with low morbidity. However, a dosimetry model for TPLA is lacking. ObjectiveTo determine (1) the three-dimensional (3D) histologically defined ablation zone of single- and multifiber TPLA treatment for PCa correlated with magnetic resonance imaging (MRI) and contrast-enhanced ultrasound (CEUS) and (2) a reliable imaging modality of ablation zone volumetry. Design, setting, and participantsThis was a prospective, multicenter, and interventional phase I/II pilot study with an ablate-and-resect design. TPLA was performed in 12 patients with localized prostate cancer divided over four treatment regimens to evaluate potential variation in outcomes. InterventionTPLA was performed approximately 4 wk prior to robot-assisted radical prostatectomy (RARP) in a daycare setting using local anesthesia. Outcome measurements and statistical analysisFour weeks after TPLA, ablation zone volumetry was determined on prostate MRI and CEUS by delineation and segmentation into 3D models and correlated with whole-mount RARP histology using the Pearson correlation index. Results and limitationsTwelve office-based TPLA procedures were performed successfully under continuous transrectal ultrasound guidance using local perineal anesthesia. No serious adverse events occurred. A qualitative analysis showed a clear demarcation of the ablation zone on T2-weighted MRI, dynamic contrast-enhanced MRI, and CEUS. On pathological evaluation, no remnant cancer was observed within the ablation zone. Ablation zone volumetry on CEUS and T2-weighted MRI compared with histology had a Pearson correlation index of r = 0.94 (95% confidence interval [CI] 0.74–0.99, p < 0.001) and r = 0.93 (95% CI 0.73–0.98, p < 0.001), respectively. ConclusionsCEUS and prostate MRI could reliably visualize TPLA ablative effects after minimally invasive PCa treatment with a high concordance with histopathological findings and showed no remnant cancer. Patient summaryThe treatment effects of a novel minimally invasive ablation therapy device can reliably be visualized with radiological examinations. These results will improve planning and performance of future procedures.

Radiomics based automated quality assessment for T2W prostate MR images

PurposeThe guidelines for prostate cancer recommend the use of MRI in the prostate cancer pathway. Due to the variability in prostate MR image quality, the reliability of this technique in the detection of prostate cancer is highly variable in clinical practice. This leads to the need for an objective and automated assessment of image quality to ensure an adequate acquisition and hereby to improve the reliability of MRI. The aim of this study is to investigate the feasibility of Blind/referenceless image spatial quality evaluator (Brisque) and radiomics in automated image quality assessment of T2-weighted (T2W) images. MethodAnonymized axial T2W images from 140 patients were scored for quality using a five-point Likert scale (low, suboptimal, acceptable, good, very good quality) in consensus by two readers. Images were dichotomized into clinically acceptable (very good, good and acceptable quality images) and clinically unacceptable (low and suboptimal quality images) in order to train and verify the model. Radiomics and Brisque features were extracted from a central cuboid volume including the prostate. A reduced feature set was used to fit a Linear Discriminant Analysis (LDA) model to predict image quality. Two hundred times repeated 5-fold cross-validation was used to train the model and test performance by assessing the classification accuracy, the discrimination accuracy as receiver operating curve - area under curve (ROC-AUC), and by generating confusion matrices. ResultsThirty-four images were classified as clinically unacceptable and 106 were classified as clinically acceptable. The accuracy of the independent test set (mean ± standard deviation) was 85.4 ± 5.5%. The ROC-AUC was 0.856 (0.851 – 0.861) (mean; 95% confidence interval). ConclusionsRadiomics AI can automatically detect a significant portion of T2W images of suboptimal image quality. This can help improve image quality at the time of acquisition, thus reducing repeat scans and improving diagnostic accuracy.

Current Trends in Incidence and Management of T1a and T1b Prostate Cancer.

Prostate cancer (PCa) identified incidentally (iPCa) after surgical treatment for symptomatic benign prostatic hyperplasia (BPH) causing lower urinary tract symptoms (LUTS) is considered low risk by the most current guidelines. Management protocols for iPCa are conservative and are identical to other prostate cancers classified as having favorable prognoses. The objectives of this paper are to discuss the incidence of iPCa stratified by BPH procedure, to highlight predictors of cancer progression, and to propose potential modifications to mainstream guidelines for the optimal management of iPCa. The correlation between the rate of iPCa detection and the method of BPH surgery is not clearly defined. Old age, small prostate volume, and high pre-operative prostate-specific antigen (PSA) are associated with an increased likelihood of detecting iPCa. PSA and tumor grade are strong predictors of cancer progression and can be used along with magnetic resonance imaging (MRI) and potential confirmatory biopsies to determine disease management. In instances that iPCa requires treatment, radical prostatectomy (RP), radiation therapy, and androgen deprivation therapy all have oncologic benefits but may be associated with increased risk after the BPH surgery. It is advised that patients with low to favorable intermediate-risk prostate cancer undergo post-operative PSA measurement and prostate MRI imaging before electing to choose between observation, surveillance without confirmatory biopsy, immediate confirmatory biopsy, or active treatment. Subdividing the binary T1a/b cancer staging into more categories with ranging percentages of malignant tissue would be a helpful first step in tailoring the management of iPCa.

Racial differences in the utilization of MRI-guided biopsy before prostate cancer diagnosis.

5093 Background: Most low-risk prostate cancer (PCa) patients undergo active surveillance (AS). Compared with Whites, Blacks are more likely to opt for AS and suffer a higher risk of cancer progression while on AS. A primary concern of AS is the likelihood of missing clinically significant cancer. Transrectal ultrasound-guided biopsy is the most common method for diagnosing and grading PCa; however, this method often misses anterior PCa. Magnetic resonance imaging (MRI) allows clinicians to identify better and detect PCa; however, current guidelines on MRI-guided biopsy (MRI-GB) are inconsistent. This population-based study aimed to examine the utilization of MRI-GB 6 months before PCA diagnosis over time by race in the US. Methods: We used SEER-Medicare data to identify men diagnosed with PCa between January 1, 2012, to December 31, 2017, and excluded patients missing the date of diagnosis, being diagnosed at death, enrolled in HMO, or having no continuous A and B coverage six months before diagnosis. We used the CPT codes to identify prostate biopsy and MRI use. Multivariable logistic regression was used to evaluate racial disparity in MRI use six months before PCa. The model included covariates age, race, SEER region, marital status, state buy-in for Medicare (a poverty indicator), education, income, Charlson comorbidity index, and procedure (MRI) year. Results: Among 38,612 eligible men, MRI use six months before PCa diagnosis increased from 2012 to 2017 (3.2% to 24.3% among Whites, 1.8% to 14.2% among Blacks). Compared with Whites, Blacks were 38% less likely to receive pre-diagnosis MRI (OR=0.62, 95% CI 0.52-0.74). Besides race, having state buy-in for Medicare (OR=0.44; 95% CI 0.30-0.65) and geographic regions (Central vs. West OR=0.40; 95% CI 0.34-0.45) were strong predictors of lower utilization of MRI. Conclusions: The gap in the utilization of MRI may exacerbate racial disparities in PCa. Future studies are needed to identify potential barriers to adopting MRI for PCa workups.

Future of prostate imaging: Artificial intelligence in assessing prostatic magnetic resonance imaging.

Prostate cancer (Pca; adenocarcinoma) is one of the most common cancers in adult males and one of the leading causes of death in both men and women. The diagnosis of Pca requires substantial experience, and even then the lesions can be difficult to detect. Moreover, although the diagnostic approach for this disease has improved significantly with the advent of multiparametric magnetic resonance, that technology has certain unresolved limitations. In recent years artificial intelligence (AI) has been introduced to the field of radiology, providing new software solutions for prostate diagnostics. Precise mapping of the prostate has become possible through AI and this has greatly improved the accuracy of biopsy. AI has also allowed for certain suspicious lesions to be attributed to a given group according to the Prostate Imaging-Reporting & Data System classification. Finally, AI has facilitated the combination of data obtained from clinical, laboratory (prostate-specific antigen), imaging (magnetic resonance), and biopsy examinations, and in this way new regularities can be found which at the moment remain hidden. Further evolution of AI in this field is inevitable and it is almost certain to significantly expand the efficacy, accuracy and efficiency of diagnosis and treatment of Pca.

A Light, 3D UNet-based Architecture for Fully Automatic Segmentation of Prostate Lesions from T2-MRI Images.

Prostate magnetic resonance imaging (MRI) has been recently integrated into the pathway of diagnosis of prostate cancer (PCa). However, the lack of an optimal contrast-to-noise ratio hinders automatic recognition of suspicious lesions, thus developing a solution for proper delimitation of the tumour and its separation from the healthy parenchyma, which is of primordial importance. As a solution to this unmet medical need, we aimed to develop a decision support system based on artificial intelligence, which automatically segments the prostate and any suspect area from the 3D MRI images. We assessed retrospective data from all patients diagnosed with PCa by MRI-US fusion prostate biopsy, who underwent prostate MRI in our department due to a clinical or biochemical suspicion of PCa (n=33). All examinations were performed using a 1.5 Tesla MRI scanner. All images were reviewed by two radiologists, who performed manual segmentation of the prostate and all lesions. A total of 145 augmented datasets were generated. The performance of our fully automated end-to-end segmentation model based on a 3D UNet architecture and trained in two learning scenarios (on 14 or 28 patient datasets) was evaluated by two loss functions. Our model had an accuracy of over 90% for automatic segmentation of prostate and PCa nodules, as compared to manual segmentation. We have shown low complexity networks, UNet architecture with less than five layers, as feasible and to show good performance for automatic 3D MRI image segmentation. A larger training dataset could further improve the results. Therefore, herein, we propose a less complex network, a slim 3D UNet with superior performance, being faster than the original five-layer UNet architecture.

Feasibility of clinical studies of chemical exchange saturation transfer magnetic resonance imaging of prostate cancer at 7 T.

Chemical exchange saturation transfer (CEST) has been explored for differentiation between tumour and benign tissue in prostate cancer (PCa) patients. With ultrahigh field strengths such as 7-T, the increase of spectral resolution and sensitivity could allow for selective detection of amide proton transfer (APT) at 3.5 ppm and a group of compounds that resonate at 2 ppm (i.e., [poly]amines and/or creatine). The potential of 7-T multipool CEST analysis of the prostate and the detection of PCa was studied in patients with proven localised PCa who were scheduled to undergo robot-assisted radical prostatectomy (RARP). Twelve patients were prospectively included (mean age 68.0 years, mean serum prostate-specific antigen 7.8ng/mL). A total of 24 lesions larger than 2 mm were analysed. Used were 7-T T2-weighted (T2W) imaging and 48 spectral CEST points. Patients received 1.5-T/3-T prostate magnetic resonance imaging and galium-68-prostate-specific membrane antigen-positron emission tomography/computerised tomography to determine the location of the single-slice CEST. Based on the histopathological results after RARP, three regions of interest were drawn on the T2W images from a known malignant zone and benign zone in the central and peripheral zones. These areas were transposed to the CEST data, from which the APT and 2-ppm CEST were calculated. The statistical significance of the CEST between the central zone, the peripheral zone, and tumour was calculated using a Kruskal-Wallis test. The z-spectra showed that APT and even a distinct pool that resonated at 2 ppm were detectable. This study showed a difference trend in the APT levels, but no difference in the 2-ppm levels when tested between the central zone, the peripheral zone, and tumour (H(2) = 4.8, p= 0.093 and H(2) = 0.86, p= 0.651, respectively). Thus, to conclude, we could most likely detect APT and amines and/or creatine levels noninvasively in prostate using the CEST effect. At group level, CEST showed a higher level of APT in the peripheral versus the central zone; however, no differences of APT and 2-ppm levels were observed in tumours.

The combined role of MRI prostate and prostate health index in improving detection of significant prostate cancer in a screening population of Chinese men.

Using prostate-specific antigen (PSA) for prostate cancer (PCa) screening led to overinvestigation and overdiagnosis of indolent PCa. We aimed to investigate the value of prostate health index (PHI) and magnetic resonance imaging (MRI) prostate in an Asian PCa screening program. Men aged 50-75 years were prospectively recruited from a community-based PSA screening program. Men with PSA 4.0-10.0 ng ml -1 had PHI result analyzed. MRI prostate was offered to men with PSA 4.0-50.0 ng ml -1 . A systematic prostate biopsy was offered to men with PSA 4.0-9.9 ng ml -1 and PHI ≥35, or PSA 10.0-50.0 ng ml -1 . Additional targeted prostate biopsy was offered if they had PI-RADS score ≥3. Clinically significant PCa (csPCa) was defined as the International Society of Urological Pathology (ISUP) grade group (GG) ≥2 or ISUP GG 1 with involvement of ≥30% of total systematic cores. In total, 12.8% (196/1536) men had PSA ≥4.0 ng ml -1 . Among 194 men with PSA 4.0-50.0 ng ml -1 , 187 (96.4%) received MRI prostate. Among them, 28.3% (53/187) had PI-RADS ≥3 lesions. Moreover, 7.0% (107/1536) men were indicated for biopsy and 94.4% (101/107) men received biopsy. Among the men received biopsy, PCa, ISUP GG ≥2 PCa, and csPCa was diagnosed in 42 (41.6%), 24 (23.8%), and 34 (33.7%) men, respectively. Compared with PSA/PHI pathway in men with PSA 4.0-50.0 ng ml -1 , additional MRI increased diagnoses of PCa, ISUP GG ≥2 PCa, and csPCa by 21.2% (from 33 to 40), 22.2% (from 18 to 22), and 18.5% (from 27 to 32), respectively. The benefit of additional MRI was only observed in PSA 4.0-10.0 ng ml -1 , and the number of MRI needed to diagnose one additional ISUP GG ≥2 PCa was 20 in PHI ≥35 and 94 in PHI <35. Among them, 45.4% (89/196) men with PSA ≥4.0 ng ml -1 avoided unnecessary biopsy with the use of PHI and MRI. A screening algorithm with PSA, PHI, and MRI could effectively diagnose csPCa while reducing unnecessary biopsies. The benefit of MRI prostate was mainly observed in PSA 4.0-9.9 ng ml -1 and PHI ≥35 group. PHI was an important risk stratification step for PCa screening.

EDITORIAL COMMENT

The authors report on the largest clinical evaluation of Prostate Health Index (PHI) assay in African American (AA) men for the prediction of Gleason grade group (GG) 2-5 prostate cancer (PCa). 1 Morris K Kotamarti S Polascik T Moul J. Re-thinking how we use prostate health index for African American Men. Urology. 2023; (S0090-4295(23)00176-0) Abstract Full Text Full Text PDF Scopus (0) Google Scholar They report that the required threshold to achieve a sensitivity >90% coincides with a poor specificity; this only permits 4.7% of Black men to avoid a biopsy even though over half of biopsies were negative or GG 1 PCa. Their findings call in to question whether the assay is accurate enough to be relied upon for Black men. There is, however, a potential for bias from its retrospective and heterogeneous nature. The fact that 18.2% of men did not proceed to biopsy, 36.4% had prostate Magnetic Resonance Imaging (MRI), 13.4% had a fusion biopsy and men were seen in a tertiary care setting could enrich the population for any PCa and GG 2-5 PCa and overestimate risk for each stratum, including the 0-26.9 strata where the authors detected any PCa (48.0%) in 13 of 22 fusion biopsy patients. A biopsy-naïve/MRI-naïve cohort would be a better comparator, since the risk strata that PHI uses is based on a cohort that predates the use of MRI Prostate Imaging–Reporting and Data System (PIRADS) scores. 2 Catalona WJ Partin AW Sanda MG et al. A multi-center study of [− 2] pro41 prostate-specific antigen (PSA) in combination with PSA and free PSA for prostate cancer detection in the 2.0 to 10.0 ng/mL PSA range. J Urol. 2011; 185: 1650 Crossref PubMed Scopus (375) Google Scholar ,3 Barentsz JO Richenberg J Clements R et al. ESUR prostate MR guidelines 2012. Eur Radiol. 2012; 22: 746-757 Crossref PubMed Scopus (1906) Google Scholar Nevertheless, it shows that the value of PHI in a ‘real-world’ setting may be limited. Their findings also dovetail with the study on Black men with elevated prostate-specific antigen (PSA) by Babajide et al where the probability of overall and GG 2-5 PCa were higher at each of the established PHI risk strata. 4 Babajide R Carbunaru S Nettey OS et al. Performance of prostate health index in biopsy naïve black men. J Urol. 2021; 205: 718-724 Crossref PubMed Scopus (9) Google Scholar Data in other racial groups also suggests that optimal thresholds of PHI vary across populations. 5 Tan LG Tan YK Tai BC et al. Prospective validation of %p2PSA and the prostate health index, in prostate cancer detection in initial prostate biopsies of Asian men, with total PSA 4-10 ng ml-1. Asian J Androl. 2017; 19: 286-290 Crossref PubMed Scopus (27) Google Scholar , 6 Ferro M Bruzzese D Perdona S et al. Prostate Health Index (PHI) and Prostate Cancer Antigen 3 (PCA3) significantly improve prostate cancer detection at initial biopsy in a total PSA range of 2–10 ng/ml. PLoS One. 2013; 8: e67687 Crossref PubMed Scopus (87) Google Scholar , 7 Loeb S Sanda MG Broyles DL et al. The prostate health index selectively identifies clinically significant prostate cancer. J Urol. 2015; 193: 1163-1169 Crossref PubMed Scopus (197) Google Scholar

PO-1675 Evaluation of synthetic CTs by 3D Cycle-GAN from prostate MR images

PO-1675 Evaluation of synthetic CTs by 3D Cycle-GAN from prostate MR images

A Multi-Scale Channel Attention Network for Prostate Segmentation

Prostate cancer is one of the most common malignant tumors in men. Magnetic resonance imaging (MRI) has evolved to an important tool for the diagnosis of prostate cancer. Targeted biopsy is required for accurate diagnosis. This often requires MRI-ultrasound (MRI-US) fusion, as the biopsy is usually performed using transrectal ultrasound. Accurate prostate segmentation on MRI is essential for MRI-US fusion biopsy. However, the variation in prostate shape, appearance, and size makes the automatic segmentation challenging, given the limit of the annotated data. In this paper, we propose a method using multi-scale and Channel-wise Self-Attention (CSA) to re-calibrate the feature maps from multiple layers. By embedding the multi-scale CSA on the skip-connection in a UNet structure, called as UCAnet, we show the consistent improvement of the prostate segmentation in Dice, IoU and ASSD. For comparison, we also investigate the single-scale CSA in the networks, and incorporate the vision transformer to test if a transformer would boost the performance. Experiments on a public dataset with 204 prostate MRI scans show that UCAnet achieves the best performance and outperforms the state-of-the-art methods for prostate segmentation such as ENet, UNet, USE-Net and TransUNet.

The Journal of Urology ® Home Study Course 2023 Volume 209/210

<i>The Journal of Urology</i> ® Home Study Course 2023 Volume 209/210

A dual attention-guided 3D convolution network for automatic segmentation of prostate and tumor

BackgroundIn middle-aged and older men, prostate cancer (PCa) is a common tumor disease with a mortality rate second only to lung cancer. The automatic and accurate segmentation of the prostate and tumor in magnetic resonance imaging (MRI) images can help doctors diagnose malignancies more efficiently. T2 weighted imaging (T2W) is now used in the majority of studies on prostate MRI image segmentation; however, diffusion-weighted imaging (DWI) is more valuable in the diagnosis of PCa. The morphological differences between the prostate and tumor regions are minimal, the tumor size is uncertain, the border between the tumor and surrounding tissue is hazy, and the categories separating normal regions from tumors are uneven. Consequently, it is challenging to segment prostate and tumor on DWI images. MethodsFor the segmentation of prostate and tumor regions on DWI images, this study offers a dual attention-guided 3D convolutional neural network (3D DAG-Net). A visual attention method is built into the encoder step to obtain the features of various receptive fields and deliver more detailed contextual information. A multiscale attention technique is proposed at the decoder stage to fuse multiscale features to acquire finer global and local details. To resolve the class discrepancies between the prostate, tumor, and background regions in segmentation tasks, we propose a hybrid loss function for handling class imbalance. ResultsWe tested the algorithm on DWI images of PCa obtained from a nearby hospital, demonstrating the uniqueness and effectiveness of the method. Dice similarity coefficient (DSC) values for prostate and tumor DWI segmentation were 92.28% and 88.73%, respectively. ConclusionWe present a unique dual-attention mechanism 3D segmentation network architecture for quantitative assessment of prostate and tumor volumes on DWI. The automatic segmentation results produced by our technology were highly correlated and consistent with expert manual segmentation findings.

Clinical utility of single-shot echo-planar diffusion-weighted imaging using L1-regularized iterative sensitivity encoding in prostate MRI.

We investigated the ability of echo-planar imaging with L1-regularized iterative sensitivity encoding-based diffusion-weighted imaging (DWI) to improve the image quality and reduce the scanning time in prostate magnetic resonance imaging. We retrospectively analyzed 109 cases of prostate magnetic resonance imaging. We compared variables in the quantitative and qualitative assessments among 3 imaging groups: conventional parallel imaging-based DWI (PI-DWI) with an acquisition time of 3 minutes 15 seconds; echo-planar imaging with L1-regularized iterative sensitivity encoding-based DWI (L1-DWI) with a normal acquisition time (L1-DWINEX12) of 3 minutes 15 seconds; and L1-DWI with a half acquisition time (L1-DWINEX6) of 1 minute 45 seconds. As a quantitative assessment, the signal-to-noise ratio (SNR) of DWI (SNR-DWI), the contrast-to-noise ratio (CNR) of DWI (CNR-DWI), and the CNR of apparent diffusion coefficient were measured. As a qualitative assessment, the image quality and visual detectability of prostate carcinoma were evaluated. In the quantitative analysis, L1-DWINEX12 showed significantly higher SNR-DWI than PI-DWI (P = .0058) and L1-DWINEX6 (P < .0001). In the qualitative analysis, the image quality score for L1-DWINEX12 was significantly higher than those of PI-DWI and L1-DWINEX6. A non-inferiority assessment demonstrated that L1-DWINEX6 was non-inferior to PI-DWI in terms of both quantitative CNR-DWI and qualitative grading of image quality with a <20% inferior margin. L1-DWI successfully demonstrated a reduced scanning time while maintaining good image quality.

Prostate imaging-reporting and data system version 2 has improved biopsy tumor grade accuracy: a single, tertiary institutional experience.

To investigate change in prostate biopsy accuracy regarding tumor grade before and after the release of Prostate Imaging-Reporting and Data System version 2 (PI-RADSv2) in a single tertiary institution. We retrospectively evaluated 1191 patients with biopsy-proven prostate cancer (PCa) who had undergone prostate magnetic resonance imaging (MRI) and surgery before (2013 cohort, n = 394) and 5 years after PI-RADSv2 release (2020 cohort, n = 797). The highest tumor grade of each biopsy and surgical specimen was recorded, respectively. We compared concordant, underestimated, and overestimated biopsy rates regarding tumor grade to surgery between two cohorts, respectively. For patients who underwent both prostate MRI and biopsy at our institution, we investigated proportion of pre-biopsy MRI, age, and prostate-specific antigen of patients, and performed logistic regression to analyze which parameters are associated with concordant biopsy. Concordant and underestimated biopsy rates were significantly different between two cohorts: Concordance and underestimation rates were 47.2% and 46.3% in 2013 and 54.5% and 36.4% in 2020 (p = .019; p = .003), respectively. Overestimated biopsy rates were similar (p = .993). Proportion of pre-biopsy MRI was significantly higher in 2020 than in 2013 (80.9% versus 4.9%; p < .001), and was independently associated with concordant biopsy results in multivariate analysis (odds ratio = 1.486; 95% confidence interval, 1.057-2.089; p = .022). There was a significant change in proportion of pre-biopsy MRI before and after the release of PI-RADSv2 in patients who underwent surgery for PCa. This change appears to have improved biopsy accuracy regarding tumor grade by reducing underestimation.

Segmentation of prostate region in magnetic resonance images based on improved V-Net

Magnetic resonance (MR) imaging is an important tool for prostate cancer diagnosis, and accurate segmentation of MR prostate regions by computer-aided diagnostic techniques is important for the diagnosis of prostate cancer. In this paper, we propose an improved end-to-end three-dimensional image segmentation network using a deep learning approach to the traditional V-Net network (V-Net) network in order to provide more accurate image segmentation results. Firstly, we fused the soft attention mechanism into the traditional V-Net's jump connection, and combined short jump connection and small convolutional kernel to further improve the network segmentation accuracy. Then the prostate region was segmented using the Prostate MR Image Segmentation 2012 (PROMISE 12) challenge dataset, and the model was evaluated using the dice similarity coefficient (DSC) and Hausdorff distance (HD). The DSC and HD values of the segmented model could reach 0.903 and 3.912 mm, respectively. The experimental results show that the algorithm in this paper can provide more accurate three-dimensional segmentation results, which can accurately and efficiently segment prostate MR images and provide a reliable basis for clinical diagnosis and treatment.

Biochemically recurrent prostate cancer: rationalisation of the approach to imaging

To assess the utility of magnetic resonance imaging (MRI) in addition to the additive benefit of the conventional imaging techniques, computed tomography (CT) and nuclear medicine (NM) bone scintigraphy, for investigation of biochemical recurrence (BCR) post-prostatectomy where access to prostate specific membrane antigen (PSMA) positron-emission tomography (PET)-CT is challenging. Relevant imaging over a 5-year period was reviewed. Ethical approval was granted by the internal review board. All patients with suspected BCR, defined as a PSA ≥0.2 ng/ml on two separate occasions, underwent a retrospective imaging review. This was performed on PACS archive search database in a single centre using search terms "PSA" and "prostatectomy" in the three imaging methods; MRI, CT, and NM bone scintigraphy. All PSMA PET CT performed were recorded. One hundred and eighty-five patients were identified. Patients with an MRI pelvis that demonstrated distant metastases (i.e., pelvic bone metastases or lymph node involvement more cranial to the bifurcation of the common iliac arteries) were more likely to have a positive CT and/or NM bone scintigraphy. The Pearson correlation coefficient between the findings of M1 disease at MRI pelvis and the presence of distant metastases at CT thorax, abdomen, pelvis and NM bone scintigraphy was calculated at 0.81 (p<0.01) and 0.91 (p<0.01) respectively. An imaging strategy based on risk stratification and technique-specific selection criteria leads to more appropriate use of resources, and in turn, increases the yield of conventional imaging methods. MRI prostate findings can be used to predict the additive value of CT/NM bone scintigraphy allowing a more streamlined approach to their use.

PD10-01 ASSOCIATIONS BETWEEN PROSTATE MRI AND GENOMIC TESTING AND TREATMENT INTENSIFICATION AMONG PATIENTS WITH LOCALIZED PROSTATE CANCER

You have accessJournal of UrologyCME1 Apr 2023PD10-01 ASSOCIATIONS BETWEEN PROSTATE MRI AND GENOMIC TESTING AND TREATMENT INTENSIFICATION AMONG PATIENTS WITH LOCALIZED PROSTATE CANCER Michael Leapman, Rong Wang, Jessica Long, Preston Sprenkle, Xiaomei Ma, and Cary Gross Michael LeapmanMichael Leapman More articles by this author , Rong WangRong Wang More articles by this author , Jessica LongJessica Long More articles by this author , Preston SprenklePreston Sprenkle More articles by this author , Xiaomei MaXiaomei Ma More articles by this author , and Cary GrossCary Gross More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000003250.01AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: Prostate magnetic resonance imaging (MRI) and tissue-based gene expression (genomic) tests improve local staging and estimates of disease prognosis, however clinical management associated with their use in the real-world setting is not well understood. METHODS: We performed a retrospective cohort study of Medicare beneficiaries diagnosed with clinically localized prostate cancer in 2013-2017 and treated with prostate radiation or radical prostatectomy in the Surveillance Epidemiology and End-Results (SEER)-Medicare database. We identified prostate MRI and genomic tests using specific administrative claims codes. The primary study outcome was treatment intensification, defined as any use of androgen deprivation therapy (ADT) for prostate radiation or pelvic lymph node dissection (PLND) for radical prostatectomy, respectively. We assessed the associations between patient-level receipt of prostate MRI and genomic testing and treatment intensification via multivariable logistic regressions, adjusting for clinical and sociodemographic factors, and further stratifying the analyses by D’Amico clinical risk status. RESULTS: We identified 26,017 patients meeting inclusion criteria, including 2,822 (11%) with low, 17,651 (68%) with intermediate and 4,664 (18%) with high-risk prostate cancer. In the overall cohort, receipt of prostate MRI (odds ratio [OR], 1.76 95% confidence interval [CI] 1.65-1.88, p<0.001) was associated with increased odds of treatment intensification, while genomic testing was not associated (OR 0.86, 95% CI 0.73-1.01, p=0.07). Prostate MRI was associated with increased odds of intensified treatment in all risk strata, while genomic testing was associated with lower odds of intensified treatment in the high-risk subset only (OR 0.59, 95% CI 0.35-1.00, p=0.048). By specific treatments, prostate MRI was associated with increased odds of PLND (OR 2.54, 95% CI 2.24-2.87) and ADT (OR 1.37, 95% CI 1.26-1.48) across all risk strata, while genomic testing was associated with decreased odds of PLND (OR 0.67, 95% CI 0.54-0.83) and ADT (OR 0.55, 95% CI 0.41-0.76). This analysis is limited by the absence of MRI and genomic testing results. CONCLUSIONS: We identified distinct patterns of management associated with prostate MRI and genomic testing in localized prostate cancer. These findings suggest the need for additional information about the independent role of diagnostic technologies in decision-making. Source of Funding: William O. Seery Foundation, Patterson Trust Mentored Research AwardNIH/National Cancer Institute © 2023 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 209Issue Supplement 4April 2023Page: e326 Advertisement Copyright & Permissions© 2023 by American Urological Association Education and Research, Inc.MetricsAuthor Information Michael Leapman More articles by this author Rong Wang More articles by this author Jessica Long More articles by this author Preston Sprenkle More articles by this author Xiaomei Ma More articles by this author Cary Gross More articles by this author Expand All Advertisement PDF downloadLoading ...

Deep learning for automated contouring of neurovascular structures on magnetic resonance imaging for prostate cancer patients.

Manual contouring of neurovascular structures on prostate magnetic resonance imaging (MRI) is labor-intensive and prone to considerable interrater disagreement. Our aim is to contour neurovascular structures automatically on prostate MRI by deep learning (DL) to improve workflow and interrater agreement. Segmentation of neurovascular structures was performed on pre-treatment 3.0T MRI data of 131 prostate cancer patients (training [n=105] and testing [n=26]). The neurovascular structures include the penile bulb (PB), corpora cavernosa (CCs), internal pudendal arteries (IPAs), and neurovascular bundles (NVBs). Two DL networks, nnU-Net and DeepMedic, were trained for auto-contouring on prostate MRI and evaluated using volumetric Dice similarity coefficient (DSC), mean surface distances (MSD), Hausdorff distances, and surface DSC. Three radiation oncologists evaluated the DL-generated contours and performed corrections when necessary. Interrater agreement was assessed and the time required for manual correction was recorded. nnU-Net achieved a median DSC of 0.92 (IQR: 0.90-0.93) for the PB, 0.90 (IQR: 0.86-0.92) for the CCs, 0.79 (IQR: 0.77-0.83) for the IPAs, and 0.77 (IQR: 0.72-0.81) for the NVBs, which outperformed DeepMedic for each structure (p<0.03). nnU-Net showed a median MSD of 0.24mm for the IPAs and 0.71mm for the NVBs. The median interrater DSC ranged from 0.93 to 1.00, with the majority of cases (68.9%) requiring manual correction times under two minutes. DL enables reliable auto-contouring of neurovascular structures on pre-treatment MRI data, easing the clinical workflow in neurovascular-sparing MR-guided radiotherapy.