Search

Objectives The aim of this study was to investigate 3 Tesla multiparametric magnetic resonance imaging (mpMRI)-based predictors for the pretherapeutic T staging of prostate cancer and their accuracy. Methods Consecutive patients with 3 Tesla mpMRI, positive systematic and MR-targeted biopsy, and subsequent radical prostatectomy (RPE) between 01/2016 and 12/2017 were included. MRI parameters such as measurable extraprostatic extension (EPE) (≥ 3 mm), length of (pseudo)capsular contact (LCC), invasion of neurovascular bundle (NVBI), and/or seminal vesicles lesion contact (SVC) or infiltration (SVI) were assessed and correlated to clinical and histopathological results. Results 136 men were included. In 76 cases, a pT2 stage was determined, in 29 cases a pT3a, and in 31 a pT3b stage. The positive and negative predictive values (PPV, NPV) for the detection of T3 by measurable EPE on MRI was 98% (CI 0.88–1) and 81% (CI 0.72–0.87). No visible NVBI was found in pT2 patients (NPV 100%; CI 0.95–1). ROC analysis for T3a prediction with LCC (AUC 0.81) showed a sensitivity of 87% and a specificity of 62% at a threshold of 12.5 mm (J = 0.485) and 93% and 58% at 11 mm (Jmax = 0.512). All patients with pT3a had a LCC > 5 mm. In case of pT3b, 29/31 patients showed a SVC (PPV 76%, CI 0.61–0.87; NPV 98%, CI 0.93–0.99), and 23/31 patients showed a SVI (PPV 100%, CI 0.86–1; NPV 93%, CI 0.87–0.96). EPE (p < 0.01), LCC (p = 0.05), and SVC (p = 0.01) were independent predictors of pT3. Conclusions MRI-measurable EPE, LCC, and SVC were reliable, independent, preoperative predictors for a histopathological T3 stage. A LCC ≥ 11 mm indicated a pT3a stage, whereas a LCC < 5 mm excluded it. On MRI, visible SVI or even SVC of the PCa lesion was reliable preoperative predictors for a pT3b stage.

Purpose Analysis of patients with pre-operative 3 T multiparametric prostate MRI (mpMRI) to determine reliable MRI-based risk predictors of patients at risk for positive surgical margins (PSM) in robotic assisted radical prostatectomy (RPE). Methods Consecutive patients with 3 T mpMRI and subsequent RPE from 01/2015 to 12/2018 were retrospectively included. Patients were compared regarding clinical and MRI related parameters such as length of capsular tumor contact (LCC) and distance to the membranous urethra (UD). Results Forty-nine of 179 patients (27%) had PSM in 70 different localizations, with the majority located at the capsule (57%, 40/70), mostly apical and/or posterior. The second most often PSM occurred at the apical urethra (22%, 15/70). PCA was visible on mpMRI at the localization of PSM in 93% at the capsule and in 80% at the urethra. PSA, PI-RADS classification, extraprostatic extension (EPE), and seminal vesicles infiltration (SVI) on MRI were significantly higher / more frequent in patients with PSM. LCC (AUC 0.710), EPE (AUC 0.693), and UD (1-AUC 0.673) predicted PSM (overall). An UD of ≤ 3.5 mm showed the highest accuracy of 95% (J = 0.946) for PSM at the urethra and a LCC of ≥ 22.5 mm with 77% (J = 0.378) for PSM at the capsule. Conclusion PSM occurred mostly in the apex and/or posteriorly at the capsule or at the apical urethra. LCC was the best MRI predictor for PSM at the capsule and UD for tumors with PSM at the apical urethra. Using these MRI parameters readers might pre-operatively determine PCA localizations at risk for PSM.

This study investigates preoperative lymph node metastases (LNM) detection accuracy by MRI and CT in patients with prostate cancer (PCA).All patients with preoperative MRI, CT or both and subsequent radical prostatectomy (RPE) and lymphadenectomy (LA) were included in this retrospective cohort study. Prostate specific antigen (PSA), PI-RADS, ISUP grade group, clinical and pathological tumor (T) stage was compared between negative and positive nodal (N) stage. LNM were assessed with size and localization and weather they were preoperatively detected or not. In patients with preoperative CT and MRI, the results were compared intermodally. The reference standard was the histopathological results after RPE and LA.A total of 228 patients were analysed including 24 patients with confirmed LNM (N1; 11%). PSA (median 9.7 vs. 14 ng/ml), PI-RADS (median 4 vs. 5), ISUP (median 2 vs. 4), and cT/pT-stage (median T2 vs. T3) was significantly higher in patients with LNM. No LNM were found in patients with ISUP-1-PCA. MRI was able to detect 67% of patients with LNM. Lymph node metastases presented on MRI predominantly small, round-shaped, located parailiacally with a minimum SAD of 4 mm (vs. CT SAD of 8 mm). In comparison, MRI was superior to CT in the detection of LNM (sensitivity 81% vs. 33%; specificity 99% vs. 97).LNM were very rare in patients with PSA 10 ng/ml, PI-RADS ≤ 4, and ≤ cT2. MRI could detect LNM up to 4 mm with a moderate sensitivity and high specificity. Thus, MRI might optimise the preoperative diagnostic and therapy planning of patients with PCA, whereas CT was clearly limited for N-stage assessment.

Supplemental Digital Content is available in the text., Background: Artificial intelligence (AI) for echocardiography requires training and validation to standards expected of humans. We developed an online platform and established the Unity Collaborative to build a dataset of expertise from 17 hospitals for training, validation, and standardization of such techniques. Methods: The training dataset consisted of 2056 individual frames drawn at random from 1265 parasternal long-axis video-loops of patients undergoing clinical echocardiography in 2015 to 2016. Nine experts labeled these images using our online platform. From this, we trained a convolutional neural network to identify keypoints. Subsequently, 13 experts labeled a validation dataset of the end-systolic and end-diastolic frame from 100 new video-loops, twice each. The 26-opinion consensus was used as the reference standard. The primary outcome was precision SD, the SD of the differences between AI measurement and expert consensus. Results: In the validation dataset, the AI’s precision SD for left ventricular internal dimension was 3.5 mm. For context, precision SD of individual expert measurements against the expert consensus was 4.4 mm. Intraclass correlation coefficient between AI and expert consensus was 0.926 (95% CI, 0.904–0.944), compared with 0.817 (0.778–0.954) between individual experts and expert consensus. For interventricular septum thickness, precision SD was 1.8 mm for AI (intraclass correlation coefficient, 0.809; 0.729–0.967), versus 2.0 mm for individuals (intraclass correlation coefficient, 0.641; 0.568–0.716). For posterior wall thickness, precision SD was 1.4 mm for AI (intraclass correlation coefficient, 0.535 [95% CI, 0.379–0.661]), versus 2.2 mm for individuals (0.366 [0.288–0.462]). We present all images and annotations. This highlights challenging cases, including poor image quality and tapered ventricles. Conclusions: Experts at multiple institutions successfully cooperated to build a collaborative AI. This performed as well as individual experts. Future echocardiographic AI research should use a consensus of experts as a reference. Our collaborative welcomes new partners who share our commitment to publish all methods, code, annotations, and results openly.

The paper presents some results concerning the secular variation of H, Z and F in Romania between 1958–1974. Curves of the variation of , Z and F at 31 repeat stations are given. The effect of solar activity on the horizontal component is separated and its characteristics discussed. For the interval 1969.5–1974.5 maps of the smoothed secular variation of H, Z and F are given, with isopores described by curves of second degree with respect to the geographical coordinates.