Despite learning based methods showing promising results in single view depth estimation and visual odometry, most existing approaches treat the tasks in a supervised manner. Recent approaches to single view depth estimation explore the possibility of learning without full supervision via minimizing photometric error. In this paper, we explore the use of stereo sequences for learning depth and visual odometry. The use of stereo sequences enables the use of both spatial (between left-right pairs) and temporal (forward backward) photometric warp error, and constrains the scene depth and camera motion to be in a common, realworld scale. At test time our framework is able to estimate single view depth and two-view odometry from a monocular sequence. We also show how we can improve on a standard photometric warp loss by considering a warp of deep features. We show through extensive experiments that: (i) jointly training for single view depth and visual odometry improves depth prediction because of the additional constraint imposed on depths and achieves competitive results for visual odometry; (ii) deep feature-based warping loss improves upon simple photometric warp loss for both single view depth estimation and visual odometry. Our method outperforms existing learning based methods on the KITTI driving dataset in both tasks. The source code is available at https://github.com/Huangying-Zhan/ Depth-VO-Feat.

Purpose:To determine whether multiparametric magnetic resonance (MR) imaging can help identify patients with prostate cancer who would most appropriately be candidates for active surveillance (AS) according to current guidelines and to compare the results with those of conventional clinical assessment scoring systems, including the D'Amico, Epstein, and Cancer of the Prostate Risk Assessment (CAPRA) systems, on the basis of findings at prostatectomy. Materials and Methods:This institutional review board-approved HIPAA-compliant retrospectively designed study included 133 patients (mean age, 59.3 years) with a mean prostate-specific antigen level of 6.73 ng/mL (median, 4.39 ng/mL) who underwent multiparametric MR imaging at 3.0 T before radical prostatectomy. Informed consent was obtained from all patients. Patients were then retrospectively classified as to whether they would have met AS eligibility criteria or were better served by surgery. AS eligibility criteria for prostatectomy specimens were a dominant tumor smaller than 0.5 mL without Gleason 4 or 5 patterns or extracapsular or seminal vesicle invasion. Conventional clinical assessment scores (the D'Amico, Epstein, and CAPRA scoring systems) were compared with multiparametric MR imaging findings for predicting AS candidates. The level of significance of difference between scoring systems was determined by using the x 2 test for categoric variables with the level of significance set at P , .05. Results:Among 133 patients, 14 were eligible for AS on the basis of prostatectomy results. The sensitivity, positive predictive value (PPV), and overall accuracy, respectively, were 93%, 25%, and 70% for the D'Amico system, 64%, 45%, and 88% for the Epstein criteria, and 93%, 20%, and 59% for the CAPRA scoring system for predicting AS candidates (P , .005 for all, x 2 test), while multiparametric MR imaging had a sensitivity of 93%, a PPV of 57%, and an overall accuracy of 92% (P , .005). Conclusion:Multiparametric MR imaging provides useful additional information to existing clinicopathologic scoring systems of prostate cancer and improves the assignment of treatment (eg, AS or active treatment).q RSNA, 2013 Supplemental material: http://radiology.rsna.org/lookup /suppl

Background Coronary endothelial function (endoFx) is abnormal in patients with established coronary artery disease (CAD) and was recently shown by MRI to relate to the severity of luminal stenosis. Recent advances in MRI now allow the non-invasive assessment of both anatomic and functional (endoFx) changes that previously required invasive studies. We tested the hypothesis that abnormal coronary endoFx is related to measures of early atherosclerosis such as increased coronary wall thickness (CWT). Methods and Results Seventeen arteries in fourteen healthy adults and seventeen arteries in fourteen patients with non-obstructive CAD were studied. To measure endoFx, coronary MRI was performed before and during isometric handgrip exercise, an endothelial-dependent stressor and changes in coronary cross-sectional area (CSA) and flow were measured. Black blood imaging was performed to quantify CWT and other indices of arterial remodeling. The mean stress-induced change in CSA was significantly higher in healthy adults (13.5%±12.8%, mean±SD, n=17) than in those with mildly diseased arteries (-2.2±6.8%, p<0.0001, n=17). Mean CWT was lower in healthy subjects (0.9±0.2mm) than in CAD patients (1.4±0.3mm, p<0.0001). In contrast to healthy subjects, stress-induced changes in CSA, a measure of coronary endoFx, correlated inversely with CWT in CAD patients (r= -0.73, p=0.0008). Conclusions There is an inverse relationship between coronary endothelial function and local CWT in CAD patients but not in healthy adults. These findings demonstrate that local endothelial-dependent functional changes are related to the extent of early anatomic atherosclerosis in mildly diseased arteries. This combined MRI approach enables the anatomic and functional investigation of early coronary disease.

BackgroundWe sought to investigate the relationship between infarct and dyssynchrony post- myocardial infarct (MI), in a porcine model. Mechanical dyssynchrony post-MI is associated with left ventricular (LV) remodeling and increased mortality.MethodsCine, gadolinium-contrast, and tagged cardiovascular magnetic resonance (CMR) were performed pre-MI, 9 ± 2 days (early post-MI), and 33 ± 10 days (late post-MI) post-MI in 6 pigs to characterize cardiac morphology, location and extent of MI, and regional mechanics. LV mechanics were assessed by circumferential strain (eC). Electro-anatomic mapping (EAM) was performed within 24 hrs of CMR and prior to sacrifice.ResultsMean infarct size was 21 ± 4% of LV volume with evidence of post-MI remodeling. Global eC significantly decreased post MI (-27 ± 1.6% vs. -18 ± 2.5% (early) and -17 ± 2.7% (late), p < 0.0001) with no significant change in peri-MI and MI segments between early and late time-points. Time to peak strain (TTP) was significantly longer in MI, compared to normal and peri-MI segments, both early (440 ± 40 ms vs. 329 ± 40 ms and 332 ± 36 ms, respectively; p = 0.0002) and late post-MI (442 ± 63 ms vs. 321 ± 40 ms and 355 ± 61 ms, respectively; p = 0.012). The standard deviation of TTP in 16 segments (SD16) significantly increased post-MI: 28 ± 7 ms to 50 ± 10 ms (early, p = 0.012) to 54 ± 19 ms (late, p = 0.004), with no change between early and late post-MI time-points (p = 0.56). TTP was not related to reduction of segmental contractility. EAM revealed late electrical activation and greatly diminished conduction velocity in the infarct (5.7 ± 2.4 cm/s), when compared to peri-infarct (18.7 ± 10.3 cm/s) and remote myocardium (39 ± 20.5 cm/s).ConclusionsMechanical dyssynchrony occurs early after MI and is the result of delayed electrical and mechanical activation in the infarct.