Ibsenviborg2039

Quantitative Susceptibility Mapping (QSM) estimates tissue magnetic susceptibility distributions from Magnetic Resonance (MR) phase measurements by solving an ill-posed dipole inversion problem. Conventional single orientation QSM methods usually employ regularization strategies to stabilize such inversion, but may suffer from streaking artifacts or over-smoothing. Multiple orientation QSM such as calculation of susceptibility through multiple orientation sampling (COSMOS) can give well-conditioned inversion and an artifact free solution but has expensive acquisition costs. On the other hand, Convolutional Neural Networks (CNN) show great potential for medical image reconstruction, albeit often with limited interpretability. Here, we present a Learned Proximal Convolutional Neural Network (LP-CNN) for solving the ill-posed QSM dipole inversion problem in an iterative proximal gradient descent fashion. This approach combines the strengths of data-driven restoration priors and the clear interpretability of iterative solvers that can take into account the physical model of dipole convolution. During training, our LP-CNN learns an implicit regularizer via its proximal, enabling the decoupling between the forward operator and the data-driven parameters in the reconstruction algorithm. More importantly, this framework is believed to be the first deep learning QSM approach that can naturally handle an arbitrary number of phase input measurements without the need for any ad-hoc rotation or re-training. We demonstrate that the LP-CNN provides state-of-the-art reconstruction results compared to both traditional and deep learning methods while allowing for more flexibility in the reconstruction process.Modern machine learning systems, such as convolutional neural networks rely on a rich collection of training data to learn discriminative representations. In many medical imaging applications, unfortunately, collecting a large set of well-annotated data is prohibitively expensive. To overcome data shortage and facilitate representation learning, we develop Knowledge-guided Pretext Learning (KPL) that learns anatomy-related image representations in a pretext task under the guidance of knowledge from the downstream target task. In the context of utero-placental interface detection in placental ultrasound, we find that KPL substantially improves the quality of the learned representations without consuming data from external sources such as IMAGENET. It outperforms the widely adopted supervised pre-training and self-supervised learning approaches across model capacities and dataset scales. Our results suggest that pretext learning is a promising direction for representation learning in medical image analysis, especially in the small data regime.We examine how combinations of systolic and diastolic blood pressure levels and pulse pressure levels predicted mortality risk. Respondents are those aged over 50 from the Health and Retirement Study (N=10,366) who provided blood pressure measures in 2006/2008. Systolic and diastolic blood pressures were measured three times; and we averaged the three readings. Pulse pressure was calculated as systolic minus diastolic blood pressure. Phenazine methosulfate clinical trial Seven combinations of systolic and diastolic blood pressure (low/normal/high of each) and three levels of pulse pressure (low/normal/high) were used to categorize blood pressure. Over 1 to 10 years of follow-up (average follow-up time of 7.8 years), 2,820 respondents died after blood pressure measurement in 2006/2008. Potential covariates including age, gender, education, BMI, total cholesterol, HbA1c, antihypertensive medication intake and lifetime-smoking pack years were adjusted in Cox proportional hazard models and survival curves. The blood pressure subgroup with low systolic blood pressure ( less then 90 mmHg) and low diastolic blood pressure ( less then 60 mmHg) had the highest relative risk of mortality (HR=2.34, 95% CI 1.45-3.80), followed by those with normal systolic blood pressure but low diastolic blood pressure (HR=1.45, 95% CI 1.17-1.81) among those with cardiovascular conditions at baseline. For those without cardiovascular conditions at baseline, low blood pressure, either systolic or diastolic, was not related to mortality. Those with high levels of both systolic and diastolic blood pressure had a higher risk of mortality than those with both blood pressures normal but no other subgroups with low blood pressure differed from normal/normal in predicting mortality. Pulse pressure did not predict mortality. How high and low blood pressures are related to mortality needs to be examined by jointly looking at systolic and diastolic blood pressure.Spectral CT has great potential for a variety of clinical applications due to the improved material discrimination with respect to conventional CT. Many clinical and preclinical spectral CT systems have two spectral channels for dual-energy CT using strategies such as split-filtration, dual-layer detectors, or kVp-switching. However, there are emerging clinical imaging applications which would require three or more spectral sensitivity channels, for example, multiple exogenous contrast agents in a single scan. Spatial-spectral filters are a new spectral CT technology which use x-ray beam modulation to offer greater spectral diversity. The device consists of an array of k-edge filters which divide the x-ray beam into spectrally varied beamlets. This design allows for an arbitrary number of spectral channels; however, traditional two-step reconstruction-decomposition schemes are typically not effective because the measured data for any individual spectral channel is sparse in the projection domain. Instead, we ons of spectral CT.Interest in spectral CT for diagnostics and therapy evaluation has been growing. Acquisitions of data from distinct energy spectra provide, among other advantages, quantitative density estimations for multiple materials. We introduce a novel spectral CT concept that includes a fine-pitch grating for prefiltration of the x-ray beam. The attenuation behavior of this grating changes significantly if x-rays are slightly angled in relation to the grating structures. To apply such an angle (i.e. switch between the different filtrations) we propose a fast, controllable, and precise solution by moving the focal spot of the x-ray tube. In this work, we performed preliminary evaluations with a grating prototype on a CT test bench. Our results include x-ray spectrometer measurements that reveal diverse and controllable spectral shaping between 4° and 6° for a specific grating design. Additional experiments with a contrast agent phantom illustrated the capability to decompose clinically relevant iodine concentrations (5, 10, 20, and 50mg/mL) - demonstrating the feasibility of the grating-based approach.