Dataset Information

Minimal Linear Networks for Magnetic Resonance Image Reconstruction.

ABSTRACT: Modern sequences for Magnetic Resonance Imaging (MRI) trade off scan time with computational challenges, resulting in ill-posed inverse problems and the requirement to account for more elaborated signal models. Various deep learning techniques have shown potential for image reconstruction from reduced data, outperforming compressed sensing, dictionary learning and other advanced techniques based on regularization, by characterization of the image manifold. In this work we suggest a framework for reducing a "neural" network to the bare minimum required by the MR physics, reducing the network depth and removing all non-linearities. The networks performed well both on benchmark simulated data and on arterial spin labeling perfusion imaging, showing clear images while preserving sensitivity to the minute signal changes. The results indicate that the deep learning framework plays a major role in MR image reconstruction, and suggest a concrete approach for probing into the contribution of additional elements.

SUBMITTER: Liberman G

PROVIDER: S-EPMC6925115 | biostudies-literature | 2019 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Minimal Linear Networks for Magnetic Resonance Image Reconstruction.

Liberman Gilad G Poser Benedikt A BA

Scientific reports 20191220 1

Modern sequences for Magnetic Resonance Imaging (MRI) trade off scan time with computational challenges, resulting in ill-posed inverse problems and the requirement to account for more elaborated signal models. Various deep learning techniques have shown potential for image reconstruction from reduced data, outperforming compressed sensing, dictionary learning and other advanced techniques based on regularization, by characterization of the image manifold. In this work we suggest a framework for ...[more]

PMID: 31862922

Similar Datasets

Project description:Cardiovascular resonance (CMR) imaging is a standard imaging modality for assessing cardiovascular diseases (CVDs), the leading cause of death globally. CMR enables accurate quantification of the cardiac chamber volume, ejection fraction and myocardial mass, providing information for diagnosis and monitoring of CVDs. However, for years, clinicians have been relying on manual approaches for CMR image analysis, which is time consuming and prone to subjective errors. It is a major clinical challenge to automatically derive quantitative and clinically relevant information from CMR images.Deep neural networks have shown a great potential in image pattern recognition and segmentation for a variety of tasks. Here we demonstrate an automated analysis method for CMR images, which is based on a fully convolutional network (FCN). The network is trained and evaluated on a large-scale dataset from the UK Biobank, consisting of 4,875 subjects with 93,500 pixelwise annotated images. The performance of the method has been evaluated using a number of technical metrics, including the Dice metric, mean contour distance and Hausdorff distance, as well as clinically relevant measures, including left ventricle (LV) end-diastolic volume (LVEDV) and end-systolic volume (LVESV), LV mass (LVM); right ventricle (RV) end-diastolic volume (RVEDV) and end-systolic volume (RVESV).By combining FCN with a large-scale annotated dataset, the proposed automated method achieves a high performance in segmenting the LV and RV on short-axis CMR images and the left atrium (LA) and right atrium (RA) on long-axis CMR images. On a short-axis image test set of 600 subjects, it achieves an average Dice metric of 0.94 for the LV cavity, 0.88 for the LV myocardium and 0.90 for the RV cavity. The mean absolute difference between automated measurement and manual measurement is 6.1 mL for LVEDV, 5.3 mL for LVESV, 6.9 gram for LVM, 8.5 mL for RVEDV and 7.2 mL for RVESV. On long-axis image test sets, the average Dice metric is 0.93 for the LA cavity (2-chamber view), 0.95 for the LA cavity (4-chamber view) and 0.96 for the RA cavity (4-chamber view). The performance is comparable to human inter-observer variability.We show that an automated method achieves a performance on par with human experts in analysing CMR images and deriving clinically relevant measures.

Project description:ObjectiveAlthough the general utility of voxel-based processing of structural magnetic resonance imaging (MRI) data for detecting occult lesions in focal epilepsy is established, many differences exist among studies, and it is unclear which processing method is preferable. The aim of this study was to compare the ability of commonly used methods to detect epileptogenic lesions in magnetic resonance MRI-positive and MRI-negative patients, and to estimate their diagnostic yield.MethodsWe identified 144 presurgical focal epilepsy patients, 15 of whom had a histopathologically proven and MRI-visible focal cortical dysplasia; 129 patients were MRI negative with a clinical hypothesis of seizure origin, 27 of whom had resections. We applied four types of voxel-based morphometry (VBM), three based on T1 images (gray matter volume, gray matter concentration, junction map [JM]) and one based on normalized fluid-attenuated inversion recovery (nFSI). Specificity was derived from analysis of 50 healthy controls.ResultsThe four maps had different sensitivity and specificity profiles. All maps showed detection rates for focal cortical dysplasia patients (MRI positive and negative) of >30% at a strict threshold of p < 0.05 (family-wise error) and >60% with a liberal threshold of p < 0.0001 (uncorrected), except for gray matter volume (14% and 27% detection rate). All maps except nFSI showed poor specificity, with high rates of false-positive findings in controls. In the MRI-negative patients, absolute detection rates were lower. A concordant nFSI finding had a significant positive odds ratio of 7.33 for a favorable postsurgical outcome in the MRI-negative group. Spatial colocalization of JM and nFSI was rare, yet showed good specificity throughout the thresholds.SignificanceAll VBM variants had specific diagnostic properties that need to be considered for an adequate interpretation of the results. Overall, structural postprocessing can be a useful tool in presurgical diagnostics, but the low specificity of some maps has to be taken into consideration.

Dataset Information

Minimal Linear Networks for Magnetic Resonance Image Reconstruction.

Publications

Minimal Linear Networks for Magnetic Resonance Image Reconstruction.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets