Fully Automatic initialization and segmentation of left and right ventricles for large-scale cardiac MRI using a deeply supervised network and 3D-ASM

BACKGROUND AND OBJECTIVE

The sheer volume of data generated by population imaging studies is unparalleled by current capabilities to extract objective and quantitative cardiac phenotypes; subjective and time-consuming manual image analysis remains the gold standard. Automated image analytics to compute quantitative imaging biomarkers of cardiac function are desperately needed. Data volumes and their variability pose a challenge to most state-of-the-art methods for endo- and epicardial contours, which lack robustness when applied to very large datasets. Our aim is to develop an analysis pipeline for the automatic quantification of cardiac function from cine magnetic resonance imaging data.

METHOD

This work adopt 4,638 cardiac MRI cases coming from UK Biobank with ground truth available for left and RV contours. A hybrid and robust algorithm is proposed to improve the accuracy of automatic left and right ventricle segmentation by harnessing the localization accuracy of deep learning and the morphological accuracy of 3D-ASM (three-dimensional active shape models). The contributions of this paper are three-fold. First, a fully automatic method is proposed for left and right ventricle initialization and cardiac MRI segmentation by taking full advantage of spatiotemporal constraint. Second, a deeply supervised network is introduced to train and segment the heart. Third, the 3D-ASM image search procedure is improved by combining image intensity models with convolutional neural network (CNN) derived distance maps improving endo- and epicardial edge localization.

RESULTS

The proposed architecture outperformed the state of the art for cardiac MRI segmentation from UK Biobank. The statistics of RV landmarks detection errors for Triscuspid valve and RV apex are 4.17 mm and 5.58 mm separately. The overlap metric, mean contour distance, Hausdorff distance and cardiac functional parameters are calculated for the LV (Left Ventricle) and RV (Right Ventricle) contour segmentation. Bland-Altman analysis for clinical parameters shows that the results from our automated image analysis pipelines are in good agreement with results from expert manual analysis.

CONCLUSIONS

Our hybrid scheme combines deep learning and statistical shape modeling for automatic segmentation of the LV/RV from cardiac MRI datasets is effective and robust and can compute cardiac functional indexes from population imaging.

Sediment source fingerprinting as an aid to large-scale landscape conservation and restoration: A review for the Mississippi River Basin

Reliable quantitative information on sediment sources to rivers is critical to mitigate contamination and target conservation and restoration actions. However, for large-scale river basins, determination of the relative importance of sediment sources is complicated by spatiotemporal variability in erosional processes and sediment sources, heterogeneity in sediment transport and deposition, and a paucity of sediment monitoring data. Sediment source fingerprinting is an increasingly adopted field-based technique that identifies the nature and relative source contribution of sediment transported in waterways. Notably, sediment source fingerprinting provides information that is independent of other field, modeling, or remotely sensed techniques. However, the diversity in sampling, analytical, and interpretive methods for sediment fingerprinting has been recognized as a problem in terms of developing standardized procedures for its application at the scale of large river basins. Accordingly, this review focuses on sediment source fingerprinting studies conducted within the Mississippi River Basin (MRB), summarizes unique information provided by sediment source fingerprinting that is distinct from traditional monitoring techniques, evaluates consistency and reliability of methodological approaches among MRB studies, and provides prospects for the use of sediment source fingerprinting as an aid to large-scale landscape conservation and restoration under current management frameworks. Most MRB studies reported credible fingerprinting results and found near-channel sources to be the dominant sediment sources in most cases, and yet a lack of standardization in procedural steps makes results difficult to compare. Findings from MRB studies demonstrated that sediment source fingerprinting is a highly valuable and reliable sediment source assessment approach to assist land and water resource management under current management frameworks, but efforts are needed to make this technique applicable in large-scale landscape conservation and restoration efforts. We summarize research needs and discuss sediment fingerprinting use for basin-scale management efforts with the aim of encouraging that this technique is robust and reliable as it moves forward.

Mass spectrometry-based metabolomics for an in-depth questioning of human health

Today, metabolomics is becoming an indispensable tool to get a more comprehensive analysis of complex living systems, providing insights on multiple aspects of physiology. Although its application in large scale population-based studies is very challenging due to the processing of large sample sets as well as the complexity of data information, its potential to characterize human health is well recognized. Technological advances in metabolomics pave the way for the efficient biomarker discovery of disease etiology, diagnosis and prognosis. Here, different steps of the metabolomics workflow, particularly mass spectrometry-based approaches, are discussed to demonstrate the potential of metabolomics to address biological questioning in human health. First an overview of metabolomics is provided with its interest in human health studies. Analytical development and advances in mass spectrometry instrumentation and computational tools are discussed regarding their application limits. Advancing metabolomics for applicability in human health and large-scale studies is presented and discussed in conclusion.

CATI: A Large Distributed Infrastructure for the Neuroimaging of Cohorts

This paper provides an overview of CATI, a platform dedicated to multicenter neuroimaging. Initiated by the French Alzheimer's plan (2008-2012), CATI is a research project called on to provide service to other projects like an industrial partner. Its core mission is to support the neuroimaging of large populations, providing concrete solutions to the increasing complexity involved in such projects by bringing together a service infrastructure, the know-how of its expert academic teams and a large-scale, harmonized network of imaging facilities. CATI aims to make data sharing across studies easier and promotes sharing as much as possible. In the last 4 years, CATI has assisted the clinical community by taking charge of 35 projects so far and has emerged as a recognized actor at the national and international levels.

Automatic initialization and quality control of large-scale cardiac MRI segmentations

Continuous advances in imaging technologies enable ever more comprehensive phenotyping of human anatomy and physiology. Concomitant reduction of imaging costs has resulted in widespread use of imaging in large clinical trials and population imaging studies. Magnetic Resonance Imaging (MRI), in particular, offers one-stop-shop multidimensional biomarkers of cardiovascular physiology and pathology. A wide range of analysis methods offer sophisticated cardiac image assessment and quantification for clinical and research studies. However, most methods have only been evaluated on relatively small databases often not accessible for open and fair benchmarking. Consequently, published performance indices are not directly comparable across studies and their translation and scalability to large clinical trials or population imaging cohorts is uncertain. Most existing techniques still rely on considerable manual intervention for the initialization and quality control of the segmentation process, becoming prohibitive when dealing with thousands of images. The contributions of this paper are three-fold. First, we propose a fully automatic method for initializing cardiac MRI segmentation, by using image features and random forests regression to predict an initial position of the heart and key anatomical landmarks in an MRI volume. In processing a full imaging database, the technique predicts the optimal corrective displacements and positions in relation to the initial rough intersections of the long and short axis images. Second, we introduce for the first time a quality control measure capable of identifying incorrect cardiac segmentations with no visual assessment. The method uses statistical, pattern and fractal descriptors in a random forest classifier to detect failures to be corrected or removed from subsequent statistical analysis. Finally, we validate these new techniques within a full pipeline for cardiac segmentation applicable to large-scale cardiac MRI databases. The results obtained based on over 1200 cases from the Cardiac Atlas Project show the promise of fully automatic initialization and quality control for population studies.