With a massive influx of multimodality data, the role of data analytics in health informatics has grown rapidly in the last decade. This has also prompted increasing interests in the generation of analytical, data driven models based on machine learning in health informatics. Deep learning, a technique with its foundation in artificial neural networks, is emerging in recent years as a powerful tool for machine learning, promising to reshape the future of artificial intelligence. Rapid improvements in computational power, fast data storage, and parallelization have also contributed to the rapid uptake of the technology in addition to its predictive power and ability to generate automatically optimized high-level features and semantic interpretation from the input data. This article presents a comprehensive up-to-date review of research employing deep learning in health informatics, providing a critical analysis of the relative merit, and potential pitfalls of the technique as well as its future outlook. The paper mainly focuses on key applications of deep learning in the fields of translational bioinformatics, medical imaging, pervasive sensing, medical informatics, and public health.

/pdf/deep-learning-for-health-informatics-3k8fzpjo4y.pdf

Deep Learning for Health Informatics.

3D deeply supervised network for automated segmentation of volumetric medical images.

The International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) is a scientific organization that encourages safe clinical practice and high-quality teaching and research related to diagnostic imaging in women’s healthcare. The ISUOG Clinical Standards Committee (CSC) has a remit to develop Practice Guidelines and Consensus Statements that provide healthcare practitioners with a consensus-based approach for diagnostic imaging. They are intended to reflect what is considered by ISUOG to be the best practice at the time at which they are issued. Although ISUOG has made every effort to ensure that Guidelines are accurate when issued, neither the Society nor any of its employees or members accept any liability for the consequences of any inaccurate or misleading data, opinions or statements issued by the CSC. The ISUOG CSC documents are not intended to establish a legal standard of care because interpretation of the evidence that underpins the Guidelines may be influenced by individual circumstances, local protocol and available resources. Approved Guidelines can be distributed freely with the permission of ISUOG (info@isuog.org).

ISUOG practice guidelines: performance of first-trimester fetal ultrasound scan.

Cardiovascular magnetic resonance (CMR) has become a key imaging modality in clinical cardiology practice due to its unique capabilities for non-invasive imaging of the cardiac chambers and great vessels. A wide range of CMR sequences have been developed to assess various aspects of cardiac structure and function, and significant advances have also been made in terms of imaging quality and acquisition times. A lot of research has been dedicated to the development of global and regional quantitative CMR indices that help the distinction between health and pathology. The goal of this review paper is to discuss the structural and functional CMR indices that have been proposed thus far for clinical assessment of the cardiac chambers. We include indices definitions, the requirements for the calculations, exemplar applications in cardiovascular diseases, and the corresponding normal ranges. Furthermore, we review the most recent state-of-the art techniques for the automatic segmentation of the cardiac boundaries, which are necessary for the calculation of the CMR indices. Finally, we provide a detailed discussion of the existing literature and of the future challenges that need to be addressed to enable a more robust and comprehensive assessment of the cardiac chambers in clinical practice.

/pdf/a-review-of-heart-chamber-segmentation-for-structural-and-2mlis4rr77.pdf

A review of heart chamber segmentation for structural and functional analysis using cardiac magnetic resonance imaging

The present and future of deep learning in radiology.

http://www.digitalimaginggroup.ca/members/Shuo/1-s2.0-S1361841515001024-main.pdf

Multi-scale deep networks and regression forests for direct bi-ventricular volume estimation

Cardiac indices estimation is of great importance during identification and diagnosis of cardiac disease in clinical routine. However, estimation of multitype cardiac indices with consistently reliable and high accuracy is still a great challenge due to the high variability of cardiac structures and the complexity of temporal dynamics in cardiac MR sequences. While efforts have been devoted into cardiac volumes estimation through feature engineering followed by a independent regression model, these methods suffer from the vulnerable feature representation and incompatible regression model. In this paper, we propose a semi-automated method for multitype cardiac indices estimation. After the manual labeling of two landmarks for ROI cropping, an integrated deep neural network Indices-Net is designed to jointly learn the representation and regression models. It comprises two tightly-coupled networks, such as a deep convolution autoencoder for cardiac image representation, and a multiple output convolution neural network for indices regression. Joint learning of the two networks effectively enhances the expressiveness of image representation with respect to cardiac indices, and the compatibility between image representation and indices regression, thus leading to accurate and reliable estimations for all the cardiac indices. When applied with five-fold cross validation on MR images of 145 subjects, Indices-Net achieves consistently low estimation error for LV wall thicknesses (1.44 ± 0.71 mm) and areas of cavity and myocardium (204 ± 133 mm2). It outperforms, with significant error reductions, segmentation method (55.1% and 17.4%), and two-phase direct volume-only methods (12.7% and 14.6%) for wall thicknesses and areas, respectively. These advantages endow the proposed method a great potential in clinical cardiac function assessment.

Direct Multitype Cardiac Indices Estimation via Joint Representation and Regression Learning

Direct and simultaneous estimation of cardiac four chamber volumes by multioutput sparse regression

Multioutput regression has recently shown great ability to solve challenging problems in both computer vision and medical image analysis. However, due to the huge image variability and ambiguity, it is fundamentally challenging to handle the highly complex input-target relationship of multioutput regression, especially with indiscriminate high-dimensional representations. In this paper, we propose a novel supervised descriptor learning (SDL) algorithm for multioutput regression, which can establish discriminative and compact feature representations to improve the multivariate estimation performance. The SDL is formulated as generalized low-rank approximations of matrices with a supervised manifold regularization. The SDL is able to simultaneously extract discriminative features closely related to multivariate targets and remove irrelevant and redundant information by transforming raw features into a new low-dimensional space aligned to targets. The achieved discriminative while compact descriptor largely reduces the variability and ambiguity for multioutput regression, which enables more accurate and efficient multivariate estimation. We conduct extensive evaluation of the proposed SDL on both synthetic data and real-world multioutput regression tasks for both computer vision and medical image analysis. Experimental results have shown that the proposed SDL can achieve high multivariate estimation accuracy on all tasks and largely outperforms the algorithms in the state of the arts. Our method establishes a novel SDL framework for multioutput regression, which can be widely used to boost the performance in different applications.

Descriptor Learning via Supervised Manifold Regularization for Multioutput Regression

Objective

To determine the visualization rates of fetal anatomic structures by three-dimensional ultrasound (3DUS) at 12–13 weeks of gestation.



Study Design

This was a prospective observational study of women presenting for nuchal translucency ultrasound. Five 3D volumes of the fetus were acquired transabdominally. Two investigators independently reviewed the stored volumes offline following a standardized protocol.



Results

One hundred singleton fetuses were examined. The mean time for 3D volumes acquisition was 4.8 min; and for 3D review 17 min. Anatomic structures were seen as follows: cranium, lateral cerebral ventricles and abdominal wall 100%; stomach, vertebrae, upper and lower limbs ≥ 94%; face 71%, bladder 58%, both kidneys 39%, skin overlying spine 26% and heart 18%. Agreement between two observers ranged from 100% (for head, abdominal wall and lower limbs) to 43% (for visualization of skin overlying spine). A complete basic anatomic survey was achieved in 11.4% of the 12-week fetuses and 33.3% of the 13-week fetuses (p-value = 0.038).



Conclusions

First-trimester transabdominal 3DUS was adequate for assessment of the head, abdominal wall, stomach, limbs and vertebral alignment. It was less effective for evaluating the heart and intactness of the skin over the spine. Copyright © 2010 John Wiley & Sons, Ltd.

Mousumi Bhaduri

Papers

Multi-scale deep networks and regression forests for direct bi-ventricular volume estimation

Direct Multitype Cardiac Indices Estimation via Joint Representation and Regression Learning

Direct and simultaneous estimation of cardiac four chamber volumes by multioutput sparse regression

Descriptor Learning via Supervised Manifold Regularization for Multioutput Regression

Fetal anatomic survey using three-dimensional ultrasound in conjunction with first-trimester nuchal translucency screening