University of Pennsylvania

Other

The Children's Hospital of Philadelphia

Jayaram K Udupa

Drew A Torigian

Yubing Tong

Dewey Odhner

Tiange Liu

Andre Souza

Punam Kumar Saha

George Grevera

Jaimin Liu

L&aacute;szl&oacute; G Ny&uacute;l

Supun Samarasekera

You Hao

Changjian Sun

Joseph M McDonough

Patrick J Cahill

Vibhu Agrawal

Methods and systems are described for determining body composition information. An example method can comprise receiving imaging data associated with a patient, causing the imaging data to be input into a convolutional neural network stored on one or more computing devices, determining, based on output data resulting from inputting the imaging data into the convolutional neural network, body composition information, and causing output of the body composition information.

Deep learning network for the analysis of body tissue composition on body-torso-wide CT images

A method of analyzing thoracic insufficiency syndrome (TIS) in a subject by performing quantitative dynamic magnetic resonance imaging (QdMRI) analysis. The QdMRI analysis includes performing four-dimensional (4D) image construction of a TIS subject's thoracic cavity. The 4D image includes a sequence of two dimensional (2D) images of the TIS subject's thoracic cavity over a respiratory cycle of the TIS subject. The QdMRI analysis also includes segmenting a region of interest (ROI) within the 4D image, determining TIS measurements within the ROI, comparing the TIS measurements to normal measurements determined from ROIs in 4D images of the thoracic cavities of normal subjects that are not afflicted by TIS, and outputting quantitative markers indicating deviation of the thoracic cavity of the TIS subject relative to the thoracic cavities of the normal subjects.

Quantitative dynamic MRI (QDMRI) analysis and virtual growing child (VGC) systems and methods for treating respiratory anomalies

A computerized method of providing automatic anatomy recognition (AAR) includes gathering image data from patient image sets, formulating precise definitions of each body region and organ and delineating them following the definitions, building hierarchical fuzzy anatomy models of organs for each body region, recognizing and locating organs in given images by employing the hierarchical models, and delineating the organs following the hierarchy. The method may be applied, for example, to body regions including the thorax, abdomen and neck regions to identify organs.

Applications of automatic anatomy recognition in medical tomographic imagery based on fuzzy anatomy models

Provided are systems and methods for analyzing medical images to localize body regions using deep learning techniques. A combination of convolutional neural network (CNN) and a recurrent neural network (RNN) can be applied to medical images, identifying axial slices of a body region. In accordance with embodiments, boundaries, e.g., superior and inferior boundaries of various body regions in computed tomography images can be automatically demarcated.

Deep learning architecture for analyzing medical images for body region recognition and delineation

Quantification of body composition plays an important role in many clinical and research applications. Radiologic imaging techniques such as Dual-energy X-ray absorptiometry, magnetic resonance imaging (MRI), and computed tomography (CT) imaging make accurate quantification of the body composition possible. This disclosure presents an automated, efficient, accurate, and practical body composition quantification method for low dose CT images; method for quantification of disease from images; and methods for implementing virtual landmarks.

Quantification and staging of body-wide tissue composition and of abnormal states on medical images via automatic anatomy recognition

Interactive non-uniformity correction (NC) and interactive intensity standardization (IS) require sample tissue regions to be specified for several different types of tissues. Interactive NC estimates the degree of non-uniformity at each voxel in a given image, builds a global function for non-uniformity correction, and then corrects the image to improve quality. Interactive IS includes two steps: a calibration step and a transformation step. In the first step, tissue intensity signatures of each tissue from a few subjects are utilized to set up key landmarks in a standardized intensity space. In the second step, a piecewise linear intensity mapping function is built between the same tissue signatures derived from the given image and those in the standardized intensity space to transform the intensity of the given image into standardized intensity. Interactive IS for MR images combined with interactive NC can substantially improve numeric characterization of tissues.

Interactive non-uniformity correction and intensity standardization of MR images

Methods for the improved interactive segmentation of medical image slice data using a computer include the novel combination of the well-known live wire and snakes methods. The improved techniques automatically insert new anchor points for each medical image slice that is processed. The improved methods called iterative live wire and live snakes result in a segmentation process that is faster, more accurate, and requires less operator interaction than the previous methods while still allowing an operator to make adjustments to the segmentation as the process moves from one image slice to the next.

Image slice segmentation using midpoints of contour anchor points

An improved method of segmenting medical images includes aspects of live wire and active shape models to determine the most likely segmentation given a shape distribution that satisfies boundary location constrains on an item of interest. The method includes a supervised learning portion to train and learn new types of shape instances and a segmentation portion to use the learned model to segment new target images containing instances of the shape. The segmentation portion includes an automated search for an appropriate shape and deformation of the shape to establish a best oriented boundary for the object of interest on a medical image.

Segmentation method using an oriented active shape model

The invention provides novel scale-based filtering methods that use local structure size or 'object scale' information to arrest smoothing around fine structures and across even low-gradient boundaries. One method teaches a weighted average over a scale-dependent neighborhood; while another employs scale-dependent diffusion conductance to perform filtering. Both methods adaptively modify the degree of filtering at any image location depending on local object scale. This permits a restricted homogeneity parameter to be accurately used for filtering in regions with fine details and in the vicinity of boundaries, while at the same time, a generous filtering parameter is used in the interiors of homogeneous regions.

Scale-based image filtering of magnetic resonance data

The present invention provides a method for standardizing MR image intensity scales by the use of post-processing intensity transformation techniques applied to routinely acquired images. The method requires neither specialized acquisition protocols, nor calibration phantoms. The standardizing method offers previously unattainable consistency of intensity meaning of tissues by devising a transformation that is specific to a given MRI protocol and/or for any body region to provide standardized images. Essentially, the histogram of a given volume image is deformed to match a &ldquo;standard&rdquo; histogram for the corresponding MRI protocol and body-region, thereby minimizing or eliminating the human interaction required in the per-case manual window adjustments needed to visualize MR images at physician viewing stations. The method offers significantly more consistent tissue meaning for MR image intensities than the images before standardization. Also provided are MR images having standardized intensity scales.

Method for standardizing the MR image intensity scale

A technique for object information extraction from images which retains fuzziness as realistically as possible. The technique is used for image segmentation of fuzzy objects for n-dimensional digital spaces based on the notion of 'hanging togetherness' of image elements specified by their fuzzy connectedness. A specified fuzzy object is extracted and all fuzzy objects present in the image data are identified by segmenting the image using the notion of 'hanging togetherness' of image elements specified by their fuzzy connectedness as defined herein. The technique is used in a preferred embodiment to quantify MS lesions of the brain via magnetic resonance imaging.

Growing community of inventors

Philadelphia, PA, United States of America

Jayaram K Udupa