Vida Diagnostics, Inc.

Neomagic Corporation

Change Healthcare Holdings, LLC

Mevis Medical Solutions, Inc.

Hologic, Inc.

Philippe Raffy

Youbing Yin

Fathy Yassa

Susan A Wood

Dan Schonfeld

Karthik Hariharakrishnan

John D Newell, Jr

Jean-Francois Pambrun

David Dubois

Ashish Kumar

A method includes receiving a plurality of images, each of the plurality of images being a two-dimensional view of a portion of a three-dimensional object; receiving a definition of a three-dimensional volume in a same frame of reference as the plurality of images; receiving a label for the three-dimensional volume; projecting the three-dimensional volume onto each of the plurality of images; determining, for each of the plurality of images an image metric; and determining whether to assign the label to respective ones of the plurality of images based on the image metric.

Artificial intelligence system including three-dimensional labeling using frame of reference projections

A method includes defining a plurality of data storage policies, each of the plurality of data storage policies providing rules for storing data among a plurality of data storage locations, each of the plurality of data storage locations having a data storage cost and a data retrieval cost associated therewith; determining a baseline policy distribution among the plurality of data storage policies for an entity; receiving new data items corresponding to the entity; storing the new data items in the plurality of data storage locations using the plurality of data storage policies based on the baseline policy distribution; and determining, using the artificial intelligence engine, a selected one of the plurality of data storage policies to use in storing the new data items corresponding to the entity based on the data storage cost for each of the plurality of data storage locations, and the data retrieval cost for each of the plurality of storage locations.

Selection of health care data storage policy based on historical data storage patterns and/or patient characteristics using an artificial intelligence engine

Systems and methods for displaying a predicted outcome of a lung volume reduction procedure for a patient including a user interface, a processor, and programing operable on the processor for displaying a predicted outcome of the bronchoscopic lung volume reduction procedure on the user interface, wherein displaying the predicted outcome of the lung volume reduction procedure includes receiving patient data comprising volumetric images of the patient, analyzing the volumetric images to identify one or more features correlated to treatment outcome prediction, predicting an outcome for a treatment modality or treatment device using the one or more identified features, and displaying the predicted outcome on the user interface.

Treatment outcome prediction for lung volume reduction procedures

Methods and systems for assessing lung function using volumetric images obtained at inspiration and expiration. The method may include processing the first and second set of images to identify known anatomical structures of the lungs, registering the first set of images to the second set of images to match voxels of the first set of images to voxels of the second set of images as matched pairs of inspiratory and expiratory voxels, calculating a continuous probability of a lung characteristic at a location of the matched pairs of voxels, and displaying the result on a display. The method may also include classifying lung tissue at each location as normal, having air trapping without emphysema, or being emphysematous.

Visualization and quantification of lung disease utilizing image registration

Analysis of pulmonary scans representative of a patient's pulmonary structure can be used to classify a patient into one or more of a plurality of populations. The patient's scan can be mapped to a reference domain and analyzed to determine one or more fissure features associated with a plurality of regions in the reference domain. Comparison of the determined fissure features with a plurality of fissure atlases, each associated with different population, can be performed to classify the patient into one or more of the populations. Data from different fissure atlases can be compared to determine regions in the fissure atlases that distinguish one population from another. Such distinguishing regions can improve the ability to classify the patient while reducing errors based on false classifications.

Systems and methods for quantifying regional fissure features

Systems and methods for visualizing pulmonary fissures including a processor and software instructions for creating a 3 dimensional model of the fissures. Creating the 3 dimensional model includes accessing volumetric imaging data of the patient's lungs, analyzing the volumetric imaging data to segment the lungs into lobes, using the segmented lobes to identify locations at which pulmonary fissures should be present where the lobes abut each other, analyzing the volumetric images to identify locations at which pulmonary fissures actually are present as existing fissure, comparing the locations at which pulmonary fissures should be present to the locations at which pulmonary fissures are present to identify locations of missing fissure, and creating a visual display comprising a 3 dimensional model of the pulmonary fissures including existing fissure portions and missing fissure portions, with the existing fissure portion visually distinct from the missing fissure portions.

Visualization and characterization of pulmonary lobar fissures

An intuitive user interface is provided that allows for selection of an abnormality level in features in a medical image on the basis of graphical depictions of the abnormalities themselves. In one embodiment, the user interface displays realistic images of actual abnormalities in a series ranging from least severe to most severe and the medical practitioner selects the abnormality level by mouse-clicking on one of the images. The user interface then displays an annotated map of suspected abnormalities with annotations only on those suspected abnormalities that are at least as severe as the selected abnormality. In alternative embodiments, the user interface displays stylized images of the abnormalities plotted by the display using mathematical functions. In yet another alternative, the images are hand-drawn images.

Method and apparatus for setting a detection threshold in processing medical images

An object in a video sequence is tracked by object masks generated for frames in the sequence. Macroblocks are motion compensated to predict the new object mask. Large differences between the next frame and the current frame detect suspect regions that may be obscured in the next frame. The motion vectors in the object are clustered using a K-means algorithm. The cluster centroid motion vectors are compared to an average motion vector of each suspect region. When the motion differences are small, the suspect region is considered part of the object and removed from the object mask as an occlusion. Large differences between the prior frame and the current frame detect suspected newly-uncovered regions. The average motion vector of each suspect region is compared to cluster centroid motion vectors. When the motion differences are small, the suspect region is added to the object mask as a disocclusion.

Occlusion/disocclusion detection using K-means clustering near object boundary with comparison of average motion of clusters to object and background motions

In a preferred embodiment a radiologist traces the pulmonary artery and pulmonary veins visible in a set of CT images and identifies the arteries and veins. The radiologist's identification of the pulmonary arteries and pulmonary veins is then received by an image analyzer and combined with the analyzer's identification of the pulmonary arteries to form a combined identification; and the analyzer then reviews this combined identification of the pulmonary arteries to detect any pulmonary embolisms. The radiologist's identification of any pulmonary embolisms is compared with the analyzer's identification of any pulmonary embolisms to determine if there are any embolisms identified by the analyzer that were not identified by the radiologist.

Method for segmenting arteries and veins

Detection inputs are classified using two thresholds. In a preferred embodiment, a multiplicity of inputs are scored and the scored inputs are searched to locate an input that has a score greater than a first threshold. If such input is found, every input having a score in excess of a second threshold lower than the first threshold is identified as belonging to a first class of interest. If no input is found having a score in excess of the first threshold, no inputs are identified as belonging to the first class.

Growing community of inventors

Edina, MN, United States of America

Philippe Raffy