Koninklijke Philips Corporation N.v.

Prescott Peter Klassen

Lyubomir Georgiev Zagorchev

A method () of training a machine-learned (ML) image classifier () to classify images () respective to a set of labels includes: generating image-based labels for the images from the set of labels and image-based label confidence values for the image-based labels by applying the ML image classifier to the images; generating report-based labels for the images from the set of labels and report-based label confidence values for the report-based labels by applying a report classifier () to corresponding radiology reports; selecting a training subset of the set of images based on the image-based labels, the report-based labels, the image-based label confidence values, and the report-based label confidence values; assigning a pseudo-label for each image of the training subset which is one of the image-based label or the report-based label for the image; and training the ML image classifier using at least the selected training subset and the assigned pseudo-labels.

Semi-supervised learning using co-training of radiology report and medical images

Systems and methods for magnetic resonance (MR) examination are provided. In an embodiment, a method for MR examination includes receiving, at a computing device in communication with a magnetic resonance imaging (MRI) device, MR data of a patient body comprising a plurality of anatomical structures, the plurality of anatomical structures including a region of interest, segmenting, by the computing device, the MR data to obtain geometries of the plurality of anatomical structures, receiving, at the computing device, a report comprising text descriptions representative of the plurality of anatomical structures, associating, by the computing device, the text descriptions with respective geometries of the plurality of anatomical structures, identifying, by the computing device, the text descriptions associated with anatomical structures within or outside of the region of interest; and outputting, by the computing device, a graphical representation based on the identified text descriptions associated with anatomical structures within or outside of the region of interest.

Growing community of inventors

CAMBRIDGE, MA, United States of America

Prescott Peter Klassen