Google Inc.

Marcin Sieniek

Jeffrey De Fauw

Scott McKinney

Shravya Shetty

Varun Godbole

Jonathan Godwin

Christopher Kelly

Natasha Antropova

A method is disclosed of processing a set of images. Each image in the set has an associated counterpart image. One or more regions of interest (ROIs) are identified in one or more of the images in the set of images. For ROI identified, a reference region is identified in the associated counterpart image. ROIs and associated reference regions are cropped out, thereby forming cropped pairs of images 1 . . . n, that are fed to a deep learning model trained to make a prediction of probability of a state of the ROI, e.g., disease state, which generates a prediction P-, (i=1 . . . n) for each cropped pair. The model generates an overall prediction P from each of the predictions P. A visualization of the set of medical images and the associated counterpart images including the cropped pair of images is generated.

System and method for interpretation of multiple medical images using deep learning

A method and system for assessing a machine learning model providing a prediction as to the disease state of a patient from a 2D or 3D image of the patient or a sample obtained therefrom. The machine learning model produces a prediction of the disease state from the image. The method involves presenting on a display of a workstation the image of the patient or a sample obtained therefrom along with a risk score or classification associated with the prediction. The image is further augmented with high-lighting to indicate one or more regions in the image which affected the prediction produced by the machine learning model. Tools are provided by which the user may highlight one or more regions of the image which the user deems to be suspicious for the disease state. Inference is performed on the user-highlighted areas by the machine learning model. The results of the inference are presented to the user via the display.

System, user interface and method for interactive negative explanation of machine learning localization models in health care applications

A method and system for assessing a machine learning model providing a prediction as to the disease state of a patient from a 2D or 3D image of the patient or a sample obtained therefrom. The machine learning model produces a prediction of the disease state from the image. The method involves presenting on a display of a workstation the image of the patient or a sample obtained therefrom along with a risk score or classification associated with the prediction. The image is further augmented with highlighting to indicate one or more regions in the image which affected the prediction produced by the machine learning model. Tools are provided by which the user may highlight one or more regions of the image which the user deems to be suspicious for the disease state. Inference is performed on the user-highlighted areas by the machine learning model. The results of the inference are presented to the user via the display.

System, user interface and method for interactive negative explanation of machine-learning localization models in health care applications

Growing community of inventors

Mountain View, CA, United States of America

Marcin Sieniek