Method and system for machine learning based assessment of fractional flow reserve

Abstract

A method and system for anatomical object detection using marginal space deep neural networks is disclosed. The pose parameter space for an anatomical object is divided into a series of marginal search spaces with increasing dimensionality. A respective deep neural network is trained for each of the marginal search spaces, resulting in a series of trained deep neural networks. Each of the trained deep neural networks can evaluate hypotheses in a current parameter space using discriminative classification or a regression function. An anatomical object is detected in a medical image by sequentially applying the series of trained deep neural networks to the medical image.

Description

[0001] This application claims the benefit of U.S. Provisional Application No. 62 / 148,273, filed April 16, 2015, and U.S. Provisional Application No. 62 / 121,782, filed February 27, 2015, and filed October 16, 2014 Continuation-in-Part of filed U.S. Application No. 14 / 516,163 claiming the benefit of U.S. Provisional Application No. 61 / 891,920 filed October 17, 2013, the disclosures of which are incorporated by reference This article. technical field [0002] The present invention relates to anatomical object detection in medical image data, and more particularly, to anatomical object detection in medical image data using deep neural networks. Background technique [0003] Fast and robust anatomical object detection is a fundamental task in medical image analysis that underpins the entire clinical imaging workflow from diagnosis, patient stratification, treatment planning, intervention, and follow-up. Automatic detection of anatomical objects is a prerequisite for many medica...

AI Technical Summary

Problems solved by technology

However, anatomical object detection using MSL is not always robust, especially for some challenging detection problems where anatomical objects exhibit large variations in anatomy, shape, or appearance in medical images

Images