Automatic image registration quality assessment

Abstract

The medical registration training component implemented within the medical registration system trains the medical registration operation on a pair of medical studies. In response to the medical registration training component determining that the training medical registration operation was successful, the medical registration training component records the medical registration instance for the pair of medical studies in the medical registration history and marks the medical registration instance as a positive instance in the medical registration history. In response to the medical registration training component determining that training the medical registration operation requires correction, the medical registration training component records the medical registration instance for the medical study pair in the medical registration history and marks the medical registration instance as the medical registration history. The medical registration training component uses machine learning to train the failure prediction machine learning model based on the medical registration history such that the failure prediction machine learning model predicts whether a new medical registration operation needs to be corrected. In response to predicting a need for correction, the mechanism takes steps to automatically correct the new medical registration operation.

Description

Background technique [0001] The present application relates generally to improved data processing apparatus and methods, and more particularly to mechanisms for automatic image registration quality assessment. [0002] Image registration is the process of transforming different sets of data into a coordinate system. Data may be multiple photos, data from different sensors, time, depth or viewpoint. It is used in computer vision, medical imaging, military automatic target recognition, and compiling and analyzing satellite imagery and data. In order to compare or integrate data obtained from these different measurements, registration is necessary. [0003] Image registration or image alignment algorithms can be categorized as intensity-based and feature-based. One of the images is called the reference or source, and the other images are called the target, sensed or subject images respectively. Image registration involves spatially registering target images to align with refe...

AI Technical Summary

Problems solved by technology

Due to the large number of applications in which image registration can be applied, it is not possible to develop a general method optimized for all uses

Images