Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12597088-B2
Publication Date
2026-04-07
Expiration Date
Abstract
Systems and methods are disclosed for performing operations comprising: receiving a new data sample comprising estimated motion; identifying a cluster of training data samples from a plurality of training data samples that corresponds to the new data sample; obtaining motion estimations associated with the cluster of training data samples; determining a reconstruction from the cluster of training data samples to the new data sample; transforming the motion estimations for the cluster of training data samples based on the reconstruction from the cluster of training data samples to the new data sample; and computing a quality factor indicating a measure of similarity between the estimated motion of the new data sample and the motion estimations associated with the cluster of training data samples.
Core Innovation
The disclosed system and method assess the reliability of motion estimates for a new data sample. They identify a corresponding cluster of training data samples, obtain its motion estimations, determine a reconstruction from the cluster to the new data sample, and transform the motion estimations based on that reconstruction. The system or method computes a quality factor indicating similarity between the new sample's estimated motion and the transformed motion estimations.
The quality factor is compared with one or more criteria, and the comparison is used to adjust a weight associated with an image-registration result for the new data sample. The partial content identifies reliability assessment as the purpose but does not explicitly describe the problem in the background section. It also describes use in radiotherapy workflows and identifies medical images as a data context.
Claims Coverage
The independent claims are claim 1, directed to a system, and claim 14, directed to a method. They cover the same core workflow through three inventive features.
Corresponding cluster selection and reconstruction
Receive a new data sample comprising estimated motion; identify a cluster of training data samples that corresponds to the new data sample; obtain motion estimations associated with the cluster; and determine a reconstruction from the cluster to the new data sample.
Transformation of cluster motion estimations
Transform the motion estimations for the cluster of training data samples based on the reconstruction from the cluster to the new data sample.
Quality-factor-based adjustment of image-registration weight
Compute a quality factor indicating a measure of similarity between the estimated motion of the new data sample and the motion estimations associated with the cluster, compare the quality factor to one or more criteria, and, based on the comparison, adjust a weight associated with a result of performing image registration for the new data sample.
Both independent claims cover selecting a corresponding training-data cluster, reconstructing the new data sample from that cluster, transforming its motion estimations, and using a similarity quality factor to adjust the weight associated with an image-registration result.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Assessing motion-estimate reliability for medical images, including in radiotherapy workflows.
Interested in licensing this patent?