Method And Apparatus To Classify Structures In An Image
AssigneeMedtronic Navigation, Inc.
InventorsRowena Vigh, Hallehsadat Ghaderi, Daniel H. Adler, Shai Ronen, Nikhil Mahendra
Life of this patent
Estimating the term…
Abstract
Disclosed is a system and method for segmentation of selected data. In various embodiments, automatic segmentation of fiber tracts in an image data may be performed. The automatic segmentation may allow for identification of specific fiber tracts in an image.
Claims22
A method to segment neuronal tracts in an image, comprising:
- accessing selected data of a subject;
- evaluating the accessed selected data to determine an image tractography including a plurality of tracts;
- evaluating at least a sub-plurality of tracts of the plurality of tracts with a trained machine learning algorithm;
- determining whether at least one of the evaluated sub-plurality of tracts is a given fiber neuronal tract based at least on the evaluation of at least the sub-plurality of tracts of the plurality of tracts;
- and
- outputting which at least one tract of the plurality of tracts is/are the given fiber neuronal tract when at least one of the evaluated plurality of tracts is determined to be the given fiber neuronal tract.
The method of claim 2, wherein the accessed selected data includes diffusion weighted gradient images.
The method of claim 3, wherein evaluating the accessed selected data based on the first criteria to determine the image tractography including the plurality of tracts includes:
- determining an anisotropy of water within the selected data;
- and
- determining tracts through an image in the selected data based on the determined anisotropy.
A system configured to segment given neuronal tracts in an image of a subject, comprising:
- a processor system configured to execute instructions to:
- access a selected data of a subject;
- evaluate the accessed selected data based on a first criteria to determine an image tractography including a plurality of tracts;
- evaluate at least a sub-plurality of tracts of the plurality of tracts with a trained machine learning algorithm;
- determine whether at least one of the evaluated sub-plurality of tracts is a given fiber neuronal tract based at least on the evaluation of at least the sub-plurality of tracts of the plurality of tracts;
- and
- output which at least one tract of the plurality of tracts is/are the given fiber neuronal tract when at least one of the evaluated plurality of tracts is determined to be the given fiber neuronal tract.
Dependencies are inferred from the claim text.