Method and system for analyzing image data
Inventors
Jain, Saurabh • SMEETS, Dirk • SIMA, Diana • RIBBENS, Annemie • MAERTENS, Anke
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-10198815-B2
Publication Date
2019-02-05
Expiration Date
Abstract
A method of analyzing image data comprises: obtaining a first image of a first part of an object; obtaining a second image of a second part of the object having overlap with the first part; obtaining a mapping between the first and second images; segmenting the second image to obtain a segmentation; detecting outliers in the first image by identifying extreme intensity values of elements within one or more classes of elements on the basis of the segmentation; replacing elements of the second image that correspond to at least some outliers of the first image, with replacement values, to obtain a corrected second image; and updating the segmentation by performing the segmenting on the corrected second image. The detecting outliers, the replacing, and the updating are performed iteratively until a predetermined convergence criterion is met, which represents a point where there is no significant change in the tissue and lesion segmentations.
Core Innovation
The invention analyzes image data by obtaining a first image of a first part of an object and obtaining a second image of a second part of the object, where the second part has substantial overlap with the first part. A mapping between the first image and the second image is obtained, and the second image is segmented to obtain a segmentation. Outliers in the first image are detected by identifying extreme intensity values of elements within one or more classes of elements on the basis of the segmentation.
Elements of the second image that correspond, according to the mapping, to at least some of the detected outliers of the first image are replaced with replacement values to obtain a corrected second image. The segmentation is updated by performing the segmenting on the corrected second image. The outlier detecting, replacing, and updating are performed iteratively until a predetermined convergence criterion is met.
The predetermined convergence criterion represents a point at which there is no significant change in the tissue and the lesion segmentation. In a neuroimaging embodiment, the first image and the second image are a FLAIR image and a T1 image, respectively, with tissue classes including gray matter, white matter, and cerebro-spinal fluid (CSF). Outliers are detected in the FLAIR image among elements classified within tissue classes, and mapped outlier-corresponding elements are replaced in the second image to form a corrected image, after which tissue/lesion segmentation and quantitative measures are updated until convergence.
Claims Coverage
The document contains three independent claims covering a method, a non-transitory computer-readable medium, and an image processing system. Each independent claim includes the same core workflow features, totaling six main inventive actions across obtaining, mapping, segmenting, detecting outliers, replacing, and iterative updating to a convergence criterion.
Iterative outlier detection and correction on overlapping images with segmentation-defined classes
Obtaining a first image and a second image having substantial overlap, obtaining a mapping between the first image and the second image, segmenting the second image to obtain a segmentation, detecting outliers in the first image by identifying extreme intensity values of elements within one or more classes of elements on the basis of the segmentation, replacing elements of the second image that correspond (according to the mapping) to at least some of the detected outliers with replacement values to obtain a corrected second image, and updating the segmentation by performing the segmenting on the corrected second image, with detecting, replacing, and updating performed iteratively until a predetermined convergence criterion is met representing no significant change in the tissue and the lesion segmentation.
Corrected second image produced from mapped outlier-corresponding elements
Replacing elements of the second image that correspond, according to the mapping, to at least some of the detected outliers of the first image with replacement values to obtain a corrected second image.
Convergence criterion defined by no significant change in tissue and lesion segmentation
Performing the detecting outliers, the replacing, and the updating iteratively until a predetermined convergence criterion is met, where the convergence criterion represents a point at which there is no significant change in the tissue and the lesion segmentation.
Across the independent method claim, the independent medium claim, and the independent system claim, the coverage centers on iterative outlier detection in a first image using extreme intensity values within segmentation-defined classes, mapped replacement in a second image to create a corrected second image, and repeated segmentation updates until a convergence criterion is met defined as no significant change in tissue and lesion segmentation.
Stated Advantages
Convergence is achieved at a point where there is no significant change in the tissue and the lesion segmentation.
Enables detection and correction of outliers in overlapping images using segmentation-defined classes and extreme intensity values.
Documented Applications
Neuroimaging analysis using FLAIR (first) and T1 (second) images, including tissue and lesion segmentation and measurement of volume or area across anatomical regions, including MS lesions.
Interested in licensing this patent?