Systems and methods for extremity analysis and modeling using medical imaging
AssigneeSpringbok, Inc.
InventorsLara Riem, Silvia Blemker, Xue Feng, Matthew Cousins, Olivia DuCharme
Life of this patent
Estimating the term…
Abstract
Systems and methods are provided for upper extremity analysis and modeling using medical imaging. The systems and methods may be used to identify and segment both muscle and fatty tissue directly from the standard clinical images, and may be fully automated, thus providing rapid assessments back to the clinician and seamless integration into the clinical decision-making workflow.
Claims35
A system for generating a report providing information regarding a clinical diagnosis or prognosis from medical images, the system comprising:
- a memory configured to store a plurality of a medical images of an upper extremity of a subject;
- a processor configured to access the plurality of medical images of the upper extremity of the subject stored in the memory and configured to:
- segment the plurality of images based on at least muscle, fat, and bone;
- identify and correct images for scan coverage variance based on a function of a distance on a bone as a percentage of total length of the bone;
- determine at least one of extra-muscular fat or intra-muscular fat using segmentation of the plurality of images;
- quantify at least one of structural volume, muscle volume, or fat infiltration in the upper extremity of the subject using the segmentation of the plurality of images and the determined at least one of extra-muscular fat or intra-muscular fat;
- and
- generate a report providing information indicating pre-surgical or a post-surgical diagnosis or prognosis for the upper extremity including at least the muscle volume or fat infiltration ratio.
A computer-implemented method for automatic muscle, fat, and bone segmentation using at least one processor, the method comprising:
- accessing image data acquired by a medical imaging system;
- segmenting at least one bone in the image data using an artificial intelligence network
- analyzing the image data and correcting for scan coverage variance based on a function of a distance on the bone as a percentage of a total length of the bone;
- segmenting at least one muscle in the image data using the artificial intelligence network;
- segmenting fat in the image data using the artificial intelligence network;
- generating label maps by combining the fat segmented image data with the muscle segmented image data;
- determining at least one of extra-muscular fat or intra-muscular fat based on the label maps;
- and
- quantifying muscle volume for the at least one muscle and a fat infiltration ratio based on the determined at least one of extra-muscular fat or intra-muscular fat.
A system for automatic muscle, fat, and bone segmentation comprising:
- a computer system configured to:
- i) access image data acquired by a medical imaging system;
- ii) segment at least one bone in the image data using an artificial intelligence network;
- iii) analyze the image data and correct for scan coverage variance based on a function of a distance on the bone as a percentage of a total length of the bone;
- iv) segment at least one muscle in the image data using the artificial intelligence network;
- v) segment fat in the image data using the artificial intelligence network;
- vi) generate label maps by combining the fat segmented image data with the muscle segmented image data;
- vii) determine at least one of extra-muscular fat or intra-muscular fat based on the label maps;
- and
- viii) quantify muscle volume for the at least one muscle and a fat infiltration ratio based on the determined at least one of extra-muscular fat or intra-muscular fat.
Dependencies are inferred from the claim text.