Patent Issued for Biomechanics abnormality identification (USPTO 11622729): Cerner Innovation Inc.

2023 APR 27 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- From Alexandria, Virginia, NewsRx journalists report that a patent by the inventors McNair, Douglas S. (Seattle, WA, US), filed on June 11, 2020, was published online on April 11, 2023.

The patent’s assignee for patent number 11622729 is Cerner Innovation Inc. (Kansas City, Kansas, United States).

News editors obtained the following quote from the background information supplied by the inventors: “Musculoskeletal diseases, which include back pain, arthritis, bodily injuries, and osteoporosis, are reported by persons in the United States more than any other health condition. In 2004, the estimated total cost of treatment and lost wages associated with musculoskeletal diseases was $850 billion, approximately 8% of the gross domestic product (GDP), and the estimated cost for treating patients with musculoskeletal conditions was $510 billion.

“More than three of every five accidental injuries that occur annually in the United States are to the musculoskeletal system. In 2004, more than 57.2 million musculoskeletal injuries were treated in health care settings, accounting for 60% of injuries of all types treated that year. Musculoskeletal injuries include sprains and strains, usually incurred during sudden movement or over-use, (16.3 million injuries in 2004); fractures (15.3 million); open wounds and other open traumatic injuries (10.3 million); and contusions and bruises and other closed traumatic injuries (8.4 million). The estimated cost in 2004 of treating all musculoskeletal injuries was $127.4 billion. Since 1996, in 2004 dollars, the cost of treating musculoskeletal injuries has risen 37%. The share of cost attributed to prescription drugs rose from 11% in 1996 to 17% in 2004.

“Thus, musculoskeletal conditions and their management are important epidemiologically and economically. Therefore, systematic and efficient diagnosis and management of these conditions have high clinical and financial value, not only in terms of direct expense, but also in terms of absenteeism among persons of employment age and lost productivity.”

As a supplement to the background information on this patent, NewsRx correspondents also obtained the inventors’ summary information for this patent: “Systems, methods, and computer-readable media are provided for the automatic identification of patients or athletes who have an existing disability or an acute or chronic injury, or who have an elevated near-term risk of musculoskeletal injury or disability, or healthy individuals whose musculoskeletal performance characteristics are the subject of optimization, training, or injury-prevention efforts. An embodiment is directed to classification and diagnosis, risk stratification, and optimization of assessment, communication, and decision-making to prevent or manage musculoskeletal injury in humans. An embodiment takes the form of a platform for analyzing 3-D motion data from high-speed multi-camera imaging devices with embedded decision support software for calculating biclusters. An embodiment takes the form of a 3-D digital motion-capture system that is connected via network to a decision support system that implements biclustering in a web-based cloud computing configuration. Thus, the aim of an embodiment relates to automatically identifying persons who potentially have a plurality of materialized abnormal musculoskeletal conditions or who have features that may dispose such persons toward such abnormal conditions by using signal-processing software and statistical predictive algorithms. This system calculates biclusters and bicluster membership properties of multi-variable static or dynamic biomechanics data acquired by the motion-capture system to enable detection and categorization of such abnormalities or predisposing features.

“The measurements and predictive and classificatory algorithms enable use in sports medicine and rehabilitation and other ambulatory environments, as well as in general acute-care and chronic-care venues, and afford a degree of robustness against variations in individual anatomy and session-to-session variations in movements executed by an individual. An embodiment provides a leading indicator of likely near-term future abnormalities, proactively notifying clinicians responsible for the care of the individual and providing the care providers sufficient advance notice to enable effective preventive maneuvers to be undertaken. In an embodiment, involving serial testing of a given individual, a clinician is notified of actionable changes in classification and bicluster membership of an individual-either favorable or unfavorable-for the purposes of adjusting the regimen for managing the individual’s condition(s). In an exemplary embodiment, a device is integrated with case-management software and electronic health record decision-support systems, including occupational health, health insurance, and disability assessment decision-support systems.”

The claims supplied by the inventors are:

“1. A system comprising: a three-dimensional (3-D) motion capture system; at least one processor; and one or more computer storage media storing computer-readable instructions that, when executed by the processor, cause the processor to perform operations comprising: capturing from the 3-D motion capture system a set of movements for reference subjects performing a regimen of physical motion; generating an index comprising a first data dimension associated with biomechanics motion data determined from the captured set of movements and a second data dimension associated with motion variables, each reference subject having an associated variable array of at least a portion of the biomechanics motion data for the motion variables; and creating biclusters from the biomechanics motion data using a biclustering technique on the generated index, wherein the generated index predicts a musculoskeletal injury of a test subject performing the regimen of physical motion based on association of the test subject with at least one bicluster.

“2. The system of claim 1, wherein creating the biclusters comprises simultaneously producing marker-condition clusters and condition-subject clusters.

“3. The system of claim 1, further comprising a physiological variable probe configured to measure physiological data of the reference subjects, wherein the second data dimension is further associated with physiological variables, and wherein the associated variable array for each reference subject comprises measured physiological data of a reference subject of the reference subjects.

“4. The system of claim 1, further comprising removing a selected variable array from the generated index, wherein the biclusters are created from the generated index without the selected variable array.

“5. The system of claim 1, further comprising receiving a test subject variable array comprising at least a portion of the biomechanics motion data that is associated with the test subject, wherein the biclusters are created from the generated index and the test subject variable array.

“6. The system of claim 1, further comprising: employing a quantile threshold to determine a score for each of the motion variables; and prior to creating the biclusters from the generated index, removing a portion of the biomechanics motion data based on the quantile threshold score.

“7. The system of claim 1, wherein the biclustering technique comprises one of Coupled Two-Way Clustering (CTWC), Order Preserving Submatrix (OPSM), Iterative Signature Algorithm (ISA), Binary Inclusion Maximal algorithm (BIMAX), association analysis based Range Support Patterns (RAP), Combinatorial Algorithm for Expression and Sequence-Based Cluster Extraction (COALESCE), Qualitative Biclustering (QUBIC), Statistical-Algorithmic Method for Bicluster Analysis (SAMBA), Factor Analysis for Bicluster Acquisition (FABIA), spectral clustering, Sparse Singular Value Decomposition (SSVD), or sparse singular value decomposition incorporating stability selection (S4VD).

“8. A computer-implemented method comprising: receiving biomechanics motions data associated with a set of movements of reference subjects performing a regimen of physical motion captured by a three-dimensional (3-D) motion capture system; generating an index comprising a first data dimension associated with the biomechanics motion data and a second data dimension associated with motion variables, each reference subject having an associated variable array comprising at least a portion of the biomechanics motion data; creating biclusters from the biomechanics motion data using a biclustering technique on the generated index; and providing the generated index to predict a musculoskeletal injury of a test subject performing the regimen of physical motion based on association of the test subject with at least one bicluster.

“9. The method of claim 8, wherein creating the biclusters comprises simultaneously producing marker-condition clusters and condition-subject clusters.

“10. The method of claim 8, wherein the second data dimension is further associated with physiological variables, and wherein the associated variable array for each reference subject comprises measured physiological data of a reference subject of the reference subjects.

“11. The method of claim 8, further comprising removing a selected variable array from the generated index, wherein the biclusters are created from the generated index without the selected variable array.

“12. The method of claim 8, further comprising receiving a test subject variable array of at least the portion of the biomechanics motion data associated with the test subject, wherein the biclusters are created from the generated index and the test subject variable array.

“13. The method of claim 8, further comprising: employing a quantile threshold to determine a score for each of the motion variables; and prior to creating the biclusters from the generated index, removing a portion of the biomechanics motion data based on the quantile threshold score.

“14. The method of claim 8, wherein the biclustering technique comprises one of Coupled Two-Way Clustering (CTWC), Order Preserving Submatrix (OPSM), Iterative Signature Algorithm (ISA), Binary Inclusion Maximal algorithm (BIMAX), association analysis based Range Support Patterns (RAP), Combinatorial Algorithm for Expression and Sequence-Based Cluster Extraction (COALESCE), Qualitative Biclustering (QUBIC), Statistical-Algorithmic Method for Bicluster Analysis (SAMBA), Factor Analysis for Bicluster Acquisition (FABIA), spectral clustering, Sparse Singular Value Decomposition (SSVD), or sparse singular value decomposition incorporating stability selection (S4VD).

“15. A computer-implemented method comprising: receiving biomechanics motions data associated with a set of movements of reference subjects performing a regimen of physical motion captured by a three-dimensional (3-D) motion capture system, wherein a portion of the reference subjects have a known musculoskeletal injury; generating an index comprising a first data dimension associated with the biomechanics motion data and a second data dimension associated with motion variables, each reference subject having an associated variable array comprising at least a portion of the biomechanics motion data for the motion variables; creating biclusters from the biomechanics motion data using a biclustering technique on the generated index; associating the known musculoskeletal injury to a bicluster; and providing the generated index to predict a musculoskeletal injury of a test subject performing the regimen of physical motion based on association of the test subject with the bicluster associated with the known musculoskeletal injury.

“16. The method of claim 15, wherein creating the biclusters comprises simultaneously producing marker-condition clusters and condition-subject clusters.

“17. The method of claim 15, wherein the second data dimension is further associated with physiological variables, and wherein the associated variable array for each reference subject comprises measured physiological data of a reference subject of the reference subjects.

“18. The method of claim 15, further comprising removing a selected variable array from the generated index, wherein the biclusters are created from the generated index without the selected variable array.

“19. The method of claim 15, further comprising receiving a test subject variable array of at least a portion of the biomechanics motion data associated with the test subject, wherein the biclusters are created from the generated index and the test subject variable array.

“20. The method of claim 15, further comprising: employing a quantile threshold to determine a score for each of the motion variables; and prior to creating the biclusters from the generated index, removing a portion of the biomechanics motion data based on the quantile threshold score.”

For additional information on this patent, see: McNair, Douglas S. Biomechanics abnormality identification. U.S. Patent Number 11622729, filed June 11, 2020, and published online on April 11, 2023. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(11622729)&db=USPAT&type=ids

(Our reports deliver fact-based news of research and discoveries from around the world.)