SBIR-STTR Award

A Probabilistic Pose Estimation Algorithm For 3d Motion Capture Data
Award last edited on: 11/7/19

Sponsored Program
SBIR
Awarding Agency
NIH : NIAMS
Total Award Amount
$1,631,028
Award Phase
2
Solicitation Topic Code
-----

Principal Investigator
William Scott Selbie

Company Information

C-Motion Inc

20030 Century Boulevard Suite 104A
Germantown, MD 20874
   (301) 540-5611
   info@c-motion.com
   www.c-motion.com
Location: Multiple
Congr. District: 06
County: Montgomery

Phase I

Contract Number: 1R43AR064620-01A1
Start Date: 7/15/13    Completed: 1/14/14
Phase I year
2013
Phase I Amount
$149,084
Orthopaedic disorders are a leading cause of disability in the U.S., with arthritis and/or spine problems adversely affecting quality of life fo more than 20% of adults. With an aging population, the rate of disability from orthopaedic disorders has been increasing steadily. While advances in diagnostic imaging (including CT, MRI and ultrasound) have greatly improved our ability to detect structural changes in musculoskeletal tissues, they typically reveal little about joint function. There is evidence that abnormal mechanical joint function contributes significantly to the development and progression of many types of joint disease. There is, therefore, a significant clinical need for the widespread use of technologies that can identify subtle abnormalities in joint function that, if left untreate, can compromise long-term joint health. The specific goal of this Phase I SBIR project is to extend C-Motion's capabilities for bone pose (position and orientation) estimation by incorporating the bi-planar radiography imaging expertise of Dr. Tashman's laboratory at the University of Pittsburgh. Dynamic Stereo X-ray is the only currently available technology that can achieve sub-mm accuracy during a wide variety of functional movements. To date, however, the potential promise of DSX to provide a unique and powerful tool for diagnosing subtle motion disorders and identifying at-risk joints has yet to be realized as a clinical tool, ad is still an under-used clinical research tool. The successful completion of this project sets the stage for the development of a fundamentally novel approach to quantitative radiography that would turn a limited-use research tool into a powerful clinical device for dynamic assessment of musculoskeletal function. Just as cine-angiography has revolutionized diagnosis and treatment of cardiovascular disorders, widespread availability of dynamic musculoskeletal imaging could significantly improve treatment for a wide variety of orthopaedic disorders.

Public Health Relevance Statement:


Public Health Relevance:
The successful completion of this Phase I proposal sets the stage for the development of a fundamentally novel approach to biplanar radiography that would turn a limited-use research tool into a powerful clinical device for rapid, dynamic assessment of musculoskeletal function. Just as cine-angiography has revolutionized diagnosis and treatment of cardiovascular disorders, widespread availability of dynamic musculoskeletal imaging could significantly improve treatment for a wide variety of orthopaedic disorders.

Project Terms:
Adult; Affect; aging population; Algorithms; Angiography; Architecture; Arthritis; arthropathies; base; Biomechanics; bone; Cardiovascular Diseases; Cartilage; Clinical; clinical application; Clinical Research; Computer software; Data; Data Set; design; Development; Devices; Diagnosis; Diagnostic Imaging; Diagnostic radiologic examination; disability; Disease; Environment; flexibility; Future; Goals; Health; Image; improved; Individual; Intervention; joint function; joint injury; Joints; kinematics; Kinetics; Knowledge; Laboratories; Left; Libraries; Ligaments; Magnetic Resonance Imaging; Mechanics; Medical; meetings; Metric; Morphologic artifacts; Motion; Movement; Musculoskeletal; musculoskeletal imaging/visualization/scanning; novel strategies; Orthopedics; Phase; Positioning Attribute; Process; public health relevance; Publishing; Quality of life; Relative (related person); Research; Research Infrastructure; Risk; Roentgen Rays; sensor; skeletal; Skin; Small Business Innovation Research Grant; soft tissue; software development; Sports; Staging; Structure; Technology; Testing; Time; Tissues; tool; Ultrasonography; United States National Institutes of Health; Universities; Vertebral column; Visual; Work

Phase II

Contract Number: 2R44AR064620-02
Start Date: 1/2/13    Completed: 3/31/17
Phase II year
2015
(last award dollars: 2017)
Phase II Amount
$1,481,944

Orthopaedic disorders are a leading cause of disability in the U.S., with arthritis and/or spine problems adversely affecting quality of life fo more than 20% of adults. While advances in diagnostic imaging have greatly improved our ability to detect structural changes in musculoskeletal tissues, they typically reveal little about joint function. There is evidence that abnormal mechanical joint function contributes significantly to the development and progression of many types of joint disease. There is, therefore, a significant clinical need for the widespread use of technologies that can identify subtle abnormalities in joint function that, if left untreated, can compromise long-term joint health. Dynamic Stereo X-ray (DSX) is the only currently available technology that can achieve sub-mm bone pose (position and orientation) estimation accuracy during a wide variety of functional movements. Over the past 15 years, Dr. Tashman has developed a sophisticated set of DSX software tools for his research involving the tracking of bones during various movements. In Phase I we implemented published key algorithms of the DSX in a modern development environment and added several important innovations that make it a better clinical tool. We used motion capture data from a 3D video-based system to seed the tracking optimization. We improved the operator interaction to manipulate seed poses manually. We designed and implemented a 4D algorithm, based on a global solution finder, that uses all time frames simultaneously. These innovations reduce the amount of operator time required to process a data set, reduce the noise in the solutions, and allow the use of asynchronous X-ray systems, which are much more common than synchronous systems. These innovations have been an important step toward making DSX software a robust clinical tool. Building on our success in Phase I, in Phase II we will introduce several innovations that will enable more regions of the body to be analyzed, and further reduce the amount of operator time and CPU time needed to analyze a movement. We will implement Dr. Tashman's published hierarchical algorithm in the 4D optimization to better track bones that overlap significantly with other bones (e.g., the spine) and to extend this algorithm to other regions of the body. We will also develop a modular system for defining anatomically meaningful coordinate systems in any bone, which is needed to represent joint kinematics of the tracked bones. We will quantify the robustness of the solution to inaccuracies in the input to provide guidelines for the required accuracy of the seed pose. Finally, we will implement the remaining DSX algorithms, including 3D calibration and distortion correction, to create a complete clinical package.

Public Health Relevance Statement:


Public Health Relevance:
Dynamic Stereo X-ray (DSX) mitigates uncertainty in 3D motion capture by directly measuring motion of the skeleton, and is the only currently available technology that can achieve sub-mm accuracy for a wide variety of functional movements. In this Phase II project we will develop an innovative clinical research application for DSX to provide rapid assessment of musculoskeletal function during dynamic activities. Just as cine- angiography has revolutionized diagnosis and treatment of cardiovascular disorders, widespread availability of DSX could significantly improve treatment for a wide variety of orthopaedic disorders.

Project Terms:
Adult; Affect; Algorithms; Angiography; animation; Architecture; Arthritis; arthropathies; base; Body Regions; bone; Bone Development; Bone Surface; Calibration; Cardiovascular Diseases; Clinical; Clinical Research; Computer software; computerized data processing; cone-beam computed tomography; Custom; Data; Data Set; design; Development; Diagnosis; Diagnostic Imaging; disability; Disease; Environment; Generic Drugs; Geometry; graphical user interface; Guidelines; Health; Implant; improved; innovation; joint function; Joints; kinematics; Left; Life; Maps; Measures; Mechanics; Methods; Modeling; Motion; Movement; Musculoskeletal; Noise; Optics; Orthopedics; Paint; parallel computer; parallel processing; Phase; Physiologic pulse; Plug-in; Positioning Attribute; Process; public health relevance; Publishing; Quality of life; Research; Roentgen Rays; Running; Scheme; Seeds; Skeleton; Software Tools; Solutions; success; Surface; System; Technology; Testing; Time; Tissues; tool; Translating; Uncertainty; Validation; Vertebral column; Work