US 11,721,437 B2
Digital twin for predicting performance outcomes
Daniel P. Nicolella, San Antonio, TX (US); Kase J. Saylor, San Antonio, TX (US); and Mark J. Libardoni, San Antonio, TX (US)
Assigned to Southwest Research Institute, San Antonio, TX (US)
Filed by Southwest Research Institute, San Antonio, TX (US)
Filed on Jun. 2, 2019, as Appl. No. 16/429,024.
Claims priority of provisional application 62/680,215, filed on Jun. 4, 2018.
Prior Publication US 2019/0371466 A1, Dec. 5, 2019
Int. Cl. G16H 50/20 (2018.01); G06T 17/00 (2006.01); G06F 30/20 (2020.01)
CPC G16H 50/20 (2018.01) [G06F 30/20 (2020.01); G06T 17/00 (2013.01)] 4 Claims
OG exemplary drawing
 
1. A method of monitoring activity of an animate real-time subject, comprising:
generating a digital model for each of a control group of subjects, using internal imaging data obtained from the subjects;
wherein each model represents at least a model of the subject's internal musculoskeletal system;
collecting movement data of the subjects, using a motion capture system applied to the subjects;
combining the model and the movement data, thereby generating a dynamic digital twin of the subjects;
activating each digital twin to perform a specified physical activity;
wherein the activating step includes at least activating muscles of the musculoskeletal system of the digital twin to produce motion of the digital twin to perform the physical activity;
identifying which subjects of the control group of subjects successfully complete the physical activity as measured from data collected from the digital twins of the control group of subjects thereby identifying a subset of nominal performance digital twins having nominal performance data;
collecting actual activity data of the real-time subject, using at least one wearable sensor worn by the subject during the same physical activity as in the activating step;
comparing the actual activity data to the nominal performance data; and
determining if the actual activity data exceeds a threshold of the nominal performance data.