Retaining observational information when transmitting markerless motion capture data

Recent advances in 3D camera technology and associated software is enabling clinical assessments of function to be undertaken in real world clinical settings, rather than in specialized bio-mechanics laboratories. Working in collaboration with Kinetisense Inc, an SME from Medicine Hat Alberta, a patented markerless motion capture technology has been developed to support a wide
range of rehabilitation assessments for patients in rural and remote settings. PROMote is a platform developed at University of Alberta to support more complex clinic-to-clinic assessments incorporating virtual health technologies. Privacy restrictions prohibit storage of identifiable video images of patients in centralized electronic medical records. Although technologies such as facial
blurring are feasible they are not currently acceptable. Another challenge when transferring video images is limited bandwidth, clinician time for file transfers and large file storage requirements. Instead the Kinetisense software generates a spreadsheet of marker locations at 30 frames a second in csv format. Therefore no patient identifiable data is retained. This project will take
typical datasets, remove artifacts through filtering, and reconstruct the movements in the form of a realistic avatar of the patient. A large library of data is already available and consented by patients for analysis.

Faculty Supervisor:

Martin Ferguson-Pell

Student:

Partner:

PSG College of Technology

Discipline:

Computer science

Sector:

Health and Related Sciences & Technology; Information and Communications Technology; New and Digital Media

University:

University of Alberta

Program: