Real-world testing of smartphone video body measurement for wheelchair fitting - ON-1264

Genre de projet: Recherche
Discipline(s) souhaitée(s): Génie -biomédical, Génie, Kinésiologie, Sciences de la vie, Informatique, Sciences mathématiques
Entreprise: Homni Technologies Inc.
Durée du projet: 4 à 6 mois
Date souhaitée de début: 01/01/2027
Langue exigée: Anglais
Emplacement(s): ON, Canada
Nombre de postes: 2
Niveau de scolarité désiré: MaîtriseDoctorat
Ouvert aux candidatures de personnes inscrites à un établissement à l’extérieur du Canada: No

Au sujet de l’entreprise: 

Homni is a health tech company based in Niagara, building agentic software for mobility and accessibility distributors, the businesses that supply wheelchairs, walkers and home accessibility equipment.

Veuillez décrire le projet.: 

Today, getting a complex wheelchair takes an average of 162 days from evaluation to delivery, and measuring the client is one of the slowest steps. We would like to start this study in 2027 using Homni's. HomniScan technology: a smartphone app that records a short video of a seated person and produces the body measurements used to fit a wheelchair. The project is anticipated to take place in a real-world setting with a partner healthcare organization.
Technical specifications: HomniScan records a short smartphone video of the client seated, with a calibration reference for scale, since many wheelchair users cannot stand. The video is reconstructed into a 3D body model using Gaussian splatting, and HomniScan extracts the measurements used in wheelchair fitting, following WHO wheelchair service guidelines: hip width, seat depth, lower leg length, seat to bottom of rib cage, seat to bottom of shoulder blade, and seat to elbow. These are translated into wheelchair dimensions: seat width, seat depth, seat-to-floor height and back height. Because HomniScan is in active development, these specifications may change before or during the study.
The research team will plan and run testing sessions with clients and clinicians, compare HomniScan's measurements with standard manual measurements, and document how accurate, fast and easy to use the tool is in real clinical settings. Findings will go directly to our engineering team to improve each new version of the prototype during the project.
The project will deliver a testing report on accuracy and usability, a recommended capture protocol for real-world use, and a set of improvements for the next version of HomniScan. HomniScan is designed to support, not replace, clinical assessment by occupational and physical therapists.

Expertise ou compétences exigées: 

This project has two complementary positions.
Position 1, technical (computer science): studying computer science or engineering, comfortable with Python and data analysis, ideally with experience in computer vision or mobile apps. Manages the testing data, analyzes HomniScan's accuracy against manual measurements, and works closely with our engineering team on each new version of the prototype.
Position 2, clinical (rehabilitation science or kinesiology): studying rehabilitation science, kinesiology or a related health field, with experience working with people in a clinical or research setting. Familiarity with body measurement or wheelchair seating and mobility is a strong asset. Plans and runs testing sessions with clients and clinicians, and takes the manual reference measurements. Strong communication skills are important for both roles, since the team works closely with clinicians, clients and our engineering team.