NRC Canada

Project 1: AI-enabled cough detection and multimodal contactless vital signs monitoring

2020 – 2021

Project goals:

  1. AI-enabled cough diagnostics: Cough is a symptom and mechanism for COVID-19 and viral spread. This project aims to distinguish cough types (dry, wet, wheeze and whooping) with data-driven approaches on previously collected cough data. Collaborating with the NRC team to integrate the models within a system for clinical validations. The COVID-19 cough will be explored to understand if there are unique features. The team will collaborate with other research organizations to obtain relevant data.
  2. Multimodal contactless vital signs monitoring: Thermal and RGB imaging has the potential to be used for non-contact assessment of vital signs (heart rate and respiration rate) through measurement of small changes in temperature or skin colour. The team has prior experience using thermal and RGB cameras within residential settings and will apply this to public space applications. The team will compare the performance of the cameras individually and in combination using existing datasets.

PI: Rafik Goubran

Publications:

  1. Wallace, B., Yassin Kassab, L., Law, A., Goubran, R., Knoefel, F., “Contactless Remote Assessment of Heart Rate and Respiration Rate using Video Magnification”, accepted IEEE Instrumentation and Measurement Magazine, 2022.
  2. Cohen-McFarlane, M., Xi, P., Wallace, B., Valdés, J., Goubran, R. Knoefel, F., “Impact of face coverings on cough measurement characterization” 2021 IEEE International Symposium on Medical Measurements and Applications (MeMeA), 2021.
  3. Valdés, J., Xi, P., Cohen-McFarlane, M., Wallace, B., Goubran, R. Knoefel, F., “Analysis of cough sound measurements including COVID-19 positive cases: A machine learning characterization”, 2021 IEEE International Symposium on Medical Measurements and Applications (MeMeA), 2021.
  4. Yassin Kassab, L., Law, A., Wallace, B., Larivière-Chartier, J., Goubran, R. Knoefel, F., “Effects of Region of Interest Size on Heart Rate Assessment through Video Magnification”, 2021 IEEE International Symposium on Medical Measurements and Applications (MeMeA), 2021.

Project 2: Integrated Decision Support for Health and Safety

2021 – 2024

Project goals: Healthy and safe aging in place requires a continuous assessment of well-being. Automating this assessment is a cost-effective solution that reduces the burden on family members and caregivers. This project proposes deploying ambient sensors in the home and using an integrated decision support system to combine the information from these sensors and assess the health and well-being of the home occupant. The system will monitor cough, vital signs (heart rate, breathing rate) and daily activities using data analysis, deep learning and machine learning methods. The result is an integrated decision support software system that can detect any decline in health conditions, help in the early detection of diseases and enable early intervention. The project will lead to a cost-effective approach to enable aging in place by providing high-quality healthcare, reducing the burden on family members, and reducing the need for long term care, in support of the NRC AiP program.

PI: Rafik Goubran

Publications:

  1. Contactless Remote Assessment of Heart Rate and Respiration Rate using Video Magnification“, Wallace, B., Yassin Kassab, L., Law, A., Goubran, R., Knoefel, F., IEEE Instrumentation and Measurement Magazine, 2022.
  2. Yassin Kassab, A. Law, B. Wallace, J. Lariviere-Chartier, R. Goubran, and F. Knoefel, “The Effect of Noise on Contactless Heart Rate Measurement using Video Magnification,” accepted at 2022 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), 2022.
  3. M. Mozafari, A. Law, J. Green, and R. Goubran, “Respiration Rate from Thermal Camera Using Tensor Decomposition,” accepted at 2022 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), 2022.
  4. M. Mozafari, A. Law, S. Béni Tchoudem Djouaka, J. Green, and R. Goubran, “Blind Source Separation for Respiration Rate from Depth Camera,” accepted at 2022 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), 2022.
  5. M. Cohen-McFarlane, B. Wallace, P. Xi, R. Goubran, F. Knoefel, “Feasibility analysis of the fusion of pressure sensors and audio measurements for respiratory evaluations”, 2022 IEEE International Symposium on Medical Measurements and Applications (MeMeA), 2022.

Project 3: Speech and eye-movements analysis for the remote early detection of diseases associated with aging

2022 – 2025

Project goals: Early identification of cognitive decline allows for earlier intervention with appropriate care so that entry to long term care can be delayed. This project explores two non-invasive assessments (eye-tracking and speech patterns) that have potential for cognitive assessment. Specifically, the project will design a system to capture voice and eye-gaze data during tasks, with the goal of detecting speech and oculomotor abnormalities associated with cognitive decline. It will combine the two modalities and leverage multimodal analysis techniques to generate more sensitive predictions. The project will analyze combinations of tasks to achieve the assessment. The project will collect longitudinal data, to allow intra-subject analysis in addition to cross-sectional analysis to validate the performance of the system. The project will lead to a more cost-effective approach for assessment that could be done at home or remotely and that is an alternative to more expensive or invasive diagnostic testing currently being done.

PI: Frank Knoefel

Publications: