2021 is finally behind us and our team took some time to look back and reflect on what was accomplished by our community of researchers & developers. Last year was also a year of tremendous growth for our Hexoskin and Astroskin solutions with our professional clients. We'd like to present you with a few publications and case studies that got our attention as we look forward to a new chapter for Hexoskin in 2022.
In this study, Hexoskin smart garments were worn by 16 firefighters to track their psychophysiological responses, specifically HR and HRV, during rescue simulations performed under various conditions (day vs night, and haptic vs audio alarm) at the University of Bourgogne-Franche-Comte. Their results showed that night rescue interventions resulted in significantly higher psychophysiological responses and lower self-confidence than day rescue interventions, and the type of alarm had little effect on psychophysiological responses.
In this study, the researchers at the University of Salzburg evaluated Hexoskin’s RIP sensors with a custom algorithm versus a reference spirometry system to determine the concurrent validity in detecting flow reversals (FR) and breathing pattern (BP). Their algorithm successfully determined 99% of FR, suggesting that the proposed system is valid and practically useful for BP assessments in the field, specifically during exercise and running.
In this study, the team at the University of Alberta recruited six people with neuromuscular conditions. The participants wore the Hexoskin smart shirts and engaged in a 15-week recreational singing and dancing program to evaluate its safety and meaningfulness on quality of life. Overall, they showed improvements in respiratory function along with speech, swallowing, strength, leisure, and relationships, demonstrating that recreational singing and dancing is a safe activity to improve physical and social quality of life for people with neuromuscular disorders.
Human activity recognition (HAR) plays an important role in remote health monitoring and emergency notification. This study is a collaboration between teams from research institutions based in Montreal (Canada) and Cartage (Tunisia) which evaluates a method for classifying acceleration data using an efficient classifier combination of feature engineering-based and learning-based representation. The result yields a 90% recognition rate and performs significantly better than individual classifiers.
This team from Peter L. Reichertz at the Institute for Medical Informatics, is working on a lightweight IoT platform, where ECG metrics from Hexoskin were sent to an edge device and analyzed using deep learning models for binary end-to-end classification.
The performance on 5G showed an average transmission latency of 110ms, data corruption in 0.07% of samples, and deep learning inference of about 170ms, suggesting that 5G networks with edge devices provide a suitable infrastructure for continuous remote patient monitoring.
Choosing is hard and there are many more publications and use cases that deserve attention. That’s why our Publication page provides a full list of the publications from our community.
The entire Hexoskin Team would like to congratulate our Hexoskin community that continues to push the boundaries of science and innovation despite the challenges we face due to the global pandemic.
Looking forward to the upcoming months, we are preparing a slate of events aimed to shine a light on some of the groundbreaking research and projects performed with Hexoskin & Astroskin. Make sure to subscribe to our Hexoskin Newsletter to receive our news and events!