Foresight offers a smart protective eye wear with an integrated software platform to monitor the health of healthcare workers in real time to assess their fatigue level.
In the light of COVID-19, we are inspired by the healthcare workers who are not only facing a daunting task of saving lives but are also fighting their own emotional stress. When we imagined ourselves in their shoes after looking at those videos on the internet where they expressed their daily battles, it burdened us. Being professionals with experience in the MedTech field, we thought about how we could leverage our skills to help them. While researching we found out that there are many programs and services set up to support and sustain their mental health, but there was a gap to address their physical well-being. Hence, started with our research on monitoring their fatigue, and here we are today with our first prototype.
Why fatigue & vital signs monitoring?
Once suited up in protective gear, healthcare workers are serving without food, water or even bathroom breaks for 6 to 8 hours which could weaken their immune system. On the other hand, the current crisis demands their swift minds to make quicker decisions, be innovative and continue doing what they do. A physiological impact study conducted among a cohort of 549 healthcare workers who served in Beijing, China during the 2003 SARS outbreak showed that about 40% suffer from post-traumatic stress even after 3 years. Today, we have millions of tireless healthcare workers working towards this crisis. While vital signs indicate their physiological health, fatigue monitoring can indicate their mental load.
What Foresight PPE does?
Foresight PPE is integrated with sensors that can continuously monitor eye blinks, temperature (Version 1.0) and respiration rate (Version 2.0). Real-time data is sent to the platform for further processing. The platform makes use of a machine learning algorithm that is capable of tracking the eye blinks and building a fatigue index. The core of our proposed method is a Hierarchical Multiscale Long Short-Term Memory (HMLSTM) network, that is fed by detected blink features in sequence. Based on the fatigue levels, the system alerts the user and this allows the user to follow up with their health status.
How we built it?
Before participating in this hackathon, we gathered all the relevant information from the scholarly articles and worked on framing the initial solution. During this hackathon, we focused on developing and evaluating different algorithms in Machine learning as a first step to process the user data based on discussions from Ghoddoosian, R., Galib, M., & Athitsos, V. (2019). A Realistic Dataset and Baseline Temporal Model for Early Drowsiness Detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (pp. 0-0). We also focused on the design of our eye wear and getting it 3D printed to demonstrate our prototype.
Challenges we ran into
One of the main challenges was the lack of dataset. So, we had to collect sufficient data within the past 3 days to train and test the Machine learning algorithm. The second challenge was to identify and select a suitable material for our eyewear which is both eco-friendly and organic.
Accomplishments that we’re proud of
We are a good combination of experience and skills across Life Sciences, Information Technology and Business domains along with a good core synergy of thoughts while respecting our complimentary background and perspectives. We are also related to each other and share a natural bond outside work which further facilitates quick turn around of our ideas into tangible results. With the above traits and the confidence from our previous successes, we believe we will make foresight, a reality.
What we learned
Our team has improved in size, knowledge and experience for further development of the project. Thousands of people around the world are interested to find solutions to deal with our current situations!.
What's next for FORESIGHT
*Marketing plan to get more data samples by crowdfunding campaigns, and working in close collaboration with experts from medical field to test and validate the efficiency of the developed algorithm.
*Fundraise to make the team 100% operational. It’s a profitable project that could be attractive to investors and investment funds from the healthcare sector and other fields. Only together we are strong to address this COVID-19 pandemics and future threats!