Authors

Inspiration

People are living in fear amid the COVID-19 outbreak. Some cannot afford the protective gear against the virus. Some feel unsafe because they cannot see who around them has a high risk of being a transmitter of the virus. Some do not have the right knowledge about what they should do during this pandemic.

The use of masks and thermometers have been acknowledged as forefront measures for protection and early detection of the coronavirus. Wouldn’t it be better if we have a solution which combines the benefits of both?

Our team, COVID-19 Early Detection Intelligence from Thailand, proposes an intelligent system consisting of a body-temperature-sensitive facial mask and a simple but smart monitoring platform, which together can create a visual map of low-risk vs. high-risk citizens. This information can be utilized to help make high-level decisions, such as zoning and resource allocation, to make the most effective and timely moves in response to this rapid widespread of COVID-19.

What it does

The first part of this “COVID-19 Early Detection Intelligence” project is the mask itself.

The mask will be made by a special paper or a paper with the thermochromic substances that can change the color when the temperature change. It can be a three-layer mask, where the outer layer is waterproof to protect any sprayed liquids; the middle layer is protected against a smaller particles; and the inner later is food-graded paper with thermochromic substances, which will not be harmful to the skin or health. For people with high-fewer wearing this mask, when they speak, cough, or sneeze, the mask can detect the humidity with high temperature above the threshold level and automatically change the color.

The second part of this “Corona Masks Detector” project is on the AI systems to detect the changed color.

The Computer Vision with artificial intelligence (CV–AI) through CCTV or Video Camera can detect the color of the mask. With the precision of the CV-AI nowadays, the error rate in capturing the pattern of the changed color is relatively low. When the CV-AI detects those patients with high temperature (the color of mask is changed), the real-time analytics can be run; for instance, 35% of people wearing this mask in this Zone “A” have a relatively high temperature, which signal that they are a very high risk group. Such information can be used for government to make better decision.

How I built it

Our mask contains a thermochromic paper whose color changes according to the temperature of the liquid it touches. When the wearer having a high body temperature coughs, sneezes, or just speaks, the color of the mask will automatically change from light brown to dark brown. The outer layer of the mask is water-repellent, giving protection to the wearer, similar to a medical grade surgical mask. Our mask is affordable and easy to wear, making it available and accessible to everyone. The color of the mask is easily detectable and interpreted by a monitoring system, built on a computer vision with artificial intelligence camera. The system can generate a visual map of individuals with normal vs. high body temperature, providing necessary guidelines for medical staff, policy makers, and law enforcers.

Our solution addresses the safety issues as our masks are made with food-grade materials and our monitoring system serves as an accurate, fast, data-rich, non-contact method for early detection of the COVID-19 symptom.

The big investment is for the city or government to invest on the computer vision AI (CV-AI) to detect the color of the mask including the database management, real-time analytics, and IoT.

Challenges I ran into

  1. How can we make people wear this mask. For this solution to be effective, everyone in the community need to use this mask.
  2. This mask must be cost effective. The cost of the mask should not cost more than 1.7 cent per piece.
  3. We are not sure how temperature or humidity might affect the level of mask’s sensitivity.
  4. The culture of the country might affect how people adopt this mask. Some people might view this mask to be intrusive

Accomplishments that I'm proud of

Our idea is to turn hidden information such as the symptoms of the coronavirus to be visible through our mask. Such valuable information can help the city to perform zoning activities (Alert Zone, Early Detection Zone), which can signal the city mayor about resource allocation, decision making involved the city, business, or school closure. The hospital can gain the benefits from such information to better utilize the staffs’ capacity, prepare for staff scheduling due to the alert from the government.

Another point is that people is afraid of reality telling the truth about being infected, for instance. Thus, our system can provide a higher level of signals rather than attacking the privacy of each person. The information should be relatively enough for the city to setup an emergency policy or early detect and signals about livelihood or virus incidents. Additionally, our COVID-19 early detection intelligence can potentially save at least 8 billion dollars per million population from cutting down on the costs of late detection and treatment at a 50% reduced infection rate.

What I learned

We learn to solve the problem as a team. Our team consists of experts from both management and technical fields. In addition, we are a group of design thinkers from Thailand. We apply design thinking approach to solve this problem. We hope that our solution will solve this crisis.

What's next for Corona Mask Detector

According to the National Coronavirus Response, American Enterprise Institute, the COVID-19 pandemic can be classified into 4 phases, i.e. “Slow the spread”, “State-by-State Reopening”, “Establish Immune Protection and Lift Physical Distancing During Phase”, and “Rebuild Our Readiness for the Next Pandemic” respectively.

Our solution could effectively perform as a surveillance system that can identify those who have a very chance of Covid-19 infection. Consequently, it can signal which districts or cities are in the red or green zone. So, it can help government to accurately create zoning for better managing their hospital facilities and businesses. That is the benefit for Phase I that every country around the world are faced currently except China. Our solution might help all of them to anticipate COVID-19 and rapidly move toward to Phase II: State-by-State Reopening with sufficient critical-care capacities.

Furthermore, it can be scalable to use around countries to monitor new cases and prevent from going back to Phase I again. Our solution could be a part of the national policy to maintain the pandemic situation at the Phase II as long as possible before the vaccine of COVID-19 will be created, i.e. Phase III.

Finally, to put our plan into work, we will be partnering with Computer Vision providers, thermochromic paper suppliers, and AI providers.

Hackathons

Technologies

computer-vision-ai, cv-ai, mask

Devpost Software Identifier

259922