During the ongoing Covid-19 pandemic, my partner and I were tasked with creating an accurate temperature detection machine with the support of Intel Corporation, where I serve as a technical ambassador. Despite the prevalence of similar devices on the market, they were found to be unreliable in measuring febrile illnesses. We sought to create a device that was grounded in science and could accurately take measurements. We aimed to address the limitations of existing devices such as cheap hardware and unreliable software by using a scientifically valid method and location for temperature measurement.
Our temperature detection machine was designed with accuracy and ease of use in mind. We partnered with a medical cart manufacturer to create a portable device that can be placed anywhere and measures subjects of various heights without the need for repositioning. The machine features a camera gantry on an automated linear slide rail that is controlled by an operator’s screen to match the height of the subject. The device uses an Intel RealSense depth sensing camera and a high-end thermal camera to map the subject’s face and take a reading from the ocular canthus, a reliable location for measuring core body temperature. To ensure accuracy, the device employs the use of a “black body” reference point for the thermal camera. The device is easy to use, with a screen that instructs the subject to step forward or backwards and remove their glasses if necessary.
The machine was designed to automatically detect the subject’s temperature and did not require them to remove their mask. It would scan the subject’s canthus for 1-2 seconds and display the temperature on the operator’s side of the machine. If the temperature was normal, the subject was permitted to proceed. If an elevated temperature was detected, they were redirected to a secondary screening area designated by the machine’s owner. The machine also had the capability of sending data to an analytical platform for remote statistical analysis.
We successfully developed and deployed 10 prototypes of our machine, which were constructed in my garage and barn and personally delivered to various schools and warehouses. At the start of the project, the market was saturated with overpriced solutions that claimed to deliver the same results as our machine but failed to do so. However, when the CDC released a report revealing the lack of reliability in these competing solutions, the market collapsed. Machines that were previously sold for $20-40K were now being sold for less than $5K, and some even for a couple hundred dollars.
Given this situation, we decided to pivot and explore new opportunities. Our solution is currently being evaluated for use in the next generation of smart buildings, which we believe could be a promising market for us in the future. The key takeaway from this experience is that instead of questioning what the competition is not doing, it’s important to question if they are doing what they claim to do reliably.