The technology’s infrared time-of-flight 3D camera automatically images a person’s bare feet and detects signs of peripheral edema (swelling of the feet), which can be a symptom of heart failure approximately two weeks before readmission.

Remote monitoring solutions were developed to reduce hospital admissions, as the two-week period provides sufficient time to administer medication. The company says its solution can prevent readmissions by up to 75%.

Unlike traditional monitoring methods that rely on patient input, this system operates autonomously and captures images of the foot multiple times a day without requiring the patient’s active participation. This data is sent to the cloud for processing.

CEO and Chief Scientific Officer Dr. Shams Hashir and Dr. Orian Shoshou said: engineer The device is third generation, he said, adding that the first version used a Raspberry Pi camera mounted on an aluminum bar.

“The problem with this problem is that you need a very broad baseline to deeply understand a particular uncertainty,” Dr. Hashir says. “But if a person’s feet are close to you, you need a narrow baseline, otherwise you won’t really get the data. On the other hand, time-of-flight technology, which is Microsoft Kinect’s technology for the Xbox One camera, In cameras, it increases almost linearly; noise increases with depth, rather than quadratically increasing noise with depth.”

Dr Husheer said the Microsoft Kinect was initially ignored for 3D data acquisition as it was thought to be insufficiently sensitive, but with “a little fun extra math”, software development and hardware changes, the team He added that he has created a device that achieves small feasibility. A test that demonstrated the validity of the measurement.

“With Raspberry PI, we demonstrated that optical measurements can be used to detect clinically important changes in the foot near heart failure,” said Dr. Husheer. “But it wasn’t practical to put them in people’s homes…so we had to build enough devices to put them in the homes of up to several hundred patients.”

The team also needed to obtain real patient data to use as a training set, and initially used Blender for this purpose.

“This is a 3D graphics program that simulates 3D models of people’s feet and trains neural networks,” Dr. Husheer said. ”

To date, the system has received regulatory approvals, including CE marking under the European Medical Device Regulation (MDR), and exempt status in the United States pending further clinical studies.

Dr. Chausioux and Dr. Husheer emphasized the importance of privacy considerations when implementing technology to ensure patient confidentiality is maintained while providing clinically meaningful insights. Although the current focus is on heart failure monitoring, the potential applications of this technology could be extended to other conditions such as high-risk pregnancy.

Heartfelt Technologies is one of the winners of the sixth annual #21toWatch Awards, which recognizes emerging innovators and entrepreneurs in Cambridge and the East of England.