Why apply AI to physical therapy?
In this interview, we speak with Mike Telem (Kemtai; Petah Tikva, Israel) about AI implementation within physical therapy and exercise. Mike explains that AI-guided feedback and guidance in physical therapy empowers individuals to get better faster and at a much lower cost.
Why did you create Kemtai? Is there a particular reason that you wanted to specialize in AI application in physical therapy and exercise?
Yes, we (the founders of Kemtai) all felt very strongly about creating something with real value, something that can help people. Given the computer vision background in our team (Mor and Naomi have built and exited two previous computer vision startups) and our personal history with sports, exercise and also back pain and physical therapy – we saw the opportunity to help people get better faster and at a much lower cost.
Since many people cannot afford or do not have convenient access to therapy and even those who do cannot get to a clinic more than once a week, the rehabilitation process is much longer. With Kemtai however, they can still see their PT but also exercise at home, with real time feedback and corrective guidance, as if a therapist or trainer was right there with them.
How can Kemtai be used in rehabilitation?
Most rehabilitation protocols include a substantial amount of required motion based on specific exercises per condition and rehabilitation stage. With Kemtai, the patient receives AI-guided feedback and guidance when exercising at home or at the clinic when the therapist is not there, and the rehabilitation process is faster and more efficient. The therapist or provider gains access to detailed adherence and performance reports that allow them to identify which patients are actually doing the exercise protocol and how well, as well as reviewing their progress.
What are some of the most successful stories of people that have been helped by your technology?
We work with several digital healthcare providers, and they typically embed our technology in their platform. This means that we do not have direct access to the patients, but we can see how an individual’s functional movements and range of motion improves with time. People seem to enjoy doing their PT with Kemtai since it is more interactive and helps users move correctly and notice the small improvements in their abilities.
Is there anything that surprised you when you founded Kemtai?
When we originally started Kemtai we thought it would take us a long time to get the accuracy and number of exercises required for physical therapy and rehabilitation so we started with home fitness. While that was fun too, we were approached, to our surprise, by two organizations (later to become our first two customers in healthcare) that “argued” with us that we are already good enough to provide an AI platform for physical therapy and rehabilitation. We were quite surprised but I am happy that we listened, and we are since focused on the healthcare market, and musculoskeletal conditions in particular. I think the fact that we use our own proprietary AI models allowed us to improve our models much faster than the rest of the market, add more data points on the body and support thousands of exercises.
Interviewee profile:
I have been active in the tech startup industry since 1999 and got a chance to enjoy many ups and downs in the market. I worked at both large and small tech companies and founded two startups of my own. The first one, Insightera (Tel-Aviv, Israel), was founded in 2009 and created a real time web personalization platform for B2B marketing and was acquired by California-based Marketo (USA) in 2014. I founded Kemtai (Petah Tikva, Israel) in 2019 with two good friends and cofounders, Mor Amitai and Naomi Keren and have been enjoying the ride ever since.
Disclaimers:
The opinions expressed in this feature are those of the interviewee and do not necessarily reflect the views of Future Medicine AI Hub or Future Science Group.
