Read the First Edition of The Lens, a new magazine by LSI arrow-icon

Hamed Hanafi, NovaResp - Improving Adherence through AI-Empowered Comfort | LSI Europe '24

The most comfortable experience for sleep apnea patients — powered by revolutionary AI.
Speakers
Hamed Hanafi
Hamed Hanafi
President & CEO, NovaResp AI

Hamed Hanafi 00:02
My name is Hamed Hanafi, the CEO of NovaResp, the company that is revolutionizing the treatment of obstructive sleep apnea through comfort innovation. Obstructive Sleep Apnea is the blockage of the airway that could happen hundreds of times a night. They could last periods of 10 seconds to up to two minutes. The comorbidities associated with obstructive sleep apnea include type two diabetes, higher chances of heart attacks, and higher chances of strokes. It is estimated that about 1 billion people are affected by it globally, 80% of which are still undiagnosed. Out of those that are diagnosed, 50% of them refuse to use CPAP therapy. Now, are there alternatives to CPAP therapy? Yes, there are oral appliances and pills that do not work on severe obstructive sleep apnea patients. Yet, there are neurostimulators that require minimally invasive surgery. CPAP remains the most effective therapy out there, except that 50% of patients do not meet the adherence criteria by Medicare to have reimbursement for it, and that's only using the machine for four hours a night. Now let's see why patients don't use the machines. Let's say your doctor tells you that you're going to get heart attacks and strokes if you don't use their CPAP; you're like, OK, I'm going to accept the mask, I'm going to accept the tube, and I'm going to try this. Throughout the night, you still get apneas. APAP, which is the conventional algorithm on the machine, ends up increasing the pressure after these apneas happen to hope that it's going to keep your airway open. This happens over and over again. Pressure keeps going higher, causing multiple ways of discomfort, air aphasia, nasal congestion. You have to tighten your mask so that you don't get leaks. The machine makes a lot of noise; there are leaks, the air draft that goes and hits your bed partner, that wakes them up. You end up being non-adherent and don't meet the criteria for reimbursement. Now, patients end up suffering from this because they don't get to take advantage of the benefits of CPAP. They have a lower quality of life. DMEs suffer because they're not making the sales of the machine; if the patient was adherent, they're also not making the sales of the consumables, which are reimbursed quarterly by Medicare. They also spend a lot of labor costs coaching patients, trying to convince them to use the CPAP. Same for the manufacturers, along with the pressure of emerging technologies. Now, manufacturers have two options: keep the pressures high, treat for all apneas, keeping the apnea hypopnea index (AHI) low, or reduce the pressure at the cost of having more apneas. Our software, our patented AI-enabled software, solves for both of these problems. We reduce the pressure without increasing the AHI. We keep the efficacy of treatment the same. We do so by predicting and preventing apneas with a gentle intervention before they would occur. As a result, throughout the night, the mean pressure of therapy remains low. Now, 50% of patients do not meet the criteria for Medicare coverage because they can't use the machine for more than four hours a night; at three months, this number is even lower; at five years, which is the shelf life of the machine, the number is around 15 to 20% adherence. By improving comfort and adherence, CPAP can improve patients' quality of life, increase DME and manufacturers' revenue, as well as reduce labor costs. To give you a monetary value of this technology, if you improve patient adherence by just 10%, that's equivalent to $700 million a year more revenue for manufacturers. So our product is pure software that would sit on the processor of the CPAP machine, any CPAP machine in the market. All it does is read from the pressure and flow sensors of the machine and tells the air blower what to do. We expect to commercialize by 2026, and on top of that, we have the possibility of adding personalization to therapy, endotyping, and phenotyping to specific patient populations, or personalization to the individual through using the digital ecosystem of the manufacturer that we'll be working with. First, in our clinical path, we first proved that we can predict apneas. Then we moved forward and said, "Okay, can we use predictions to prevent apneas?" Then we spent 2023 perfecting the algorithm, and we spent 2024 with the results that I'm about to present to you, running two clinical trials to prove the efficacy of our algorithm. The first one is called the Comfort Study. It's a randomized crossover trial with 50 patients already adherent; they're used to their CPAP. They start on this machine, either on their own home algorithm, APAP algorithm, or on our algorithm, CPAP, half of them, and then they switch over. Halfway through, they fill out questionnaires to test out where they're comfortable in each arm, and they are wearing a sleep image ring that would be scoring their sleep staging through cardiopulmonary coupling. Results were amazing. This is an unknown population to our algorithm, where we're reducing the pressure by 20% without compromising the therapy; we keep the efficacy of therapy the same. Now these are already adherent patients; they're used to their CPAP. We did not expect to improve their comfort or sleepiness. We did improve their comfort and sleepiness in a statistically significant way. We also improved their sleep quality index, scored by the sleep image ring. We improved their stable sleep, significantly reduced their unstable sleep, improved their REM sleep, and reduced fragmentation. Most importantly, we reduced the 95th percentile leak. That's the type of leak that your mask is whistling, waking you and your partner up. We reduced that significantly. We received a lot of nice comments from the patients, but most importantly, their partners went out of their way to tell us that they could sleep better because there was less noise and hissing coming from the machine and less blow to their face. The second study is an adherence study, a 200-patient study where we're comparing our algorithm versus the conventional algorithm to see how much adherence we're actually improving. Because we're in the middle of it, I can only report on our observations, where we're significantly improving patient adherence. We're increasing the nightly usage of the machine. We're lowering the labor cost; less coaching is needed for the patients. Patients are acclimating to therapy faster. They meet that four-hour criteria earlier in the trial period, and there is much less leak. Now our patents are issued in public in multiple geographies and pending in some others. We have a favorable FTO as well as trade secrets. Our board of directors and Medical Advisory Board is filled with superstars. Dr. Neil Smith is the founder of one of North America's largest privately owned DMEs and an Ear, Nose, and Throat surgeon. Mr. Raj Sodi is a global leader in digital health and the former President of Software as a Service at ResMed. Dr. Ingo Fieze is the head of sleep at Sheraton Institute of Berlin. Dr. Mehra is the same at the University of Washington. We address patient adherence and patient comfort without compromising therapy efficacy. This is a large and undeniably growing market. We've proven that our algorithm works much better than the conventional algorithm, and we've protected our IP. Thank you very much.

LSI USA ‘25 is filling fast. Secure your spot today to join Medtech and Healthtech leaders.

March 17-21, 2025 Waldorf Astoria, Monarch Beach | Dana Point, CA Register arrow