The combined impact of respiratory illness such as COPD, asthma, chronic cough and pneumonia on individuals and health systems is staggering. COPD alone affects more than 15 million people and costs the healthcare system close to $50 billion each year.

Effective daily maintenance therapy enables many people who suffer from these chronic conditions to live healthy and productive lives, but exacerbations are inevitable and if not promptly recognized and treated can be fatal.

And, even when not fatal, poor monitoring leads to increased ER visits and hospital stays. For example, COPD results in more than two million ER and hospital visits per year, which are costly in both monetary and quality-of-life terms.

Some health IT experts believe remote patient monitoring and artificial intelligence have a role to play here. That RPM and AI can help reduce costs and increase quality of life. We interviewed Dr. Peter Small, chief medical officer at Hyfe AI, a vendor of AI algorithms that arm remote patient monitoring technologies, including for cough tracking, to talk about these very issues.

Q. What are some of the challenges that healthcare today faces with respiratory health where remote patient monitoring could prove helpful?

A. Early recognition of exacerbations is complicated by the stochastic nature of their symptoms. One example? Cough – a protean manifestation of many of these diseases that waxes and wanes over good days and bad.

Furthermore, patients are insensitive to changes in cough and often unable to recognize worsening symptoms until they are severe and require a visit to an emergency room or intensified therapy.

I see the greatest value of remote patient monitoring as the ability to passively, continuously and unobtrusively monitor their symptoms so they can better manage their disease. For example, it has been shown that monitoring cough can detect COPD exacerbations more than three days in advance of when patients seek care – enabling early intervention to avoid that exacerbation.

Q. Please explain what acoustic AI is and how it can be used in remote patient monitoring.

A. Large language models like ChatGPT have garnered a lot of attention recently. But the use of deep learning models in health is not new. These models are intensely data hungry and thus it is not surprising that the first “language” was data-rich images. There now are well-established tools to assist healthcare providers to interpret images in the fields of dermatology, ophthalmology and radiology.

The next frontier in AI in health is acoustic AI, which is simply the application of these computer methods to sound. When trained on sufficiently large and diverse collections of annotated sounds, these approaches can recognize and classify sounds such as a heart murmur, a wheeze or a cough.

And, when incorporated into digital stethoscopes, they can improve the diagnostic capabilities of clinicians. By tracking their frequency, they can assist with medical management decisions. And ultimately, when incorporated in consumer products, they will empower patients to have a better understanding of and be better able to cope with their chronic medical conditions.

Q. What are the expected patient outcomes with this kind of RPM for people with everything from simple bad coughs to chronic diseases like asthma or COPD?

A. Cough is one of the most common problems for which people seek medical care, accounting for nearly one in five reasons for visiting a healthcare provider. However, despite living in the era of “precision health,” cough is simply not measured, which dramatically complicates its diagnosis and treatment.

Most coughing is caused by transient self-limited diseases such as upper respiratory infections. However, in many people, cough persists for months or years and can be extremely problematic. Particularly since COVID, coughers are stigmatized and often wary of going out in public, attending movies, religious services and other everyday activities.

Chronic cough is associated with urinary incontinence, broken ribs, and has been linked to feelings of social isolation and depression.

Fortunately, there have been significant advances in the science of cough in the last decade. Chronic cough is now understood to not only be a symptom of common conditions, like reflux, asthma and post-nasal drip, but also a disease unto itself caused by neural hypersensitivity.

Understanding of the neural pathways of cough has sparked a pharmaceutical race to develop effective antitussives. It’s likely the FDA will approve a new cough drug by the end of this year – the first in more than 60 years.

Objective, passive, automated and privacy-preserving cough monitoring will be a game changer for patients with chronic cough.

It will empower them to better understand their disease and communicate its severity to their doctors. It will enable providers to better diagnose, select effective therapies and monitor the course of treatment. And it will help health systems to intervene promptly and avoid expensive and dangerous exacerbations.

Q. Looking ahead, what is the potential for RPM in respiratory health and where do you see this space heading?

A. We are in the midst of a revolution in medicine based on empowering patients to monitor their own health with devices, such as home spirometers and sphygmomanometers. The ability to monitor signals continuously, like blood glucose, is becoming the standard of quality care.

As someone with a chronic cough (I have coughed 40-80 times a day for more than three decades), I’m particularly excited about how continuous cough monitoring can empower me to have more informed conversations with my doctor and to feel more in control – to actually be more in control – of my health and healthcare.

As a tuberculosis expert who spent about the same number of years helping coughers get tested for TB, I’m excited by how better monitoring of cough will help doctors and health systems improve the care of a dozen or so serious medical conditions.

I predict within a few years cough-related conditions will be managed like hypertension – if you have the disease, you should have a device in your home for measuring it. FDA-cleared cough monitoring devices will become commercially available and routinely used by those with cough-related conditions.

And soon thereafter, cough monitoring will become like step counters that are ubiquitous on consumer devices and folks only pay attention if they care. I look forward to the day when counting coughs is as easy as counting steps and as important as continuous glucose monitoring.