Using IoT for Triage and Monitoring: a Collaboration with NYU Langone Health with Wider Applications

African-Americans and Hispanics with untreated moderate or severe sleep apnea are twice as likely to have hard-to-control high blood pressure.^[1][2] That may seem like a highly specific medical situation, but in at-risk and underserved populations, the potential for heart disease, hypertension, and diabetes is of great concern. That is something Azizi Seixas, Ph.D., an Assistant Professor in the Departments of Population Health and Psychiatry at NYU Langone Health, is now addressing.

Underserved populations have a more difficult time accessing healthcare in general and are much less likely to be enrolled in sleep studies. In the best of circumstances, barriers are high, since in-patient sleep studies are expensive and beds are scarce. Yet diagnosing and treating sleep apnea is crucial to managing medical risks like hypertension, heart disease, memory and mood issues. Without diagnosis and treatment, high risk and underserved populations can face these additional medical problems as a result, decreasing quality of life and increasing medical costs.

The condition is thought to be underdiagnosed, partly due to lack of knowledge, and partly because diagnosis is expensive.^[3] To address both factors, Dr. Seixas’ study uses wearable internet of technology (IoT) devices and a cloud-based platform to collect data points for assessing, treating and monitoring in a primarily non-white Hispanic population. “We want to build a decision aid for patients and providers, that can help assess risk for sleep apnea, and triage who should be seen, referred and treated,” he said.

Using Technology for Triage and Adherence

Maven Wave is working with NYU Langone on the project to create a technology framework and workflow so a primary care physician or nurse can gather enough information to determine apnea risk and triage appropriately. The progressive web application will capture sleep tracking data, patient activity levels, caloric intake, BMI, and other health inputs.

If sleep apnea issues present with monitoring, the patient can be referred for advanced diagnostic care. Sleep studies are sometimes done at home, which is easier for patients and less expensive as well. Treating sleep apnea at an earlier stage can potentially improve health outcomes and reduce treatment costs in the long run.

In a related study, Dr. Seixas will use the Maven Wave technology architecture to try increasing treatment adherence for those already diagnosed with sleep apnea. The first two weeks of initial treatment are critical in optimizing adherence. The study will promote adherence by monitoring the patient with wearables, and sending push-based patient reminders about treatment follow-through. “We want to provide health-related and actionable messages to patients to increase their adherence and optimize their behaviors,” Dr. Seixas said.

Designing Data Architecture

Dr. Seixas and his team could not conduct their sleep apnea studies without customized technology. Maven Wave is using Google Cloud Composer for workflow orchestration. Raw FitBit biometric data are pushed to Google Cloud Storage (GCS) and correlated and warehoused with patient profile data and historical data. Data Studio and Datalab are used by the NYU Langone research team for data analysis where data can be correlated with population and geographic datasets such as census data and Google Places API data. Patients are able to visualize progress and receive personalized messaging to encourage adherence through a web-based health and wellness portal.

The data are available in real-time, so clinicians and patients have immediate access. To realize the advantages of real-time architecture, a patient mobile application will be developed to create a truly seamless real-time experience.

FitBit and Bitesnap offer individual dashboard interfaces; however, usefulness for clinicians are only fully realized when FitBit data are integrated with disparate sources across the cloud, for a holistic view, laying the groundwork for data science efforts.

Using an outside technology partner allows scientists to focus on the data analysis and patient aspects of the studies.

Scaling Up the Technology

Once the architecture is fully developed and tested with the initial study population, it can easily be scaled up for wider applications. Sleep apnea is an interesting use case for integrating wearables and data from multiple sources and presenting it in an easy-to-access format. The concept and technology can be used in situations that don’t require in-patient monitoring, opening up possibilities to serve rural communities, patients with limited mobility, and even those in developing countries. Wearables provide continuous and reliable biometric data, which is often more reliable than the subjective information patients can provide in surveys.

Dr. Seixas noted that the technology also allows physicians to work more efficiently. They can provide specialized care to patients who may not otherwise get it, especially in underserved populations. Physicians can follow more patients remotely, providing automated health recommendations based on actual data. “It’s about how health systems can assist and partner with patients in obtaining and achieving health and wellness,” he said. It’s a more nuanced way of building partnerships and relationships. “These are strategies that can stem the significant burden of chronic diseases. That’s what Maven Wave is allowing us to do.”

Interested in learning more? Reach out to Maven Wave today to find out how we can build a solution to meet your healthcare organization’s needs.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4144432/
[2] https://www.ncbi.nlm.nih.gov/pubmed/24392863
[3] https://www.nhlbi.nih.gov/health-topics/sleep-studies