How wearables and human-augmented machine learning can eliminate preventable hospitalization

Dean Sawyer


A 55-year-old man with a history of heart disease is driving to work on a Monday morning when his mobile phone rings. It’s his care manager with a message: A pattern from the continuous streams of data from his wearable biometric device indicates that he is in the early stages of heart failure decompensation and is likely to be hospitalized in 14 days without an intervention. The care manager goes on to say not to worry - his cardiologist increased the dose of his diuretic in an intervention that has a 94% chance of stopping the decompensation within 48 hours. 

This is a fictitious example of “remote patient intelligence,” but the breakthrough technology for collecting and analyzing actionable medical data – for individuals as well as whole populations – is finally here. It represents our best hope for gaining control over a mass pandemic of chronic illnesses, from heart disease to diabetes.

In a presentation titled “How wearables and human-augmented machine learning can eliminate preventable hospitalization,” Sentrian CEO Dean Sawyer will explain the current barriers to collecting and applying useful patient data from wearable devices and physician-prescribed “remote patient monitoring” devices, and then reveal how advances in machine learning and Big Data analytics can detect health deterioration days or weeks in advance so less invasive, less costly interventions can be implemented.

The topic is extremely timely. Apple’s HealthKit is currently being tested by a dozen of the nation’s largest health systems that would like to track real-time health indicators from patients who use wearables. However, doctors don’t trust the data from consumer wearable devices, most of which were designed as informal fitness trackers. Moreover, HealthKit and similar initiatives are likely to overwhelm clinicians with data that cannot be integrated into their everyday practices.

By contrast, as Sawyer will explain, “remote patient intelligence” leverages wearable biometric devices, cloud-based Big Data analytics, machine learning and natural language processing ,to enable care teams to make sense of vast streams of data from remote medical devices, and monitor thousands of patients at once. New, cloud-based analytics engines can compare the remote sensor data with a patient’s medical history, co-morbidities, exacerbations, and other benchmarks (for example, two-day average blood pressure readings compared to the last months’ average), and look for contrary indications.

The data is also analyzed using disease models (algorithms) that physicians can for the first time create in natural language without the assistance of data scientists or programmers. Machine learning can then optimize those rules and look for new patterns not obvious to clinicians. With remote patient intelligence, the disease management process can at last be affordably scaled to the population level for real-time population health management.

This technology is currently being tested in four large scientific trials with thousands of patients, including one of the largest health insurance companies in the U.S.

We believe the topic of remote patient intelligence will be of deep interest to academics, clinicians, and patient advocates alike. Dean Sawyer is an experienced and sought-after speaker known for his ability to translate complex technology and business models into a format everyone interested in the future of medicine can understand. Please see Dean’s bio below.  

Contact Us

We're not around right now. But you can send us an email and we'll get back to you, asap.

Not readable? Change text. captcha txt

Start typing and press Enter to search