New research from the Icahn School of Medicine at Mount Sinai reveals that wearable devices, like the Apple Watch, can provide valuable data for training machine learning models to assess patient well-being and resilience. By analyzing heart rate variability and resting heart rate, these devices offer insights into mental health and stress management. The findings highlight the potential of wearable technology in expanding access to psychological assessment and care, improving treatment outcomes, and predicting disease progression.
The Icahn School of Medicine at Mount Sinai in New York has conducted new research indicating that data collected from Apple Watches, such as heart rate variability and resting heart rate, can be utilized to train machine learning models for assessing patient well-being and resilience.
According to the Centers for Disease Control and Prevention (CDC), mental illness affects more than 20% of adults in the United States. Mental health diagnoses are among the most prevalent health conditions in the country, highlighting the need for effective tools and technologies to support patients.
This recent study demonstrates the potential of wearable devices to assist individuals with mental health diagnoses by gathering relevant data. The researchers emphasize the strong correlation between resilience and stress reduction, morbidity management, and disease control. While wearable devices are not designed explicitly for this purpose, the passive collection of data through these devices can offer valuable insights when training machine learning algorithms.
To arrive at their conclusions, the researchers analyzed data from a previous study that utilized wearable devices for monitoring patient health. In this study, 329 healthcare workers wore Apple Watch Series 4 or 5 devices capable of capturing heart rate variability and resting heart rate.
By combining this data with baseline survey results, the researchers established a link between the collected information and measures of resilience and well-being.
The positive results obtained from this study lay the groundwork for future applications of wearable devices in assessing both physical and psychological health. The researchers believe that the combination of these capabilities with advancements in machine learning could play a vital role in treating various conditions.
Micol Zweig, MPH, co-author of the paper and Associate Director of Clinical Research at the Hasso Plattner Institute for Digital Health at Mount Sinai, expressed optimism about the potential of this approach to expand access to psychological assessment and care. Zweig also stated that further evaluation of this technique in other patient populations is essential to refine the algorithm and enhance its applicability.
Previous instances have shown that the integration of wearable device data and machine learning capabilities can aid in predicting disease outcomes.
In a separate study published in JMIR Formative Research in March, researchers described how the use of an Apple Watch enabled the prediction of pain scores in hospitalized patients with sickle cell disease. They further demonstrated how this data could be leveraged to develop machine learning algorithms for predicting pain scores during vaso-occlusive crises (VOCs).
The study involved sickle cell disease patients admitted to Duke University SCD Day Hospital for a VOC. These patients were provided with Apple Watch Series 3 devices, which they wore throughout their hospital stay. The devices collected continuous heart rate, heart rate variability, and caloric information. Three machine learning models were then trained using half of the Apple Watch data along with vital sign and pain score data from patient electronic medical records (EMRs).
The researchers concluded that the machine learning models accurately predicted pain scores, validating the feasibility of using wearable device data for predicting pain scores during VOCs.
These studies highlight the potential of wearable devices, such as the Apple Watch, in collecting relevant health data and utilizing it to train machine learning models for various applications, ranging from assessing patient well-being and resilience to predicting disease outcomes. The integration of these technologies has the potential to revolutionize healthcare delivery and improve patient outcomes.
