An innovative study has emerged to determine the feasibility of utilizing Apple Watch as a way to predict the pain of those with Sickle Cell Disease. Researchers from numerous institutions—including Duke University and Northwestern University—set out to conduct this investigation and recently shared the outcomes of their exploration.
Sickle Cell Disease (SCD) is a serious, genetic red blood cell disorder that can cause potential complications such as chronic anemia, stroke, and vaso-occlusive crises (VOCs). These episodes often bring about significant amounts of pain and are quite difficult to predict, making them especially troublesome to treat. As a result, the medical professionals involved with this study began searching for ways to find patterns in the data around VOCs in order to more efficiently predict patient pain.
To accomplish this, the researchers enrolled 20 patients diagnosed with Sickle Cell to wear an Apple Watch Series 3 over their visit. This Apple Watch tracked important pieces of data including heart rate, heart rate variability (calculated), and calories. The data from the device was then correlated with metrics from the electronic medical record such as pain scores and vital signs.
The collected data was then used to build out and evaluate multiple machine learning algorithms. A variety of machine learning models were put to the test in order to find the most accurate form of prediction, including multiple null models. The evaluation metrics these different algorithms were assessed by include accuracy (F1-Score), Area Under the Receiving Operating Characteristic Curve, and Root Mean Square Error (RMSE).
In the end, all models outperformed the null models, with the most successful being the Random Forest model. This model was able to predict the pain scores with an accuracy of 84.5%, as well as a RMSE of 0.84.
The results of this exploration serve to validate the potential of using data collected from an Apple Watch to detect pain levels in Sickle Cell patients. This is extremely insightful as it provides an alternative option to current methods that is low-cost and noninvasive. Should further exploration be conducted, this could prove to be an extremely impactful for patients with this disorder, helping them and their physicians to better understand the pain experience.
For those interested in reading more about the study, literature is available for review at the National Library of Medicine.
Are you or anyone you know affected by Sickle Cell Disease? Have you utilized an Apple Watch or any other mechanism to try and predict or manage pain? Let us know down in the comments!
