According to the Parkinson Disease Foundation, more than 60,000 Americans are diagnosed with Parkinson’s disease every year. With one million patients in the United States, Parkinson’s disease is one the most prevalent neurodegenerative disease.
Even if some cases are hereditary, researchers think that the disease is mainly linked to exposure to environmental factors still to be identified.
Among a wide range of motor and psychological symptoms, patients who have Parkinson’s Disease will experience uncontrollable tremors of the hands, legs, or jaw, impaired balance and coordination, and stiffness of the limbs.
Parkinson’s patients will be initially treated with Levopoda (L-Dopa) used to increase the decreasing levels of Dopamine in the brain and control dyskinesia and tremor.
Digital health technologies’ benefits to detect dyskinesia have been a subject of research for numerous years. Electromyography, electromagnetic motion detectors, Kinect, or even laser Doppler vibrometers proved helpful but lacked portability.
Wearables, especially smartwatches such as the Apple Watch, now have the sensors and processing power needed to detect blood pressure, atrial fibrillation, and blood oxygen saturation. However, their potential Parkinson’s disease symptoms still need to be assessed in robust clinical trials.
This article will go through the latest research using smartwatch inertial sensors to assess tremor symptoms and medication efficacy.
Quantifying tremors with consumer products built-in accelerometers
A clinical study conducted in the Netherlands in 2017 was published in The Journal of Clinical Disorders.
In their pilot study, the team investigated the possibility of using smartwatches or smartphone built-in accelerometers to measure the tremor severity in Parkinson Diseases’s patients.
In patients who have either Parkinson’s Disease or Essential Tremors, they compared the technical capabilities of smartphone or smartwatch inertial sensors to record dyskinesia.
Seven Consumers Products accelerometers (Apple iPhone 7, Apple iPod Touch 5, Apple Watch 2, Huawei Nexus 6P, Huawei watch, mbientlabMetaWear watch, and mbientlab MW clip) were compared to a Laboratory Grade Accelerometer (Biometrics ACL300).
The results were shocking. At least for Laboratory Equipment makers, the consumers’ products performed comparably to their more expensive and specialized counterparts.
This study proved the potential of consumer products for either clinical trials or measuring the tremor patterns or symptom changes at home over a long period.
As for most neurological disorders, Parkinson’s patient treatment will often have to be adopted over time. The Apple Watch Parkinson’s tracking capabilities are especially interesting for measuring motor symptoms evolution.
Detecting Early Parkinson’s symptoms using consumers wearable devices
Resting tremor is one of the characteristics signs of Parkinson’s Disease. However, researchers in Portugal and Spain recently showed that combining the sensing abilities of a smartwatch with advanced deep learning algorithms, Multitask Convolutional Neural Networks, can detect the onset of the disease.
One of the main benefits of the underlying neural network is distinguishing between resting tremor and action tremor.
The current diagnostic and therapeutic protocol involve a consultation with a neurologist and a brief assessment of the types of involuntary movements.
Doctors also rely on the patient’s assessment of the evolution of its movement disorders. But, unfortunately, Parkinson’s patients can also suffer from gradual cognitive declines, making their judgment of their symptoms doubtful at best and unreliable at worse.
The benefits of the Apple Watch and other wearable devices are that the smartwatch sensors, including gyroscopes and accelerometers, are now reliable enough to allow acquiring clinically relevant data.
The Apple Watch for continuous monitoring of PD patient’s motor fluctuations
In collaboration with researchers from the Harvard Medical School, an Apple team analyzed the benefits of the Motor fluctuations Monitor for Parkinson’s Disease (MM4PD), a system designed to make the best of the Apple Watch inertial sensors to monitor PD symptoms and hand tremors.
The clinical study results they conducted on 343 subjects were recently published in Science Translational Medicine Journal.
The study aimed to evaluate the potential benefits of the Apple Watch as a clinical tool to improve the granularity of the tremor severity to assess the disease progression better and ultimately improve treatment.
Apple researchers developed MM4PD to detect dyskinesia and Parkinson-related tremors continuously.
The algorithm development was based on the data collected during three preliminary studies, including 343 patients with Parkinson’s and 171 healthy individuals as controls.
The Longitudinal patient study lasted six months, and the data was rated by a movement disorder specialist familiar with the patient’s history. The system is aligned with the standardized MDS-UPDRS assessment scale for easing the translation between the findings and the clinical outcomes.
The performance of the Apple Watch coupled to the MM4PD algorithm significantly detected dyskinesia and tremor differences in patients and control subjects.
Before being implemented in the real world, the MM4PD still requires some development though. A smartwatch is, per definition, wore at the wrist, which in turn limits the number of observation points and will not consider tremors or dyskinesia occurring in other limbs.
Moreover, the combination of an Apple Watch with MM4PD appeared has highly correlated with clinical symptoms and sensitive enough to detect the effect of a change in treatment.
Future improvements will relate to incorporating other motor symptoms such as postural tremor, dyskinesia severity, and periods of simultaneous tremor and dyskinesias.
To wrap up
Using a smartwatch to complement clinical exams and evaluate the benefits of a change in therapy for PD patients is undoubtedly a promising approach that proved valuable in numerous studies.
Smartwatch sensors are now sensitive enough to detect the tremors caused by a neurological disease. However, the key to meaningful clinical use is to widen the scope of analysis and not limit the analyses to the wrist only.
The Apple Watch is considered the best health smartwatch available for a reason. The company has already proven its commitment to developing an FDA-cleared algorithm with the potential to change how patients can manage chronic diseases.
The MM4PD is very promising but do not expect to find it anytime soon. Clinical trials and FDA clearances take time, and we can only hope that the Cupertino company will continue its efforts in a direction that may ultimately improve the lives of millions of patients affected by Parkinson’s diseases.