29-11-2021 | By Robin Mitchell
Recently, researchers have combined low-cost wearable technology with AI to help Parkinson’s patients deal with freezing of gait. What exactly is FoG, how did the researchers use low-cost wearable devices to help treat it, and how does this demonstrate the power of AI?
Freezing of Gait (FoG) is a response in some Parkinson’s patients whereby their motor functions are briefly interrupted. Specifically, FoG will see patients unable to walk forward despite the full intention to do so, which essentially results in the patient remaining still. It is not entirely clear what causes this response due to the complex nature of neurological conditions, but it is believed that it links to how the brain processes information and cognition.
FoG rarely occurs in those walking in a straight line and generally occurs when obstacles are met, or a change of direction is needed. This provides evidence that the act of making decisions could be a cause of FoG. Furthermore, FoG can also occur on environmental and/or emotional triggers, but these vary between patients.
FoG can be hazardous for sufferers as it can interfere with essential activities. For example, walking across a street, one could see someone suddenly unable to move out of traffic. Another example could be navigating around dangerous environments (such as hills and cliff edges). Trying to treat FoG is difficult as it needs to be identified when it occurs. Once detected, a patient can stimulate their body by shock or vibration to remind them to start moving again.
One of the biggest challenges currently faced is the cost of equipment needed to detect FoG. Researchers from William and Mary University and Virginia Commonwealth University have recently developed a system that can detect FoG in Parkinson’s patients with a high degree of accuracy using off-the-shelf wearable devices. The researchers specifically pointed out one piece of technology called the Protokinetics Zeno Walkway mat, which can measure an individual’s FoG. Still, the price tag of $50,000 makes it out of reach for most.
Instead, the research team look towards an UltiGesture wristband that can gather basic data, including temperature, acceleration, and gyroscopic measurements. Combining this data with an AI algorithm, the team created an FoG detection system whose ability was similar to the mat while only costing tens of dollars.
Furthermore, the team combined the wearable with a secondary ankle-worn wearable that vibrates when a freeze is detected. The need for the AI algorithm helps distinguish between a natural stop and an unintentional freeze, and the use of the ankle device helps remind the patient to start moving again.
The use of low-cost sensors combined with AI clearly demonstrates the power of AI. Instead of relying on expensive equipment and sensors, an AI-based system can achieve a high degree of accuracy using off-the-shelf equipment available to the masses.
The use of AI in the medical field will help provide predictive diagnostics and lower the costs of getting a professional medical diagnosis. In many cases, catching diseases before they become problematic is far more cost-effective (such as cancerous moles), meaning that treatment costs can also be lower. Preventing diseases before they worsen can also help improve the quality of life (such as reversing eye damage caused by glaucoma), which is something that AI would be able to do very effectively.
AI is a powerful tool that only continues to grow in strength, and this use of AI with a low-cost wearable sensor is a clear example of its capability to change lives for the better.