SBIR Phase I:Prototyping a Wearable Device that Continuously Monitors Biometrics using Machine Learning to Predict Meltdowns in Children with Autism

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is in its ability to use machine learning and wearable technology to reduce uncontrolled destructive episodees, known as meltdowns, in children with autism. Meltdowns are highly distressing events for these children and their families and may require intervention on behalf of emergency response personnel and healthcare providers. Treating individuals with autism by proactively detecting meltdowns will allow caregivers time to intervene, mitigate, and prevent the onset of destructive behavioral episodes. The ability to predict a meltdown, and then implement strategic intervention to prevent the meltdown, may have positive life-changing effects for the children, their families, and their caretakes by reducing social stigma, enabling more mainstreaming of school and family activities, and reducing significant financial healthcare burdens. This technology may also be used to mitigate panic attacks in individuals with post-traumatic stress disorders.This Small Business Innovation Research (SBIR) Phase I project seeks to develop a wearable device that detects, predicts, and helps prevent meltdowns in children with autism. Wearable devices that measure physiological parameters are available in the market, but none of them are specific to autism, and none of them proactively predict behavior episodes. A unique feature of this wearable device is that it uses machine learning to predict meltdowns. Incorporating machine learning allows each device to learn the unique biometric signature of the wearer so it can predict meltdowns with high accuracy. When a child is at high risk for a meltdown, the device will detect the relevant physiology and alert caregivers and therapists in time to intervene. The objectives of this project are to create a prototype which includes the wearable product and the individualization enabled by machine learning to correlate a child’s biometric measures with behavioral states. The goal is to achieve decreased frequency and/or severity of meltdowns. By enabling caregivers of children with autism to take control of and prevent meltdowns, this project has the potential to eliminate the stigma these children face during a meltdown in public, and empowers the children to become more independent as they age.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.