Self-Injury Prevention Device for Autistic Children in Exercise Introduction Autistic spectrum disorder (ASD) is a neurodevelopmental disorder that affects social communication, interaction, behaviors, and interests that affects 0.7% of children in China[1]. One common treatment method is exercise, which is found to bring many positive behavioral changes[2]. One of the key symptoms of people with ASD is self-injurious behaviors (SIB) such as head-banging, hair-pulling, and self-punching[3]. However, the currently available SIB prevention devices on the market cannot function properly when autistic children are doing sports since they cannot differentiate the signals from sports and SIB. In this project, I developed an inertial sensor-based sensing system to prevent autistic children from performing head-banging behaviors during sports. Methods The design utilizes an inertial sensor to track the head movement of the wearer, specifically the MPU6050 sensor. Inertial sensors, also known as inertial measurement units (IMUs), consist of accelerometers and gyroscopes that measure linear acceleration and angular velocity respectively. The system circuit consists of an Arduino UNO controller, an MPU6050 chip, three resistors, and three LEDs. Each LED and resistor connected in series are connected in parallel with each other, and this system is connected to the Arduino UNO and MPU6050 in series. The algorithm design sets a threshold, in this case 50 degrees per second, for angular acceleration along the z-axis of the gyroscope. If the absolute value of the angular acceleration exceeds the threshold value, the blue or white LED will light up. If the absolute value of the angular acceleration does not exceed the threshold value, the green LED will light up. Results The final demo consists of a cardboard monitoring panel with the Arduino UNO board, breadboard, resistors and LEDs attached to it, and a sports headband with the MPU6050 attached to it. The reason why this device will function during exercise is that most exercises only involve linear and not angular acceleration around the head area. For instance, sensory integration therapy, an exercise-based rehabilitation therapy for children with ASD, mostly involves linear acceleration and low angular acceleration of the head. Discussion There are still many possible improvements to the final product. The most important and necessary modification is utilizing wireless connection between the sensor and control board, which can ensure a full range of movement of the child. Another modification for improving accuracy is tracking the frequency of the head-movement detected before notification, such as only alerting the caregiver after the head moves back-and-forth multiple times, further preventing misinterpretation of exercise and self-injurious behaviors. Conclusion The final prototype is a wearable self-injury prevention device specifically designed for autistic children during exercise, utilizing the Arduino UNO platform, MPU6050 inertial sensor, and sensing algorithm. I implemented a detection system that measures the angular velocity of the head, detects the occurrence of head-banging, and alerts caregivers when the angular velocity exceeds a pre-defined safe threshold. This device also established a versatile platform that allows customization for different exercise scenarios by adjusting threshold values and updating sensing logic, enabling the device to adapt to individual needs and varying exercise routines. [1] Zhou, Hao, et al. "Prevalence of autism spectrum disorder in China: a nationwide multi- center population-based study among children aged 6 to 12 years." Neuroscience Bulletin 36 (2020): 961-971. [2]Sowa, Michelle, and Ruud Meulenbroek. "Effects of physical exercise on autism spectrum disorders: A meta-analysis." Research in autism spectrum disorders 6.1 (2012): 46-57. [3] Minshawi, Noha F., et al. "The association between self-injurious behaviors and autism spectrum disorders." Psychology research and behavior management (2014): 125-136.