# Development and Validation of an IMU Sensor-Based Behaviour-Alert Detection Collar for Assistance Dogs: A Proof-of-Concept Study

## TL;DR

This study developed a wearable collar that detects trained alert behaviors in assistance dogs, aiming to improve seizure response for people with epilepsy.

## Contribution

The study introduces a novel wearable system that uses machine learning and motion sensors to detect specific alert behaviors in assistance dogs.

## Key findings

- The system achieved up to 92.4% cross-dog accuracy in detecting a standardized spin alert behavior.
- Random Forest was the top-performing model with an F1-score of 0.65 and accuracy of 92%.
- The prototype demonstrates technical feasibility for automated detection of trained alert behaviors in assistance dogs.

## Abstract

Seizure-alert dogs can offer early warnings of seizures to individuals with epilepsy, yet existing approaches to using alert dogs rely on spontaneous behaviours that are difficult to validate or replicate. This study presents a wearable behaviour-alert detection collar designed to recognise signalling behaviours in trained assistance dogs using machine learning and motion sensors. Data were collected from six trained dogs performing a standardised spin alert behaviour, producing 135 labelled spin events. By standardising the alert behaviour and automating detection, the system achieved reliable recognition of spins across dogs, with cross-dog accuracy reaching up to 92.4%. This prototype demonstrates a novel, animal-integrated solution for improving seizure response and care.

Assistance dogs have shown promise in alerting to epileptic seizures in their owners, but current approaches often lack consistency, standardisation, and objective validation. This proof-of-concept study presents the development and initial validation of a wearable behaviour-alert detection collar developed for trained assistance dogs. It demonstrates the technical feasibility for automated detection of trained signalling behaviours. The collar integrates an inertial sensor and machine learning pipeline to detect a specific, trained alert behaviour of two rapid clockwise spins used by dogs to signal a seizure event. Data were collected from six trained dogs, resulting in 135 labelled spin alerts. Although the dataset size is limited compared to other machine learning applications, this reflects the real-world constraint that it is not practical for assistance dogs to perform excessive spin signalling during their training. Four supervised machine learning models (Random Forest, Logistic Regression, Naïve Bayes, and SVM) were evaluated on segmented accelerometer and gyroscope data. Random Forest achieved the highest performance (F1-score = 0.65; accuracy = 92%) under a Leave-One-DOG-Out (LODO) protocol. The system represents a novel step toward combining intentional canine behaviours with wearable technology, aligning with trends on the Internet of Medical Things. This proof-of-concept demonstrates technical feasibility and provides a foundation for future development of real-time seizure-alerting systems, representing an important first step toward scalable animal-assisted healthcare innovation.