Real-Time Rodent Pupillometry on an Embedded Platform for Neuromodulation

Area of Science:

• Biomedical Engineering
• Neuroscience
• Ophthalmology

Background:

• Pupillometry is a key non-invasive biomarker for autonomic and cortical activity in neuromodulation research.
• Existing pupillometry systems are unsuitable for rodents due to anatomical and artifact challenges.

Purpose of the Study:

• To develop a low-cost, real-time rodent pupillometry system for preclinical neuromodulation studies.
• To overcome limitations of current systems in rodent ocular morphology and fur artifacts.

Main Methods:

• An integrated, embedded system with a rule-based algorithm featuring adaptive ellipse fitting, RGB masking, and greedy tracking.
• Optimization for infrared illumination, avoiding GPU acceleration.
• Validation through in vivo vagus nerve stimulation (VNS) experiments in rats.

Main Results:

• Achieved an 86.0% detection rate in rat pupillometry, significantly outperforming existing methods (63.1%).
• Demonstrated robust performance minimizing fur and ocular artifacts.
• Successfully correlated dynamic pupil size changes with VNS intensity.

Conclusions:

• The developed system provides a practical, low-cost solution for rodent pupillometry in preclinical research.
• This tool enhances the evaluation of VNS and highlights pupillometry's potential as a non-invasive neuromodulation biomarker.