Eyes-On
Vital Vision AI is a technology company dedicated to providing high-quality, continuous observation and monitoring for medically complex children at home. We created Eyes-On, an artificial intelligence–assisted system for visual observation and vital sign monitoring, designed to prevent complications in children with tracheostomies.
Eyes-On Background
Understanding the background and need for a deep learning–assisted visual tracking and predictive cardiopulmonary monitoring system.
Tracheostomy is a surgical procedure that creates an opening in the front of the neck into the windpipe (trachea) to provide an alternative airway. It is often performed in children to bypass an obstructed airway caused by a blockage in the upper throat or to address chronic breathing problems related to prematurity, particularly when mechanical ventilation is required. Children who undergo a tracheostomy require ongoing care and support to manage their airway and maintain optimal respiratory function. Thanks to advances in critical care, many of these children now survive hospitalization and are able to continue living at home with their tracheostomy.
Currently, tracheostomy-dependent children are observed at home by their parents or other caregivers, much like any other infant. Depending on availability, home healthcare providers may periodically visit to assess the tracheostomy and identify potential breathing issues. Home monitoring is typically limited to an oxygen saturation monitor that sounds an audible alarm when the child’s oxygen level drops below a preset threshold.
Children with a tracheostomy require vigilant monitoring to prevent complications related to a malfunctioning tracheostomy tube. The risk of sudden death due to a tracheostomy-related complication at home is 1% per year, with an overall mortality rate of 20% for infant tracheostomy patients. Additionally, many of these children suffer debilitating anoxic brain injuries due to an occluded or displaced tracheostomy tube. The risk is particularly high in infants under age three, who are entirely dependent on a functioning tracheostomy tube for breathing.
Ideally, tracheostomy-dependent children would receive continuous caregiver monitoring, including 24/7 observation for tracheostomy tube malfunctions. However, due to caregiver time constraints and limited access to home health nursing, continuous observation in the home is not feasible.
The Product
To reduce the risk of severe complications in tracheostomy-dependent children, we developed Eyes-On, a deep learning–assisted visual tracking and predictive cardiopulmonary monitoring system integrated with an infant’s crib. The video subsystem continuously tracks the tracheostomy tube’s position and the patient’s movements in real time, processing video at the bedside. Advanced AI software also monitors the child’s vital signs, detecting trends in heart and lung function.
The system leverages modern computer vision models, time-series analysis, generative AI, and symbolic techniques to alert caregivers and remote monitoring facilities of potential adverse events. By issuing real-time alerts for abnormal tube positions and identifying trends in vital signs, Eyes-On can predict respiratory conditions before they become life-threatening. It also generates fewer false alarms than current state-of-the-art monitoring systems and adapts to environmental changes such as noise, lighting, background, and camera distance.
Our Team
Adele K Evans, MD
Chief Clinical Officer & Director of Data Acquisition
Adele has been a practicing pediatric otolaryngologist for over 20 years. She specializes in managing children with complex airway disorders and is nationally recognized for her expertise and advocacy in caring for children with tracheostomies. She earned her MD from Emory University in Atlanta and completed her Otolaryngology–Head and Neck Surgery residency at Harvard University’s Massachusetts Eye and Ear Infirmary. Following her residency, she pursued a fellowship in pediatric otolaryngology at the Children’s Hospital of Philadelphia. Adele currently serves as the medical director of Tracheostomy and Complex Airway Care at Nemours Children’s Hospital, Florida, and as a professor of Otolaryngology–Head and Neck Surgery at the University of Central Florida College of Medicine.
Supratik Mukhopadhyay, PhD
Chief Technology Officer
Supratik is a professor at Louisiana State University in the School of Environmental Science. He has applied AI to solve problems in drug discovery, histopathology, climate change, agriculture, infrastructure, transportation, and education. He is also an expert in video analytics and computer vision and serves as an associate editor for IEEE Transactions on Artificial Intelligence. He earned his PhD in engineering from the Max Planck Institute for Informatics at Saarland University in Germany.
Michael E. Dunham, MD, MS
Acting CEO & Chief Software Engineer
Michael has been a practicing pediatric otolaryngologist for over 30 years, caring for numerous infants who require a tracheostomy to maintain respiratory function. He earned his MD and completed a residency in Otolaryngology at Tulane University, followed by a fellowship in Pediatric Otolaryngology at Northwestern University in Chicago. Additionally, he holds a master’s degree in computer science with a specialization in computer vision from DePaul University.