AI Guided Contact-less Infant Monitoring Platform for Non-nutritive Sucking
Northeastern spinout, NeuroSense, has developed a computer vision-based monitoring platform to assess sucking skills of infants

Background
Infants with feeding challenges are at higher risk for developmental delays. Non-nutritive sucking (NNS) refers to the sucking action that occurs when a finger, pacifier, or other object is placed in a baby’s mouth, but there is no nutrient delivered. It is one of the earliest motor behaviors that occurs when the baby is born. Data such as sucking frequency, the number of cycles and their amplitude provides important clues about the brain development of infants and preterm infants. There is increasing evidence that NNS behavior can be linked to childhood language, motor abilities, and overall neurodevelopment. However, existing methods for collecting NNS data include devices with embedded sensors and bulky transducers that are inserted into the baby’s mouth. These devices are not only expensive, and largely prohibitive for at-home use, but their invasive nature has a direct impact on the baby’s natural sucking behavior. When sensors are not available, clinicians will also often assess NNS using a gloved finger, which is a highly subjective method of acquiring data. With pediatric disorders increasing and affecting around 20-50% of developing children, there is a crucial need for a contact-less NNS data collection and analysis system, that is accurate, affordable, and that can be used in a baby’s natural environment.
Technology Overview
NeuroSense has developed a facial landmark tracking technique using machine learning and computer vision algorithms to track jaw movements remotely from video. This contact-less NNS data acquisition and quantification platform successfully measures parameters such as neonatal sucking amplitude, frequency in a burst, suck cycle count in a burst, and cycle and burst counts in a session. Analysis of such indicators can provide insight into key developmental information of infants and developing children. The remote tracking approach allows the infants to perform their natural NNS patterns without the intervention of clinicians and invasive devices. So far, the inventors have completed a beta-test of the application using the facial landmark estimation model and real-time video streaming on Android and iOS platforms.
Benefits
- Remote tracking allows easy analysis of NSS without intervening with natural process
- Inexpensive method as compared to existing technology ($2,500-$150,000)
- At-home measurement system with physician or pediatrician connections
Applications
- Baby monitoring device for assessing growth and developmental issues
Opportunity
Connecting with prospective investors, applying for grant funding
IP Status
- Provisional patent
Seeking
- Development partner
- Commercial partner
- University spin out
- Seeking investment