AI-powered computer vision system to support non-invasive, early diagnosis of feeding challenges in newborns.
In collaboration with researchers at Northeastern University, we helped bring to life NeuroSense Diagnostics—a research-to-product initiative aimed at improving developmental outcomes for infants. The solution uses computer vision and facial landmark detection to monitor infant sucking behavior via mobile devices, offering an affordable and accessible alternative to traditional NICU tools.
Weeks to launch from client brief
to MVP
Premature and at-risk infants often struggle with feeding, and traditional clinical assessments are subjective, manual, and inconsistent—typically relying on gloved-finger evaluations or observation. There was no scalable, remote, or affordable solution to track and assess non nutritive sucking (NNS), which is critical for early intervention and neurodevelopmental support.
The NEU team needed a technology partner to turn their cutting-edge academic research into a functioning mobile app with a robust AI model running in real-time.
We partnered with NEU’s Speech and Neurodevelopment Lab and the Augmented Cognition Lab to design and build a mobile application that uses facial landmark detection models to assess infant mouth movement during pacifier use. The model captures subtle muscle movements and identifies the presence of sucking patterns, which are essential for diagnosing NNS development.
The app is designed to be used by clinicians and researchers in both hospital and home settings, using only a smartphone or tablet camera.
The project resulted in a fully functional AI-powered mobile prototype capable of running video-based sucking assessments and logging patient sessions for further review. This version is currently supporting further research and grant applications, and is expected to dramatically reduce the cost and subjectivity of infant feeding assessments while improving clinical outcomes through earlier interventions.
AI-powered internal analytics platform for enhanced decision-making
Empowering developers to integrate AI into real-world software development workflows.
Customizing a domain-specific LLM to power intelligent search and knowledge discovery across scientific datasets.
AI-powered computer vision system to support non-invasive, early diagnosis of feeding challenges in newborns.
Modernizing data infrastructure and enabling intelligent decision-making for one of the world’s leading investment firms.
Empowering employees to document experiences, understand rights, and navigate workplace challenges with confidence.
e.
Simplifying the mortgage journey for homebuyers and lenders through intelligent automation and personalized guidance.
Helping organizations improve employee health through personalized coaching, smart goal tracking, and AI-powered feedback.
RadCard set out to create a gamified loyalty and discovery platform that helps users discover local businesses, events, and promotions while earning rewards…
Facilitating spontaneous connections through intelligent activity recommendations and user matching.
Empowering individuals worldwide to engage in fitness challenges, track progress, and achieve their health goals through a competitive and supportive community.
Revolutionizing dance education through personalized class recommendations and enhanced social engagement.
Connecting individuals with personalized local experiences through intelligent recommendations and community engagement.
Launched a full-stack IoT solution that helped Atomi break into the smart home market and secure major retail partnerships.
.
WaBu (short for “Walking Buddy”) is a mobile app designed to connect university students and older adults for walk-based social wellness.
Merging fan commentary and real-time engagement into a community-driven sports debate experience.
Empowering businesses with a unified, modular security solution to combat evolving cyber threats.
Empowering employees to document experiences, understand rights, and navigate workplace challenges with confidence.
Skip to content
