Enterprise AI Analysis
Artificial Intelligence-Based Video Analysis for Assessing Sucking Behavior in Preterm Infants: A Feasibility Study
Background/Objectives: Preterm infants often experience impaired swallowing function, and objective assessments for this population remain limited. In this prospective single-center study, we aimed to propose and validate an automated framework that quantitatively assesses neonatal sucking behavior by tracking facial key points in bottle feeding videos.
Methods: Fifty-eight preterm infants (corrected age [CA] ≤ 2 months) were enrolled, and 2 min videos of bottle-feeding were recorded. Certified therapists manually evaluated the videos using the Neonatal Oral Motor Assessment Scale (NOMAS), and an artificial intelligence (AI)-based analysis classified the videos into the following three groups: Normal, Disorganization, and Dysfunction. At 12 months CA, developmental outcomes were assessed using the Mental Development Index (MDI) and the Psychomotor Development Index (PDI) of the Bayley Scales of Infant Development, Second Edition (BSID-II).
Results: Among the 58 infants, the AI-based tool correctly classified 47 and misclassified 11. The classification accuracy was 82.76 for the Normal group, 82.76 for Disorganization, and 96.55 for Dysfunction. The mean PDI was lower in the Dysfunction group than in other groups; however, the differences were not statistically significant.
Conclusions: This novel AI-based video analysis demonstrates preliminary potential as a noninvasive tool for evaluating sucking behavior in preterm infants, potentially enabling early identification of dysphagia even by non-specialists in the neonatal intensive care unit (NICU) without hazard exposure. This feasibility study demonstrates preliminary technical viability of a video-based framework for neonatal sucking behavior assessment; however, further validation is required before clinical implementation.
Executive Impact: Transforming Neonatal Care
Leverage cutting-edge AI for objective and early detection of feeding difficulties in preterm infants, reducing clinical burden and improving patient outcomes.
0
Classification Accuracy (Normal/Disorganization)
0
Classification Accuracy (Dysfunction)
0
Radiation Exposure
0
Ease of Use (Non-Specialists)
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
Explore the core results and technical methodology behind the AI-based assessment of neonatal sucking behavior.
96.55%
Accuracy for Detecting Dysfunction (vs. NOMAS)
The AI-based video analysis achieved a high classification accuracy of 96.55% for detecting Dysfunction in preterm infant feeding sessions, demonstrating strong potential for objective assessment of severe swallowing impairments.
AI Sucking Behavior Analysis Process
Video Recording (2 min)
→
Facial Region Cropping
→
Manual Keypoint Annotation (First Frame)
→
CoTracker2 Propagation
→
Geometry-Aware Correction
→
Parameters Extraction (Distance/Angle)
→
Peak Detection & Validation (NNS/NS)
→
Segment Identification (≥3 valid peaks)
→
NOMAS Classification (Normal/Disorganization/Dysfunction)
The proposed framework utilizes a sophisticated multi-step process for analyzing neonatal sucking behavior, integrating advanced computer vision techniques with NOMAS clinical criteria. This includes precise facial keypoint tracking and a hierarchical classification algorithm.
| Feature | Manual NOMAS | AI-Based Analysis |
|---|---|---|
| Assessment Method | Subjective bedside observation by certified therapists | Quantitative facial keypoint tracking from video |
| Inter-rater Reliability | Moderate (κ=0.598), variability issues noted | Consistent analytical criteria, reduced variability (potential) |
| Radiation Exposure | None (screening tool) | None (video-based) |
| Accessibility/Scalability | Requires certified specialists, time-consuming | Potentially usable by non-specialists in NICU, scalable for remote monitoring |
| Detection Strengths | Identifies disrupted rhythmic jaw movement, structural abnormalities | High accuracy for Normal/Disorganization (82.76%) and Dysfunction (96.55%) |
While manual NOMAS relies on expert observation, the AI system provides objective, consistent analysis of sucking patterns, addressing challenges like inter-rater variability and enabling broader applicability.
Understand the practical benefits and future potential of integrating AI-powered sucking analysis into your neonatal care workflows.
Transforming Dysphagia Screening in NICU
This AI-based video analysis offers a promising future for early dysphagia detection in preterm infants, especially in busy NICU environments.
The Challenge
Current diagnostic tools like VFSS involve radiation, limiting routine use, while NOMAS, though non-invasive, suffers from subjective interpretation and inter-rater variability. Early detection is crucial for timely intervention to prevent poor weight gain and developmental delays.
The AI Solution
The AI framework provides a non-invasive, objective, and consistent method for assessing sucking behavior. By tracking facial key points and classifying feeding patterns against NOMAS criteria, it allows non-specialists to identify feeding difficulties without hazard exposure.
Expected Outcome
Potentially enables earlier identification of dysphagia, facilitating timely clinical intervention and supporting optimal neurodevelopmental outcomes for preterm infants. It reduces reliance on specialized personnel for initial screening and offers a scalable solution for continuous monitoring.
The implications of this technology are significant for neonatal care, offering a non-invasive and objective screening tool that can be deployed in NICUs to support early identification of dysphagia by non-specialists.
Calculate Your Potential ROI
Estimate the efficiency gains and cost savings your organization could achieve by implementing AI-powered solutions.
Estimated Annual Savings
Annual Hours Reclaimed
Your AI Implementation Roadmap
A phased approach to integrating AI into your operations for maximum impact and minimal disruption.
Phase 1: Discovery & Strategy
Initial consultations to understand your specific needs, data landscape, and strategic objectives. We define project scope, KPIs, and success metrics.
Phase 2: Data & Model Development
Data collection, annotation, and custom AI model training. This includes validating models against your specific clinical criteria and data.
Phase 3: Integration & Pilot
Seamless integration of the AI framework into your existing systems (e.g., NICU monitoring platforms). Pilot deployment in a controlled environment for testing and refinement.
Phase 4: Scaling & Optimization
Full-scale deployment across your enterprise. Continuous monitoring, performance optimization, and iterative improvements based on real-world feedback.
Ready to Transform Your Enterprise with AI?
Connect with our experts to explore how AI can drive efficiency, reduce costs, and unlock new opportunities for your organization.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat