Personalized Adaptive Neuro-symbolic Data-fusion for Infant Assessment
Abstract
PANDIA is a novel multimodal AI system for robust, interpretable, and personalized infant pain assessment (ages 0–3 months) in NICUs and family medicine clinics. It combines: Lightweight CNN/TCN encoders (MobileNetV3 + TCN) for video, audio, and physiological signals Concept Bottleneck Layer mapping raw inputs to 12 clinically meaningful concepts Graph Neural Network (GraphSAGE) for inter-modal relational reasoning MAML-based meta-learning for rapid infant-specific personalization (≥5 labeled samples) Evidential uncertainty quantification for safe abstention and clinician deferral Symbolic rule engine for human-interpretable explanations Federated contrastive pretraining for privacy-preserving multi-site collaboration Key Results (2,847 infants, 4 datasets, 4 continents)
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.