Saved in:
Bibliographic Details
Main Authors: Holmberg, Daniel, Airaksinen, Manu, Marchi, Viviana, Guzzetta, Andrea, Kivi, Anna, Haataja, Leena, Vanhatalo, Sampsa, Roos, Teemu
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.14400
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915047241940992
author Holmberg, Daniel
Airaksinen, Manu
Marchi, Viviana
Guzzetta, Andrea
Kivi, Anna
Haataja, Leena
Vanhatalo, Sampsa
Roos, Teemu
author_facet Holmberg, Daniel
Airaksinen, Manu
Marchi, Viviana
Guzzetta, Andrea
Kivi, Anna
Haataja, Leena
Vanhatalo, Sampsa
Roos, Teemu
contents Reliable methods for the neurodevelopmental assessment of infants are essential for early detection of problems that may need prompt interventions. Spontaneous motor activity, or 'kinetics', is shown to provide a powerful surrogate measure of upcoming neurodevelopment. However, its assessment is by and large qualitative and subjective, focusing on visually identified, age-specific gestures. In this work, we introduce Kinetic Age (KA), a novel data-driven metric that quantifies neurodevelopmental maturity by predicting an infant's age based on their movement patterns. KA offers an interpretable and generalizable proxy for motor development. Our method leverages 3D video recordings of infants, processed with pose estimation to extract spatio-temporal series of anatomical landmarks, which are released as a new openly available dataset. These data are modeled using adaptive graph convolutional networks, able to capture the spatio-temporal dependencies in infant movements. We also show that our data-driven approach achieves improvement over traditional machine learning baselines based on manually engineered features.
format Preprint
id arxiv_https___arxiv_org_abs_2402_14400
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Learning Developmental Age from 3D Infant Kinetics Using Adaptive Graph Neural Networks
Holmberg, Daniel
Airaksinen, Manu
Marchi, Viviana
Guzzetta, Andrea
Kivi, Anna
Haataja, Leena
Vanhatalo, Sampsa
Roos, Teemu
Computer Vision and Pattern Recognition
Machine Learning
Image and Video Processing
68T06
I.2; I.4; J.3
Reliable methods for the neurodevelopmental assessment of infants are essential for early detection of problems that may need prompt interventions. Spontaneous motor activity, or 'kinetics', is shown to provide a powerful surrogate measure of upcoming neurodevelopment. However, its assessment is by and large qualitative and subjective, focusing on visually identified, age-specific gestures. In this work, we introduce Kinetic Age (KA), a novel data-driven metric that quantifies neurodevelopmental maturity by predicting an infant's age based on their movement patterns. KA offers an interpretable and generalizable proxy for motor development. Our method leverages 3D video recordings of infants, processed with pose estimation to extract spatio-temporal series of anatomical landmarks, which are released as a new openly available dataset. These data are modeled using adaptive graph convolutional networks, able to capture the spatio-temporal dependencies in infant movements. We also show that our data-driven approach achieves improvement over traditional machine learning baselines based on manually engineered features.
title Learning Developmental Age from 3D Infant Kinetics Using Adaptive Graph Neural Networks
topic Computer Vision and Pattern Recognition
Machine Learning
Image and Video Processing
68T06
I.2; I.4; J.3
url https://arxiv.org/abs/2402.14400