Related Experiment Video
Updated: Oct 16, 2025
Preterm EEG: A Multimodal Neurophysiological Protocol
Published on: February 18, 2012
Deep Multimodal Learning From MRI and Clinical Data for Early Prediction of Neurodevelopmental Deficits in Very
Lili He1,2,3, Hailong Li1,2, Ming Chen1,2,4
1Imaging Research Center, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States.
Insights
Predicting neurodevelopmental deficits in very preterm infants (VPIs) is crucial. Multimodal MRI and clinical data integrated into deep learning models accurately predict cognitive, language, and motor outcomes by age two.
Area of Science:
- Neuroscience
- Medical Imaging
- Artificial Intelligence
Background:
- The number of disabled survivors of prematurity, particularly very preterm infants (VPIs) born before 32 weeks gestational age, has risen significantly.
- VPIs face a high risk of neurodevelopmental impairments, necessitating early and effective prediction for timely interventions during critical neuroplasticity windows.
- Multimodal MRI offers a comprehensive, non-invasive approach to visualize infant brain development and pathology, surpassing unimodal MRI.
Purpose of the Study:
- To investigate the predictive capability of multimodal magnetic resonance imaging (MRI) and clinical data for neurodevelopmental deficits in VPIs.
- To test the hypothesis that integrating multimodal MRI (T2-weighted, diffusion tensor imaging, resting-state functional MRI) with clinical data enhances prediction accuracy.
- To develop and evaluate novel end-to-end deep multimodal models for predicting cognitive, language, and motor deficits at two years corrected age.
Main Methods:
- Collected quantitative multimodal MRI data (T2-weighted, DTI, rs-fMRI) and clinical information from VPIs.
- Developed novel end-to-end deep multimodal learning models to integrate diverse data sources.
- Independently predicted cognitive, language, and motor deficits at two years corrected age using the developed models.
Main Results:
- The integrated multimodal models achieved high prediction accuracies: 88.4% for cognitive, 87.2% for language, and 86.7% for motor deficits.
- Performance significantly surpassed predictions made using individual data modalities.
- This study serves as a proof-of-concept for the efficacy of multimodal data integration in predicting neurodevelopmental outcomes.
Conclusions:
- Integrating multimodal MRI and clinical data with deep learning models provides a powerful tool for predicting neurodevelopmental deficits in very preterm infants.
- The findings demonstrate the potential for improved early diagnosis and personalized treatment strategies for at-risk infants.
- Further validation with larger cohorts and external datasets is recommended to confirm clinical utility and generalizability.
Abstract:
The prevalence of disabled survivors of prematurity has increased dramatically in the past 3 decades. These survivors, especially, very preterm infants (VPIs), born ≤ 32 weeks gestational age, are at high risk for neurodevelopmental impairments. Early and clinically effective personalized prediction of outcomes, which forms the basis for early treatment decisions, is urgently needed during the peak neuroplasticity window-the first couple of years after birth-for at-risk infants, when intervention is likely to be most effective. Advances in MRI enable the noninvasive visualization of infants' brains through acquired multimodal images, which are more informative than unimodal MRI data by providing complementary/supplementary depicting of brain tissue characteristics and pathology. Thus, analyzing quantitative multimodal MRI features affords unique opportunities to study early postnatal brain development and neurodevelopmental outcome prediction in VPIs. In this study, we investigated the predictive power of multimodal MRI data, including T2-weighted anatomical MRI, diffusion tensor imaging, resting-state functional MRI, and clinical data for the prediction of neurodevelopmental deficits. We hypothesize that integrating multimodal MRI and clinical data improves the prediction over using each individual data modality. Employing the aforementioned multimodal data, we proposed novel end-to-end deep multimodal models to predict neurodevelopmental (i.e., cognitive, language, and motor) deficits independently at 2 years corrected age. We found that the proposed models can predict cognitive, language, and motor deficits at 2 years corrected age with an accuracy of 88.4, 87.2, and 86.7%, respectively, significantly better than using individual data modalities. This current study can be considered as proof-of-concept. A larger study with external validation is important to validate our approach to further assess its clinical utility and overall generalizability.
More Related Videos
11:14A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants
Published on: October 4, 2015
08:05Measuring Statistical Learning Across Modalities and Domains in School-Aged Children Via an Online Platform and Neuroimaging Techniques
Published on: June 30, 2020