Maternal Risk, Fetal-Neonatal Brain Connectivity, and Early Neurodevelopment: A Longitudinal Observational Study

Overview

This study aims to understand how a pregnant woman's health, lifestyle, and psychological state-especially when associated with known risk factors-might influence the developing brain of her baby, both before and after birth. Specifically, the research investigates whether differences in brain connectivity observed through fetal and neonatal magnetic resonance imaging (MRI) can predict how a child will develop cognitively, emotionally, and behaviorally from birth through early childhood. This is a prospective, observational study that will follow 160 pregnant women and their children over time. Participants will be enrolled at the Gynecology and Obstetrics Unit of San Raffaele Hospital in Milan. Using advanced brain imaging techniques (resting-state functional MRI), the study will examine how key brain systems-such as those involved in movement, hearing, vision, language, and attention-are connected during fetal life and shortly after birth. The study also evaluates how these patterns of brain connectivity relate to later developmental outcomes, assessed through standard neuropsychological tests from birth up to 6 years of age. One of the study's core hypotheses is that early patterns of brain connectivity-especially when combined with detailed profiles of maternal health and risk-can serve as early markers of a child's neurodevelopmental path. To explore this, the study uses an integrated approach that combines imaging data with clinical and psychological information from the mother (e.g., her stress levels, medical history, and lifestyle habits). Participants are grouped based on the "Maternal Frailty Inventory," a tool that captures the cumulative risk profile of each mother. The sample will include mothers with both low and medium-high risk scores. This grouping allows researchers to investigate how varying degrees of maternal risk are reflected in the baby's early brain organization and how this, in turn, influences developmental milestones. A secondary aim of the study is to investigate how emotional responses to music may affect fetal brain activity. During the fetal MRI, mothers will listen to selected musical pieces. Researchers will examine if the baby's brain is influenced by the mother's emotional state. Ultimately, the study hopes to build predictive models-using artificial intelligence and advanced statistical techniques-that can estimate a child's developmental trajectory based on early brain imaging and maternal data. This could provide an important step toward early identification of children who might benefit from developmental support or intervention, even before symptoms appear.

This single-center, prospective longitudinal observational cohort study-entitled Maternal Risk, Fetal-Neonatal Brain Connectivity, and Early Neurodevelopment (MaMRI-NeUCogI)-is designed to explore the relationship between maternal risk profiles, early-life brain connectivity, and developmental outcomes from birth to early childhood (up to 72 months). The protocol aims to trace the temporal continuity between functional neurodevelopmental markers present in utero or shortly after birth and subsequent cognitive, behavioral, and emotional trajectories during early childhood. Scientific Rationale A key challenge in developmental neuroscience is identifying early biomarkers that can predict individual differences in neurodevelopmental trajectories. The fetal and neonatal periods represent critical windows during which the brain undergoes major organizational changes. Disruptions or variations in these processes-particularly in the presence of maternal medical, psychological, or environmental risks-may lead to atypical connectivity patterns that forecast later neurodevelopmental difficulties. This study leverages resting-state functional MRI (rs-fMRI) in fetuses and neonates to map the functional architecture of core neural systems (sensorimotor, auditory, visual, language, and attention). The project builds upon prior work from the Italian Ministry of Health's "Ricerca Finalizzata 2016" (grant number RF-2016-02364081; Principal Investigator: Dr. Pasquale Anthony Della Rosa), expanding its focus to include a multivariate risk framework and an artificial intelligence-based predictive modeling approach. Study Population and Grouping A total of 160 pregnant women will be enrolled from the Gynecology and Obstetrics Unit at San Raffaele Hospital, Milan. They will be stratified into two groups based on the Maternal Frailty Inventory (MaFra) developed by Della Rosa et al. (2021), which integrates clinical (e.g., obstetric, gynecological) and non-clinical (e.g., psychological, lifestyle) risk factors: * Medium-to-high risk group (n = 96): Representing mothers with significant maternal frailty indices. * Low-risk group (n = 64): Reflecting minimal clinical and psychosocial risk burden. This stratification is established a posteriori based on a risk profile classification aligned with research goals, and is not connected to clinical diagnoses or intervention decisions. Imaging Protocol and Data Collection All participants will undergo fetal and/or neonatal rs-fMRI, depending on clinical indications and risk group membership. Imaging data will be used to derive metrics of functional connectivity, specifically: * Local connectivity: Connectivity between regions within the same system (e.g., sensorimotor, auditory). * Global connectivity: Connectivity between regions across different systems. * Segregation indices: Reflecting within-system connectivity. * Integration indices: Reflecting cross-system connectivity. Functional connectivity parameters will be estimated for each subject using region-based parcellations aligned with validated fetal and neonatal brain templates. Structural MRI will also be acquired to confirm normative brain development and rule out major anomalies. Longitudinal Neurodevelopmental Follow-up Children born to participating mothers will undergo standardized neuropsychological assessment at several developmental milestones from birth to 72 months. These assessments will yield dimensional scores across various cognitive, behavioral, and emotional domains, including: * Sensorimotor processing * Language development * Attention and executive function * Socioemotional regulation * Adaptive behavior The association between early brain connectivity and later neurodevelopmental performance will be analyzed using both correlational methods and predictive modeling frameworks. Artificial Intelligence and Prediction Modeling A core innovation of the MaMRI-NeUCogI study lies in the use of ML models trained on imaging-derived connectivity features and maternal risk indices. The goal is to predict multidimensional developmental trajectories. The resulting predictive framework is intended to quantify deviation from typical developmental trajectories and may serve in the future to inform early intervention strategies. Secondary Aims: Maternal Emotional State influence on fetal brain connectivity A secondary component of the study investigates the impact of emotional responses to music on fetal brain connectivity. During fetal rs-fMRI, participating mothers will listen to emotionally evocative music. The study will examine how maternal emotional valence and arousal ratings relate to fetal connectivity patterns. Data Integration and Analytic Plan The study adopts a multi-tiered analytic approach: 1. Descriptive statistics for maternal risk profiles and neurodevelopmental scores. 2. Group comparisons across maternal risk strata. 3. Correlation and regression analyses between functional connectivity metrics and neurodevelopmental outcomes. 4. Predictive modeling using machine learning to predict later developmental profiles. All analyses will consider longitudinal dependencies, potential confounders (e.g., gestational age, birth outcomes), and interactions between maternal risk variables and imaging biomarkers.