Preterm birth is associated with significant mortality and a risk for lifelong morbidity. The complex multifactorial aetiology hampers accurate prediction and thus optimal care. A pipeline consisting of bespoke machine learning methods for data imputation, feature selection, and regression models to predict gestational age (GA) at birth was developed and evaluated from comprehensive multi-modal morphological and functional fetal MRI data from 333 control cases and 93 preterm birth cases. The GA at birth predictions were classified into term and preterm categories and their accuracy, sensitivity, and specificity were reported. An ablation study was performed to further validate the design of the pipeline. Performance was evaluated using stratified 10-fold cross-validation. The pipeline achieves an R2 score of 0.13 and a mean absolute error of 2.74 weeks. It also achieves a 0.77 accuracy, 0.59 sensitivity, and 0.82 specificity across folds. The predominant features selected by the pipeline include cervical length and statistics derived from placental T2* values. The confluence of fast, motion-robust and multi-modal fetal MRI techniques and machine learning prediction allowed the prediction of the gestation at birth. This information is essential for any pregnancy. To the best of our knowledge, preterm birth had only been addressed as a classification problem in the literature. Therefore, this work provides a proof of concept. Future work will increase the cohort size to allow for finer stratification within the preterm birth cohort. Our code is available at https://github.com/dfajardorojas/ml-for-preterm-birth-.

Cortical folding of the brain is widely regarded as an interplay between genetic programming and biomechanical forces, closely linked to cytoarchitectonic regionalisation. Abnormal folding patterns are frequently observed in neurodevelopmental conditions and psychiatric disorders. However, significant inter-individual variability of secondary and tertiary folds obscures the detection of shape biomarkers and confounds the investigation of folding-functional relationships. Here, we investigate cortical folding heterogeneity at a fine scale, using Multimodal Surface Matching with Hierarchical Templates (MSM-HT), a hierarchical surface registration, to parse cortical folding patterns into a representative family of distinct anatomical templates. By applying this technique both to young adults from the Human Connectome Project (HCP) and neonates in the Developing HCP and Brain Imaging in Babies (BIBS) cohorts, we identify and characterise common lobe-wise folding patterns: observing consistency across both age groups, with neonatal samples showing less variation. Crucially, we highlight significant hemispheric asymmetry within the temporal lobe in adults, with a consistent trend in neonates. This study provides a critical step towards understanding brain asymmetry and complex relationships between folding and function, offering a robust framework to generalise the uncovered cortical folding motifs across datasets and developmental stages. Guo and colleagues parse cortical folding variability into a family of distinct anatomical motifs that generalise across adults and neonates. These motifs shed light on structural asymmetry, heritability, and folding-function links.

Every human brain folds differently, and such natural variation confounds the search for imaging biomarkers of neurodevelopmental disorders. Physics-based simulation can help determine the causal mechanisms that underpin this variability. Yet every simulation must be initiated from a volumetric mesh of the brain's interior, tetrahedral or hexahedral, and it is the worst element in that mesh, not the average, that decides whether the simulation runs at all. Building that mesh from fetal MRI currently requires labour-intensive manual intervention. We therefore present CORTET (CORtical TETrahedral meshing): a fully automated pipeline that converts a triangulated cortical surface into a solver-ready tetrahedral mesh whose worst-element quality meets a strict quality target with no manual repair. By benchmarking against a general-purpose tetrahedral mesher on the same input surfaces, we isolate the pipeline's contribution from that of the input geometry, and we validate quality across a cohort of nearly 200 fetal subjects spanning the folding period. A mesh taken straight from the pipeline sustains a numerically stable morphoelastic folding simulation of a real fetal subject.

Lateralization is a fundamental principle of structural brain organization. In vivo imaging of brain asymmetry is essential for deciphering lateralized brain functions and their disruption in neurodevelopmental and neurodegenerative disorders. Here, we present a normative framework for benchmarking brain asymmetry across the lifespan, developed from an aggregated sample of 128 primary neuroimaging studies, including 177,701 scans from 138,231 individuals, jointly spanning the age range from 20 post menstrual weeks to 102 years. This resource includes comprehensive, hemisphere-specific brain growth charts for multiple neuroimaging phenotypes: regional cortical grey matter volume, thickness, surface area, and subcortical volumes. Our findings reveal distinct spatial patterns of asymmetry, with early leftward asymmetry observed in association cortices and late rightward asymmetry in sensory regions. These trajectories support theories of the neuroplasticity of asymmetry and the role of both genetic and environmental factors in shaping brain lateralization. Additionally, we provide tools to generate asymmetry centile scores, which allow the quantification of individual deviations from typical asymmetry throughout the lifespan and can be applied to unseen data or clinical populations. We demonstrate the utility of these models by highlighting group-level differences in asymmetry in autism spectrum disorder, schizophrenia, and Alzheimer's disease, and exploring genetic correlations with hemispheric specialization. To facilitate further research, we have made this normative framework freely available as an interactive open-access resource (upon publication), offering an essential tool to advance both basic and clinical neuroscience.

Current AI frameworks for brain decoding and encoding, typically train and test models within the same datasets. This limits their utility for cognitive training (neurofeedback) for which it would be useful to pool experiences across individuals to better simulate stimuli not sampled during training. A key obstacle to model generalisation is the degree of variability of inter-subject cortical organisation, which makes it difficult to align or compare cortical signals across participants. In this paper we address this through use of surface vision transformers, which build a generalisable model of cortical functional dynamics, through encoding the topography of cortical networks and their interactions as a moving image across a surface. This is then combined with tri-modal self-supervised contrastive (CLIP) alignment of audio, video, and fMRI modalities to enable the retrieval of visual and auditory stimuli from patterns of cortical activity (and vice-versa). We validate our approach on 7T task-fMRI data from 174 healthy participants engaged in the movie-watching experiment from the Human Connectome Project (HCP). Results show that it is possible to detect which movie clips an individual is watching purely from their brain activity, even for individuals and movies *not seen during training*. Further analysis of attention maps reveals that our model captures individual patterns of brain activity that reflect semantic and visual systems. This opens the door to future personalised simulations of brain function. Code \& pre-trained models will be made available at https://github.com/metrics-lab/sim.

The human cerebral cortex undergoes a complex developmental process during gestation, characterised by rapid cortical expansion and gyrification. This study investigates in vivo cortical growth trajectories using longitudinal MRI data from fetal and preterm cohorts. We employed anatomically constrained multimodal surface matching (aMSM) to quantify cortical surface area expansion and compare in-utero versus ex-utero cortical growth using cortical surface data from 22 to 44 weeks post-menstrual age (PMA), acquired as part of the Developing Human Connectome Project (dHCP). Our findings revealed distinct regional and temporal growth patterns during the 2nd and 3rdrd trimesters of healthy fetal cortical expansion. Ex utero brain development following preterm birth was shown to follow a modified trajectory compared to normal gestation, with potential implications for cortical organisation. Our methodology, combining biomechanically constrained surface registration with high quality fetal and neonatal imaging, provides a powerful framework for understanding early cortical development and deviations associated with preterm birth.

Preterm birth is a major cause of mortality and lifelong morbidity in childhood. Its complex and multifactorial origins limit the effectiveness of current clinical predictors and impede optimal care. In this study, a dual-branch deep learning architecture (PUUMA) was developed to predict gestational age (GA) at birth using T2* fetal MRI data from 295 pregnancies, encompassing a heterogeneous and imbalanced population. The model integrates both global whole-uterus and local placental features. Its performance was benchmarked against linear regression using cervical length measurements obtained by experienced clinicians from anatomical MRI and other Deep Learning architectures. The GA at birth predictions were assessed using mean absolute error. Accuracy, sensitivity, and specificity were used to assess preterm classification. Both the fully automated MRI-based pipeline and the cervical length regression achieved comparable mean absolute errors (3 weeks) and good sensitivity (0.67) for detecting preterm birth, despite pronounced class imbalance in the dataset. These results provide a proof of concept for automated prediction of GA at birth from functional MRI, and underscore the value of whole-uterus functional imaging in identifying at-risk pregnancies. Additionally, we demonstrate that manual, high-definition cervical length measurements derived from MRI, not currently routine in clinical practice, offer valuable predictive information. Future work will focus on expanding the cohort size and incorporating additional organ-specific imaging to improve generalisability and predictive performance.

Understanding individual cortical development is essential for identifying deviations linked to neurodevelopmental disorders. However, current normative modelling frameworks struggle to capture fine-scale anatomical details due to their reliance on modelling data within a population-average reference space. Here, we present a novel framework for learning individual growth trajectories from biomechanically constrained, longitudinal, diffeomorphic image registration, implemented via a hierarchical network architecture. Trained on neonatal MRI data from the Developing Human Connectome Project, the method improves the biological plausibility of warps, generating growth trajectories that better follow population-level trends while generating smoother warps, with fewer negative Jacobians, relative to state-of-the-art baselines. The resulting subject-specific deformations provide interpretable, biologically grounded mappings of development. This framework opens new possibilities for predictive modeling of brain maturation and early identification of malformations of cortical development.

Gestational age plays a crucial role in neurodevelopment, and individuals born very preterm (VPT; <32 weeks’ gestation) are at elevated risk for cognitive, behavioural and psychiatric problems across the lifespan. Better understanding of the impact of very preterm birth on cortical maturation trajectories could inform mechanistic insights into the origins of these sequelae. Here we compared cortical morphology between VPT individuals and full-term controls in three datasets spanning birth, childhood and adulthood. We identified a consistent cortical signature of VPT birth, characterized by reduced surface area and cortical folding in the frontal, temporal, parietal and insular regions, which persisted across development. Furthermore, in two large infant cohorts, we found that this cortical signature was significantly associated with neonatal clinical factors and with poorer motor outcomes at follow-up, suggesting its potential as a neuroimaging marker for long-term neurodevelopmental risk. Given that early motor development plays a key role in shaping infants’ interactions with the environment and supporting later cognitive and behavioural development, our findings provide insights into the neurobiological pathways linking VPT birth to subsequent neurodevelopmental difficulties. ### Competing Interest Statement The authors have declared no competing interest. the National Institute for Health Research (NIHR) Medical Research Council, https://ror.org/03x94j517

Affective biases influence cognitive and emotional behaviour and play a significant role in major depressive disorder. We have shown that the NMDA antagonist and rapid-acting antidepressant, ketamine, selectively modulates affective biases, a neuropsychological effect which may underlie its efficacy. Clinical studies with different NMDA antagonists have found mixed effects but the reasons for these differences in efficacy are not understood. This study used a rat model of negative affective biases to investigate if different NMDA antagonists also differentially modulated affective biases. Dose-response experiments using the NMDA antagonists lanicemine, memantine, CP101,606, phencyclidine (PCP) and ephenidine, and ketamine metabolite, (2R,6R)-hydroxynorketamine (HNK), were performed to test their acute and sustained (24hrs) effects, and specificity of affective bias modulation. Our results showed that HNK, PCP, lanicemine, and ephenidine acutely attenuate negative biases. HNK, CP-101,606, and ephenidine's effects were sustained at 24hrs post-treatment with a positive bias observed for HNK and a tendency towards a positive affective bias for CP101,606 and ephenidine. Lanicemine's effects were sustained at 24hrs but only at the highest dose and PCP and memantine had no effects. Considering these findings in the context of clinical observations, we suggest that the ability to induce a sustained modulation of affective biases corresponds with therapeutic effects. The differences in efficacy observed with NMDA antagonists may be related to their ion trapping properties and those with very high and low ion trapping properties are less effective than those with more moderate ion trapping effects such as ketamine or ephenidine or the subunit selective antagonist, CP101,606. ### Competing Interest Statement ESJR has obtained research funding from Boehringer Ingelheim, Compass Pathways plc, Eli Lilly, IRLab Therapeutics, MSD, Pfizer and Small Pharma. ESJR has been paid as a consultant or invited speaker by Compass Pathways, Pangea Botanicals and Charles River. BH and RA are currently employed by Boehringer Ingelheim GmbH & Co. KG. The authors declare no conflict of interest.

Importance:A leading cause of surgically remediable, drug-resistant focal epilepsy is focal cortical dysplasia (FCD). FCD is challenging to visualize and often considered magnetic resonance imaging (MRI) negative. Existing automated methods for FCD detection are limited by high numbers of false-positive predictions, hampering their clinical utility. Objective:To evaluate the efficacy and interpretability of graph neural networks in automatically detecting FCD lesions on MRI scans. Design, Setting, and Participants:In this multicenter diagnostic study, retrospective MRI data were collated from 23 epilepsy centers worldwide between 2018 and 2022, as part of the Multicenter Epilepsy Lesion Detection (MELD) Project, and analyzed in 2023. Data from 20 centers were split equally into training and testing cohorts, with data from 3 centers withheld for site-independent testing. A graph neural network (MELD Graph) was trained to identify FCD on surface-based features. Network performance was compared with an existing algorithm. Feature analysis, saliencies, and confidence scores were used to interpret network predictions. In total, 34 surface-based MRI features and manual lesion masks were collated from participants, 703 patients with FCD-related epilepsy and 482 controls, and 57 participants were excluded during MRI quality control. Main Outcomes and Measures:Sensitivity, specificity, and positive predictive value (PPV) of automatically identified lesions. Results:In the test dataset, the MELD Graph had a sensitivity of 81.6% in histopathologically confirmed patients seizure-free 1 year after surgery and 63.7% in MRI-negative patients with FCD. The PPV of putative lesions from the 260 patients in the test dataset (125 female [48%] and 135 male [52%]; mean age, 18.0 [IQR, 11.0-29.0] years) was 67% (70% sensitivity; 60% specificity), compared with 39% (67% sensitivity; 54% specificity) using an existing baseline algorithm. In the independent test cohort (116 patients; 62 female [53%] and 54 male [47%]; mean age, 22.5 [IQR, 13.5-27.5] years), the PPV was 76% (72% sensitivity; 56% specificity), compared with 46% (77% sensitivity; 47% specificity) using the baseline algorithm. Interpretable reports characterize lesion location, size, confidence, and salient features. Conclusions and Relevance:In this study, the MELD Graph represented a state-of-the-art, openly available, and interpretable tool for FCD detection on MRI scans with significant improvements in PPV. Its clinical implementation holds promise for early diagnosis and improved management of focal epilepsy, potentially leading to better patient outcomes.

Preterm birth disrupts the typical trajectory of cortical neurodevelopment, increasing the risk of cognitive and behavioral difficulties. However, outcomes vary widely, posing a significant challenge for early prediction. To address this, individualized simulation offers a promising solution by modeling subject-specific neurodevelopmental trajectories, enabling the identification of subtle deviations from normative patterns that might act as biomarkers of risk. While generative models have shown potential for simulating neurodevelopment, prior approaches often struggle to preserve subject-specific cortical folding patterns or to reproduce region-specific morphological variations. In this paper, we present a novel graph-diffusion network that supports controllable simulation of cortical maturation. Using cortical surface data from the developing Human Connectome Project (dHCP), we demonstrate that the model maintains subject-specific cortical morphology while modeling cortical maturation sufficiently well to fool an independently trained age regression network, achieving a prediction accuracy of 0.85 ± 0.62.

This paper introduces GeoMorph, a novel geometric deep-learning framework designed for image registration of cortical surfaces. The registration process consists of two main steps. First, independent feature extraction is performed on each input surface using graph convolutions, generating low-dimensional feature representations that capture important cortical surface characteristics. Subsequently, features are registered in a deep-discrete manner to optimize the overlap of common structures across surfaces by learning displacements of a set of control points. To ensure smooth and biologically plausible deformations, we implement regularization through a deep conditional random field implemented with a recurrent neural network. Experimental results demonstrate that GeoMorph surpasses existing deep-learning methods by achieving improved alignment with smoother deformations. Furthermore, GeoMorph exhibits competitive performance compared to classical frameworks. Such versatility and robustness suggest strong potential for various neuroscience applications.

Cortical folding of brain is widely regarded as an interplay between genetic programming and biomechanical forces, closely linked to cytoarchitectonic regionalisation. Abnormal folding patterns are frequently observed in neurodevelopmental conditions and psychiatric disorders. However, significant inter-individual variability of secondary and tertiary folds obscures detection of shape biomarkers and confounds investigation of folding-functional relationships. Here we investigate cortical folding heterogeneity at fine scale, using novel hierarchical surface registration (MSM-HT) to parse cortical folding patterns into a representative family of distinct anatomical templates. By applying this technique both to young adults from the Human Connectome Project (HCP) and neonates in the Developing HCP and Brain Imaging in Babies (BIBS) cohorts, we identify and characterize common lobe-wise folding patterns: observing consistency across both age groups, with neonatal samples showing less variation. Crucially, we highlight significant hemispheric asymmetry within the temporal lobe for both adults and neonates, and show that improved correspondence of shape does not translate to improved areal correspondence, affirming previous studies that have pointed to dissociation of folding and functional organisation. This study provides a critical step towards understanding brain asymmetry and complex relationships between folding and function, offering a robust framework to generalise the uncovered cortical folding motifs across datasets and developmental stages. ### Competing Interest Statement The authors have declared no competing interest.

Antidepressant-induced apathy syndrome is reported in a high number of patients. It is characterised by loss of motivation for daily activities and emotional blunting. It has a negative impact on quality of life and treatment outcome, yet the changes in underlying neurobiology driving this syndrome remain unclear. To begin to address this, a comprehensive understanding of how different classes of antidepressant treatment impact on behaviours relevant to apathy is critical. Rodent motivation for reward is commonly assessed using effort-based operant conditioning paradigms such as the Effort for Reward task. However, motivation to perform spontaneous/innate behaviours may provide additional insight into changes in behaviour reflective of daily activities. We tested the acute and chronic effects of antidepressants on the Effort for Reward task, and the spontaneous/innate Effort-Based Forage task. Acute treatment revealed important divergence in drug effect between tasks, where selective serotonin reuptake inhibitor (SSRI)/serotonin and noradrenaline reuptake inhibitor (SNRI) treatment impaired foraging behaviour in the Effort Based Forage task, but enhanced high-effort, high-value reward responding in the Effort for Reward task. Treatment with a noradrenaline reuptake inhibitor (NRI) or multimodal agent impaired foraging behaviour but did not affect high reward responding in the Effort for Reward task. Conversely, chronic treatment with an SSRI but not SNRI enhanced motivated foraging behaviour but led to a general impairment in Effort for Reward task output. Together, these data demonstrate that SSRI treatment induces opposing effects on conditioned versus innate motivation which may have significant translational relevance when interpreting drug effect. Further, these behavioural effects differ depending on whether antidepressants are acutely or chronically administered.

We describe how multiple linear layers and non-linear activation functions are combined to create deep neural network architectures. We cover training of neural networks using backpropagation. We show a complete Pytorch deep learning solution for a real world biomedical problem to predict age from structural brain connectivity in newborn babies.

We developed a computational pipeline (now provided as a resource) for measuring morphological similarity between cortical surface sulci to construct a sulcal phenotype network (SPN) from each magnetic resonance imaging (MRI) scan in an adult cohort (n = 34,725; 45-82 years). Networks estimated from pairwise similarities of 40 sulci on 5 morphological metrics comprised two clusters of sulci, represented also by the bimodal distribution of sulci on a linear-to-complex dimension. Linear sulci were more heritable and typically located in unimodal cortex, and complex sulci were less heritable and typically located in heteromodal cortex. Aligning these results with an independent fetal brain MRI cohort (n = 228; 21-36 gestational weeks), we found that linear sulci formed earlier, and the earliest and latest-forming sulci had the least between-adult variation. Using high-resolution maps of cortical gene expression, we found that linear sulcation is mechanistically underpinned by trans-sulcal gene expression gradients enriched for developmental processes.

We introduce basic machine learning concepts, provide Scikit-learn tutorial, and teach the reader how to train and evaluate machine learning models. We cover regression, classification, clustering and dimensionality reduction. We also discuss overfitting, cross-validation and evaluation using test set.

How rapid-acting antidepressants (RAADs), such as ketamine, induce immediate and sustained improvements in mood in patients with major depressive disorder (MDD) is poorly understood. A core feature of MDD is the prevalence of cognitive processing biases associated with negative affective states, and the alleviation of negative affective biases may be an index of response to drug treatment. Here, we used an affective bias behavioral test in rats, based on an associative learning task, to investigate the effects of RAADs. To generate an affective bias, animals learned to associate two different digging substrates with a food reward in the presence or absence of an affective state manipulation. A choice between the two reward-associated digging substrates was used to quantify the affective bias generated. Acute treatment with the RAADs ketamine, scopolamine, or psilocybin selectively attenuated a negative affective bias in the affective bias test. Low, but not high, doses of ketamine and psilocybin reversed the valence of the negative affective bias 24 hours after RAAD treatment. Only treatment with psilocybin, but not ketamine or scopolamine, led to a positive affective bias that was dependent on new learning and memory formation. The relearning effects of ketamine were dependent on protein synthesis localized to the rat medial prefrontal cortex and could be modulated by cue reactivation, consistent with experience-dependent neural plasticity. These findings suggest a neuropsychological mechanism that may explain both the acute and sustained effects of RAADs, potentially linking their effects on neural plasticity with affective bias modulation in a rodent model.