Phylogenetic differences in the morphology and shape of the central sulcus in great apes and humans: implications for the evolution of motor functions.

The central sulcus divides the primary motor and somatosensory cortices in many anthropoid primate brains. Differences exist in the surface area and depth of the central sulcus along the dorso-ventral plane in great apes and humans compared to other primate species. Within hominid species, there are variations in the depth and aspect of their hand motor area, or knob, within the precentral gyrus. In this study, we used post-image analyses on magnetic resonance images to characterize the central sulcus shape of humans, chimpanzees (Pan troglodytes), gorillas (Gorilla gorilla), and orangutans (Pongo pygmaeus and Pongo abelii). Using these data, we examined the morphological variability of central sulcus in hominids, focusing on the hand region, a significant change in human evolution. We show that the central sulcus shape differs between great ape species, but all show similar variations in the location of their hand knob. However, the prevalence of the knob location along the dorso-ventral plane and lateralization differs between species and the presence of a second ventral motor knob seems to be unique to humans. Humans and orangutans exhibit the most similar and complex central sulcus shapes. However, their similarities may reflect divergent evolutionary processes related to selection for different positional and habitual locomotor functions.

A framework to improve the alignment of individual cytoarchitectonic maps of the Julich-Brain atlas using cortical folding landmarks.

The segregation of the cortical mantle into cytoarchitectonic areas provides a structural basis for the specialization of different brain regions. In vivo neuroimaging experiments can be linked to this postmortem cytoarchitectonic parcellation via Julich-Brain. This atlas embeds probabilistic maps that account for inter-individual variability in the localization of cytoarchitectonic areas in the reference spaces targeted by spatial normalization. We built a framework to improve the alignment of architectural areas across brains using cortical folding landmarks. This framework, initially designed for in vivo imaging, was adapted to postmortem histological data. We applied this to the first 14 brains used to establish the Julich-Brain atlas to infer a refined atlas with more focal probabilistic maps. The improvement achieved is significant in the primary regions and some of the associative areas. This framework also provides a tool for exploring the relationship between cortical folding patterns and cytoarchitectonic areas in different cortical regions to establish new landmarks in the remainder of the cortex.

Exploring the potential of representation and transfer learning for anatomical neuroimaging: Application to psychiatry.

The perspective of personalized medicine for brain disorders requires efficient learning models for anatomical neuroimaging-based prediction of clinical conditions. There is now a consensus on the benefit of deep learning (DL) in addressing many medical imaging tasks, such as image segmentation. However, for single-subject prediction problems, recent studies yielded contradictory results when comparing DL with Standard Machine Learning (SML) on top of classical feature extraction. Most existing comparative studies were limited in predicting phenotypes of little clinical interest, such as sex and age, and using a single dataset. Moreover, they conducted a limited analysis of the employed image pre-processing and feature selection strategies. This paper extensively compares DL and SML prediction capacity on five multi-site problems, including three increasingly complex clinical applications in psychiatry namely schizophrenia, bipolar disorder, and Autism Spectrum Disorder (ASD) diagnosis. To compensate for the relative scarcity of neuroimaging data on these clinical datasets, we also evaluate three pre-training strategies for transfer learning from brain imaging of the general healthy population: self-supervised learning, generative modeling and supervised learning with age. Overall, we find similar performance between randomly initialized DL and SML for the three clinical tasks and a similar scaling trend for sex prediction. This was replicated on an external dataset. We also show highly correlated discriminative brain regions between DL and linear ML models in all problems. Nonetheless, we demonstrate that self-supervised pre-training on large-scale healthy population imaging datasets (N≈10k), along with Deep Ensemble, allows DL to learn robust and transferable representations to smaller-scale clinical datasets (N≤1k). It largely outperforms SML on 2 out of 3 clinical tasks both in internal and external test sets. These findings suggest that the improvement of DL over SML in anatomical neuroimaging mainly comes from its capacity to learn meaningful and useful abstract representations of the brain anatomy, and it sheds light on the potential of transfer learning for personalized medicine in psychiatry.

What do brain endocasts tell us? A comparative analysis of the accuracy of sulcal identification by experts and perspectives in palaeoanthropology.

Palaeoneurology is a complex field as the object of study, the brain, does not fossilize. Studies rely therefore on the (brain) endocranial cast (often named endocast), the only available and reliable proxy for brain shape, size and details of surface. However, researchers debate whether or not specific marks found on endocasts correspond reliably to particular sulci and/or gyri of the brain that were imprinted in the braincase. The aim of this study is to measure the accuracy of sulcal identification through an experiment that reproduces the conditions that palaeoneurologists face when working with hominin endocasts. We asked 14 experts to manually identify well-known foldings in a proxy endocast that was obtained from an MRI of an actual in vivo Homo sapiens head. We observe clear differences in the results when comparing the non-corrected labels (the original labels proposed by each expert) with the corrected labels. This result illustrates that trying to reconstruct a sulcus following the very general known shape/position in the literature or from a mean specimen may induce a bias when looking at an endocast and trying to follow the marks observed there. We also observe that the identification of sulci appears to be better in the lower part of the endocast compared to the upper part. The results concerning specific anatomical traits have implications for highly debated topics in palaeoanthropology. Endocranial description of fossil specimens should in the future consider the variation in position and shape of sulci in addition to using models of mean brain shape. Moreover, it is clear from this study that researchers can perceive sulcal imprints with reasonably high accuracy, but their correct identification and labelling remains a challenge, particularly when dealing with extinct species for which we lack direct knowledge of the brain.

Genetic determinants of individual variation in the superior temporal sulcus of chimpanzees (Pan troglodytes).

The superior temporal sulcus (STS) is a conserved fold that divides the middle and superior temporal gyri. In humans, there is considerable variation in the shape, folding pattern, lateralization, and depth of the STS that have been reported to be associated with social cognition and linguistic functions. We examined the role that genetic factors play on individual variation in STS morphology in chimpanzees. The surface area and depth of the STS were quantified in sample of 292 captive chimpanzees comprised of two genetically isolated population of individuals. The chimpanzees had been previously genotyped for AVPR1A and KIAA0319, two genes that play a role in social cognition and communication in humans. Single nucleotide polymorphisms in the KIAA0319 and AVPR1A genes were associated with average depth as well as asymmetries in the STS. By contrast, we found no significant effects of these KIA0319 and AVPR1A polymorphism on surface area and depth measures for the central sulcus. The overall findings indicate that genetic factors account for a small to moderate amount of variation in STS morphology in chimpanzees. These findings are discussed in the context of the role of the STS in social cognition and language in humans and their potential evolutionary origins.

Exploring the emergence of morphological asymmetries around the brain's Sylvian fissure: a longitudinal study of shape variability in preterm infants.

Brain folding patterns vary within the human species, but some folding properties are common across individuals, including the Sylvian fissure's inter-hemispheric asymmetry. Contrarily to the other brain folds (sulci), the Sylvian fissure develops through the process of opercularization, with the frontal, parietal, and temporal lobes growing over the insular lobe. Its asymmetry may be related to the leftward functional lateralization for language processing, but the time course of these asymmetries' development is still poorly understood. In this study, we investigated refined shape features of the Sylvian fissure and their longitudinal development in 71 infants born extremely preterm (mean gestational age at birth: 26.5 weeks) and imaged once before and once at term-equivalent age (TEA). We additionally assessed asymmetrical sulcal patterns at TEA in the perisylvian and inferior frontal regions, neighbor to the Sylvian fissure. While reproducing renowned strong asymmetries in the Sylvian fissure, we captured an early encoding of its main asymmetrical shape features, and we observed global asymmetrical shape features representative of a more pronounced opercularization in the left hemisphere, contrasting with the previously reported right hemisphere advance in sulcation around birth. This added novel insights about the processes governing early-life brain folding mechanisms, potentially linked to the development of language-related capacities.

In vivo mapping of the deep and superficial white matter connectivity in the chimpanzee brain.

Mapping the chimpanzee brain connectome and comparing it to that of humans is key to our understanding of similarities and differences in primate evolution that occurred after the split from their common ancestor around 6 million years ago. In contrast to studies on macaque species' brains, fewer studies have specifically addressed the structural connectivity of the chimpanzee brain and its comparison with the human brain. Most comparative studies in the literature focus on the anatomy of the cortex and deep nuclei to evaluate how their morphology and asymmetry differ from that of the human brain, and some studies have emerged concerning the study of brain connectivity among humans, monkeys, and apes. In this work, we established a new white matter atlas of the deep and superficial white matter structural connectivity in chimpanzees. In vivo anatomical and diffusion-weighted magnetic resonance imaging (MRI) data were collected on a 3-Tesla MRI system from 39 chimpanzees. These datasets were subsequently processed using a novel fiber clustering pipeline adapted to the chimpanzee brain, enabling us to create two novel deep and superficial white matter connectivity atlases representative of the chimpanzee brain. These atlases provide the scientific community with an important and novel set of reference data for understanding the commonalities and differences in structural connectivity between the human and chimpanzee brains. We believe this study to be innovative both in its novel approach and in mapping the superficial white matter bundles in the chimpanzee brain, which will contribute to a better understanding of hominin brain evolution.

Genetic variations within human gained enhancer elements affect human brain sulcal morphology.

The expansion of the cerebral cortex is one of the most distinctive changes in the evolution of the human brain. Cortical expansion and related increases in cortical folding may have contributed to emergence of our capacities for high-order cognitive abilities. Molecular analysis of humans, archaic hominins, and non-human primates has allowed identification of chromosomal regions showing evolutionary changes at different points of our phylogenetic history. In this study, we assessed the contributions of genomic annotations spanning 30 million years to human sulcal morphology measured via MRI in more than 18,000 participants from the UK Biobank. We found that variation within brain-expressed human gained enhancers, regulatory genetic elements that emerged since our last common ancestor with Old World monkeys, explained more trait heritability than expected for the left and right calloso-marginal posterior fissures and the right central sulcus. Intriguingly, these are sulci that have been previously linked to the evolution of locomotion in primates and later on bipedalism in our hominin ancestors.

Structural basis of envelope and phase intrinsic coupling modes in the cerebral cortex.

Intrinsic coupling modes (ICMs) can be observed in ongoing brain activity at multiple spatial and temporal scales. Two families of ICMs can be distinguished: phase and envelope ICMs. The principles that shape these ICMs remain partly elusive, in particular their relation to the underlying brain structure. Here we explored structure-function relationships in the ferret brain between ICMs quantified from ongoing brain activity recorded with chronically implanted micro-ECoG arrays and structural connectivity (SC) obtained from high-resolution diffusion MRI tractography. Large-scale computational models were used to explore the ability to predict both types of ICMs. Importantly, all investigations were conducted with ICM measures that are sensitive or insensitive to volume conduction effects. The results show that both types of ICMs are significantly related to SC, except for phase ICMs when using measures removing zero-lag coupling. The correlation between SC and ICMs increases with increasing frequency which is accompanied by reduced delays. Computational models produced results that were highly dependent on the specific parameter settings. The most consistent predictions were derived from measures solely based on SC. Overall, the results demonstrate that patterns of cortical functional coupling as reflected in both phase and envelope ICMs are both related, albeit to different degrees, to the underlying structural connectivity in the cerebral cortex.

Physical activity, biomarkers of brain pathologies and dementia risk: Results from the Memento clinical cohort.

INTRODUCTION: This study aims to examine whether physical activity moderates the association between biomarkers of brain pathologies and dementia risk. METHODS: From the Memento cohort, we analyzed 1044 patients with mild cognitive impairment, aged 60 and older. Self-reported physical activity was assessed using the International Physical Activity Questionnaire. Biomarkers of brain pathologies comprised medial temporal lobe atrophy (MTA), white matter lesions, and plasma amyloid beta (Aß)42/40 and phosphorylated tau181. Association between physical activity and risk of developing dementia over 5 years of follow-up, and interactions with biomarkers of brain pathologies were tested. RESULTS: Physical activity moderated the association between MTA and plasma Aß42/40 level and increased dementia risk. Compared to participants with low physical activity, associations of both MTA and plasma Aß42/40 on dementia risk were attenuated in participants with high physical activity. DISCUSSION: Although reverse causality cannot be excluded, this work suggests that physical activity may contribute to cognitive reserve. HIGHLIGHTS: Physical activity is an interesting modifiable target for dementia prevention. Physical activity may moderate the impact of brain pathology on dementia risk. Medial temporal lobe atrophy and plasma amyloid beta 42/40 ratio were associated with increased dementia risk especially in those with low level of physical activity.

Associations among hypertension, dementia biomarkers, and cognition: The MEMENTO cohort.

INTRODUCTION: Approximately 40% of dementia cases could be delayed or prevented acting on modifiable risk factors including hypertension. However, the mechanisms underlying the hypertension-dementia association are still poorly understood. METHODS: We conducted a cross-sectional analysis in 2048 patients from the MEMENTO cohort, a French multicenter clinic-based study of outpatients with either isolated cognitive complaints or mild cognitive impairment. Exposure to hypertension was defined as a combination of high blood pressure (BP) status and antihypertensive treatment intake. Pathway associations were examined through structural equation modeling integrating extensive collection of neuroimaging biomarkers and clinical data. RESULTS: Participants treated with high BP had significantly lower cognition compared to the others. This association was mediated by higher neurodegeneration and higher white matter hyperintensities load but not by Alzheimer's disease (AD) biomarkers. DISCUSSION: These results highlight the importance of controlling hypertension for prevention of cognitive decline and offer new insights on mechanisms underlying the hypertension-dementia association. HIGHLIGHTS: Paths of hypertension-cognition association were assessed by structural equation models. The hypertension-cognition association is not mediated by Alzheimer's disease biomarkers. The hypertension-cognition association is mediated by neurodegeneration and leukoaraiosis. Lower cognition was limited to participants treated with uncontrolled blood pressure. Blood pressure control could contribute to promote healthier brain aging.

Disruption of Conscious Access in Psychosis Is Associated with Altered Structural Brain Connectivity.

According to global neuronal workspace (GNW) theory, conscious access relies on long-distance cerebral connectivity to allow a global neuronal ignition coding for conscious content. In patients with schizophrenia and bipolar disorder, both alterations in cerebral connectivity and an increased threshold for conscious perception have been reported. The implications of abnormal structural connectivity for disrupted conscious access and the relationship between these two deficits and psychopathology remain unclear. The aim of this study was to determine the extent to which structural connectivity is correlated with consciousness threshold, particularly in psychosis. We used a visual masking paradigm to measure consciousness threshold, and diffusion MRI tractography to assess structural connectivity in 97 humans of either sex with varying degrees of psychosis: healthy control subjects (n = 46), schizophrenia patients (n = 25), and bipolar disorder patients with (n = 17) and without (n = 9) a history of psychosis. Patients with psychosis (schizophrenia and bipolar disorder with psychotic features) had an elevated masking threshold compared with control subjects and bipolar disorder patients without psychotic features. Masking threshold correlated negatively with the mean general fractional anisotropy of white matter tracts exclusively within the GNW network (inferior frontal-occipital fasciculus, cingulum, and corpus callosum). Mediation analysis demonstrated that alterations in long-distance connectivity were associated with an increased masking threshold, which in turn was linked to psychotic symptoms. Our findings support the hypothesis that long-distance structural connectivity within the GNW plays a crucial role in conscious access, and that conscious access may mediate the association between impaired structural connectivity and psychosis.

Disentangling the variability of the superficial white matter organization using regional-tractogram-based population stratification.

Each variation of the cortical folding pattern implies a particular rearrangement of the geometry of the fibers of the underlying white matter. While this rearrangement only impacts the ends of the long pathways, it may affect most of the trajectory of the short bundles. Therefore, mapping the short fibers of the human brain using diffusion-based tractography requires a dedicated strategy to overcome the variability of the folding patterns. In this paper, we propose a fiber-based stratification strategy splitting the population into homogeneous groups for disentangling the superficial white matter bundle organization. This strategy introduces a new refined fiber distance which includes angular considerations for inferring fine-grained atlases of the short bundles surrounding a specific sulcus and a subtractogram distance that quantifies the similitude between fiber sets of two different subjects. The stratification splits the population into groups with similar regional fiber organization using manifold learning. We first successfully test the hypothesis that the main source of variability of the regional fiber organization is the variability of the regional folding pattern. Then, in each group, we proceed with the automatic identification of the most stable bundles, at a higher granularity level than what can be achieved with the non-stratified whole population, enabling the disentanglement of the very variable configuration of the short fibers. Finally, the method searches for bundle correspondence across groups to build a population level atlas. As a proof of concept, the atlas refinement achieved by this strategy is illustrated for the fibers that surround the central sulcus and the superior temporal sulcus using the HCP dataset.

Superficial white matter bundle atlas based on hierarchical fiber clustering over probabilistic tractography data.

The study of short association fibers is still an incomplete task due to their higher inter-subject variability and the smaller size of this kind of fibers in comparison to known long association bundles. However, their description is essential to understand human brain dysfunction and better characterize the human brain connectome. In this work, we present a multi-subject atlas of short association fibers, which was computed using a superficial white matter bundle identification method based on fiber clustering. To create the atlas, we used probabilistic tractography from one hundred subjects from the HCP database, aligned with non-linear registration. The method starts with an intra-subject clustering of short fibers (30-85 mm). Based on a cortical atlas, the intra-subject cluster centroids from all subjects are segmented to identify the centroids connecting each region of interest (ROI) of the atlas. To reduce computational load, the centroids from each ROI group are randomly separated into ten subgroups. Then, an inter-subject hierarchical clustering is applied to each centroid subgroup, followed by a second level of clustering to select the most-reproducible clusters across subjects for each ROI group. Finally, the clusters are labeled according to the regions that they connect, and clustered to create the final bundle atlas. The resulting atlas is composed of 525 bundles of superficial short association fibers along the whole brain, with 384 bundles connecting pairs of different ROIs and 141 bundles connecting portions of the same ROI. The reproducibility of the bundles was verified using automatic segmentation on three different tractogram databases. Results for deterministic and probabilistic tractography data show high reproducibility, especially for probabilistic tractography in HCP data. In comparison to previous work, our atlas features a higher number of bundles and greater cortical surface coverage.

Association of APOE ε4 with cerebral gray matter volumes in non-demented older adults: The MEMENTO cohort study.

Data on 2,045 non-demented individuals with memory complaints were drawn from the Memento cohort study to examine the association between Apolipoprotein E ε4 allele (APOE4) and regional brain gray matter volumes. Linear regression was used to examine the association of APOE4 and measures of regional gray matter volumes in cross-sectional analysis and change therein using longitudinal analyses based on two brain MRI performed at baseline and at two-year follow-up. Overall, in analyses adjusted for age, sex, and intracranial volume, the presence of APOE4 was associated with lower total gray matter volume at baseline and with a higher atrophy rate over the follow-up. The hippocampus and entorhinal cortex were the two gray matter regions most associated with APOE4. Further adjustment for cardiovascular risk factors had little impact on these associations. There was an interaction between age, APOE4 status and total brain volume atrophy rate, with evidence of an earlier age at onset of atrophy in hippocampal volume in APOE4 carriers compared to non-carriers. Those results are in accordance with the role of medial temporal structures in the greater risk of dementia observed in people carrying the APOE4 allele.

Shape variability of the central sulcus in the developing brain: A longitudinal descriptive and predictive study in preterm infants.

Despite growing evidence of links between sulcation and function in the adult brain, the folding dynamics, occurring mostly before normal-term-birth, is vastly unknown. Looking into the development of cortical sulci in infants can give us keys to address fundamental questions: what is the sulcal shape variability in the developing brain? When are the shape features encoded? How are these morphological parameters related to further functional development? In this study, we aimed to investigate the shape variability of the developing central sulcus, which is the frontier between the primary somatosensory and motor cortices. We studied a cohort of 71 extremely preterm infants scanned twice using MRI - once around 30 weeks post-menstrual age (w PMA) and once at term-equivalent age, around 40w PMA -, in order to quantify the sulcus's shape variability using manifold learning, regardless of age-group or hemisphere. We then used these shape descriptors to evaluate the sulcus's variability at both ages and to assess hemispheric and age-group specificities. This led us to propose a description of ten shape features capturing the variability in the central sulcus of preterm infants. Our results suggested that most of these features (8/10) are encoded as early as 30w PMA. We unprecedentedly observed hemispheric asymmetries at both ages, and the one captured at term-equivalent age seems to correspond with the asymmetry pattern previously reported in adults. We further trained classifiers in order to explore the predictive value of these shape features on manual performance at 5 years of age (handedness and fine motor outcome). The central sulcus's shape alone showed a limited but relevant predictive capacity in both cases. The study of sulcal shape features during early neurodevelopment may participate to a better comprehension of the complex links between morphological and functional organization of the developing brain.

Brain simulation as a cloud service: The Virtual Brain on EBRAINS.

The Virtual Brain (TVB) is now available as open-source services on the cloud research platform EBRAINS (ebrains.eu). It offers software for constructing, simulating and analysing brain network models including the TVB simulator; magnetic resonance imaging (MRI) processing pipelines to extract structural and functional brain networks; combined simulation of large-scale brain networks with small-scale spiking networks; automatic conversion of user-specified model equations into fast simulation code; simulation-ready brain models of patients and healthy volunteers; Bayesian parameter optimization in epilepsy patient models; data and software for mouse brain simulation; and extensive educational material. TVB cloud services facilitate reproducible online collaboration and discovery of data assets, models, and software embedded in scalable and secure workflows, a precondition for research on large cohort data sets, better generalizability, and clinical translation.

Neuroimaging correlates of persistent fatigue in older adults: A secondary analysis from the Multidomain Alzheimer Preventive Trial (MAPT) trial.

OBJECTIVES: Fatigue has been suggested as a marker of biological aging. It seems plausible that this symptom might be associated with changes in brain health. The objective of this study was to examine the associations between persistent fatigue and neuroimaging correlates in a non-disease-specific population of community-dwelling older adults. METHODS: We performed a cross-sectional analysis using data from The Multidomain Alzheimer Preventive Trial (MAPT). We included 458 subjects. Persistent fatigue was defined as meeting exhaustion criterion of Fried frailty phenotype in two consecutive clinical visits six months apart between study baseline and one year. Brain imaging correlates, assessed by magnetic resonance imaging (MRI), were the outcomes. The associations between persistent fatigue and brain correlates were explored using mixed model linear regressions with random effect at the center level. RESULTS: The mean age of the participants was 74.8 ± 4 years old, and 63% of the subjects were women. Forty-seven participants (10%) exhibited a persistent fatigue profile. People with persistent fatigue were older compared to subjects without persistent fatigue (76.2 years ± 4.3 vs.74.7 ± 3.9 p = 0.009). Persistent fatigue was associated with higher white matter hyperintensity volume in the fully adjusted analysis. We did not find any cross-sectional association between persistent fatigue and sub-cortical volumes and global and regional cortical thickness. CONCLUSION: Persistent fatigue was cross-sectionnally associated with higher white matter hyperintensity volume in older adults. Further longitudinal studies, using an assessment tool specifically designed and validated for measuring fatigue, are needed to confirm our findings.

Plasma amyloid beta predicts conversion to dementia in subjects with mild cognitive impairment: The BALTAZAR study.

INTRODUCTION: Blood-based biomarkers are the next challenge for Alzheimer's disease (AD) diagnosis and prognosis. METHODS: Mild cognitive impairment (MCI) participants (N = 485) of the BALTAZAR study, a large-scale longitudinal multicenter cohort, were followed-up for 3 years. A total of 165 of them converted to dementia (95% AD). Associations of conversion and plasma amyloid beta (Aß)1-42 , Aß1-40 , Aß1-42 /Aß1-40 ratio were analyzed with logistic and Cox models. RESULTS: Converters to dementia had lower level of plasma Aß1-42 (37.1 pg/mL [12.5] vs. 39.2 [11.1] , P value = .03) and lower Aß1-42 /Aß1-40 ratio than non-converters (0.148 [0.125] vs. 0.154 [0.076], P value = .02). MCI participants in the highest quartile of Aß1-42 /Aß1-40 ratio (>0.169) had a significant lower risk of conversion (hazard ratio adjusted for age, sex, education, apolipoprotein E ε4, hippocampus atrophy = 0.52 (95% confidence interval [0.31-0.86], P value = .01). DISCUSSION: In this large cohort of MCI subjects we identified a threshold for plasma Aß1-42 /Aß1-40 ratio that may detect patients with a low risk of conversion to dementia within 3 years.

Automatic recognition of specific local cortical folding patterns.

The study of local cortical folding patterns showed links with psychiatric illnesses as well as cognitive functions. Despite the tools now available to visualize cortical folds in 3D, manually classifying local sulcal patterns is a time-consuming and tedious task. In fact, 3D visualization of folds helps experts to identify different sulcal patterns but fold variability is so high that the distinction between these patterns sometimes requires the definition of complex criteria, making manual classification difficult and not reliable. However, the assessment of the impact of these patterns on the functional organization of the cortex could benefit from the study of large databases, especially when studying rare patterns. In this paper, several algorithms for the automatic classification of fold patterns are proposed to allow morphological studies to be extended and confirmed on such large databases. Three methods are proposed, the first based on a Support Vector Machine (SVM) classifier, the second on the Scoring by Non-local Image Patch Estimator (SNIPE) approach and the third based on a 3D Convolution Neural Network (CNN). These methods are generic enough to be applicable to a wide range of folding patterns. They are tested on two types of patterns for which there is currently no method to automatically identify them: the Anterior Cingulate Cortex (ACC) patterns and the Power Button Sign (PBS). The two ACC patterns are almost equally present whereas PBS is a particularly rare pattern in the general population. The three models proposed achieve balanced accuracies of approximately 80% for ACC patterns classification and 60% for PBS classification. The CNN-based model is more interesting for the classification of ACC patterns thanks to its rapid execution. However, SVM and SNIPE-based models are more effective in managing unbalanced problems such as PBS recognition.