Detection of Abnormal Folding Patterns with Unsupervised Deep Generative Models

Guillon, Louise; Cagna, Bastien; Dufumier, Benoit; Chavas, Joël; Rivière, Denis; Mangin, Jean François

doi:10.1007/978-3-030-87586-2_7

Cited by 5 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fourth, because ECT is part of the recommended treatment for catatonia [31,32], schizophrenia patients with catatonia received significantly more ECT treatment compared to schizophrenia patients without catatonia. ECT can induce local changes of cortical volume [67], however, it could unlikely affect sulcal patterns which are nonplastic, very stable [61,68], markers of the cortex anatomy. Fifth, we used the BFCRS, the current gold standard scale for assessing catatonia and its severity [33].…”

Section: Discussionmentioning

confidence: 99%

Orbitofrontal sulcal patterns in catatonia

Moyal,

Haroche,

Attali

et al. 2023

Eur. Psychiatr.

View full text Add to dashboard Cite

Background Catatonia is a psychomotor syndrome frequently observed in disorders with neurodevelopmental impairments, including psychiatric disorders such as schizophrenia. The orbitofrontal cortex (OFC) has been repeatedly associated with catatonia. It presents with an important interindividual morphological variability, with three distinct H-shaped sulcal patterns, types I, II, and III, based on the continuity of the medial and lateral orbital sulci. Types II and III have been identified as neurodevelopmental risk factors for schizophrenia. The sulcal pattern of the OFC has never been investigated in catatonia despite the role of the OFC in the pathophysiology and the neurodevelopmental component of catatonia. Methods In this context, we performed a retrospective analysis of the OFC sulcal pattern in carefully selected homogeneous and matched subgroups of schizophrenia patients with catatonia (N = 58) or without catatonia (N = 65), and healthy controls (N = 82). Results Logistic regression analyses revealed a group effect on OFC sulcal pattern in the left (χ 2 = 18.1; p < .001) and right (χ 2 = 28.3; p < .001) hemispheres. Catatonia patients were found to have more type III and less type I in both hemispheres compared to healthy controls and more type III on the left hemisphere compared to schizophrenia patients without catatonia. Conclusion Because the sulcal patterns are indirect markers of early brain development, our findings support a neurodevelopmental origin of catatonia and may shed light on the pathophysiology of this syndrome.

show abstract

Section: Discussionmentioning

confidence: 99%

Orbitofrontal sulcal patterns in catatonia

Moyal,

Haroche,

Attali

et al. 2023

Eur. Psychiatr.

View full text Add to dashboard Cite

show abstract

“…Moreover, we added a decoder to SimCLR and analyzed SimCLR and β-VAE reconstructions to recover folding patterns. Last, we proposed a preprocessing based on a mask, enabling to focus on the region of interest, while avoiding the disadvantages of parallelepipedic bounding boxes used in [12].…”

Section: Discussionmentioning

confidence: 99%

“…However, characterizing the full diversity of folding patterns remains out of reach for these automatic geometric methods. Unsupervised deep learning methods is a natural next step: they have been used for detecting anomalies in folding shapes [12], but they have not been used yet to characterize the normal inter-individual variability of folding patterns.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Representation Learning of Cingulate Cortical Folding Patterns

Chavas

Guillon

Pascucci

et al. 2022

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

The human cerebral cortex is folded, making sulci and gyri over the whole cortical surface. Folding presents a very high inter-subject variability, and some neurodevelopmental disorders are correlated to local folding structures, named folding patterns. However, it is tough to characterize these patterns manually or semi-automatically using geometric distances. Here, we propose a new methodology to identify typical folding patterns. We focus on the cingulate region, known to have a clinical interest, using so-called skeletons (3D representation of folding patterns). We compare two models, β − V AE and SimCLR, in an unsupervised setting to learn a relevant representation of these patterns. We add a decoder to SimCLR to be able to analyse latent space. Specifically, we leverage the data augmentations used in SimCLR to propose a novel kind of augmentations based on folding topology. We then apply a clustering on the latent space. Cluster folding averages, interpolation in the latent space and reconstructions reveal new pattern structures. This structured representation shows that unsupervised learning can help in the discovery of still unknown patterns. We will gain further insights into folding patterns by using new priors in the unsupervised algorithms and integrating other brain data modalities. Code and experiments are available at github.com/neurospin-projects/2021 jchavas lguillon deepcingulate.

show abstract

“…Recently, β-VAE 6 and SimCLR 7 models have been trained to build a latent space for folding representation. [8][9][10] However, these studies have not used a manually labeled dataset for hyperparameter tuning and evaluation.…”

Section: Introductionmentioning

confidence: 99%

Optimizing contrastive learning for cortical folding pattern detection

Gaudin,

Guillon,

Fischer

et al. 2024

Medical Imaging 2024: Image Processing

View full text Add to dashboard Cite

The human cerebral cortex has many bumps and grooves called gyri and sulci. Even though there is a high interindividual consistency for the main cortical folds, this is not the case when we examine the exact shapes and details of the folding patterns. Because of this complexity, characterizing the cortical folding variability and relating them to subjects' behavioral characteristics or pathologies is still an open scientific problem. Classical approaches include labeling a few specific patterns, either manually or semi-automatically, based on geometric distances, but the recent availability of MRI image datasets of tens of thousands of subjects makes modern deep-learning techniques particularly attractive. Here, we build a self-supervised deep-learning model to detect folding patterns in the cingulate region. We train a contrastive self-supervised model (SimCLR) on both Human Connectome Project (1101 subjects) and UKBioBank (21070 subjects) datasets with topological-based augmentations on the cortical skeletons, which are topological objects that capture the shape of the folds. We explore several backbone architectures (convolutional network, DenseNet, and PointNet) for the SimCLR. For evaluation and testing, we perform a linear classification task on a database manually labeled for the presence of the "double-parallel" folding pattern in the cingulate region, which is related to schizophrenia characteristics. The best model, giving a test AUC of 0.76, is a convolutional network with 6 layers, a 10-dimensional latent space, a linear projection head, and using the branch-clipping augmentation. This is the first time that a self-supervised deep learning model has been applied to cortical skeletons on such a large dataset and quantitatively evaluated. We can now envisage the next step: applying it to other brain regions to detect other biomarkers. The GitHub repository is publicly available on https://github.com/neurospin-projects/2022 jchavas cingulate inhibitory control.

show abstract

Detection of Abnormal Folding Patterns with Unsupervised Deep Generative Models

Cited by 5 publications

References 22 publications

Orbitofrontal sulcal patterns in catatonia

Orbitofrontal sulcal patterns in catatonia

Unsupervised Representation Learning of Cingulate Cortical Folding Patterns

Optimizing contrastive learning for cortical folding pattern detection

Contact Info

Product

Resources

About