Novel age-dependent cortico-subcortical morphologic interactions predict fluid intelligence: A multi-cohort geometric deep learning study

Wu, Yunan; Besson, Pierre; Azcona, Emanuel A.; Bandt, S. Kathleen; Parrish, Todd B.; Breiter, Hans C.; Katsaggelos, Aggelos K.

doi:10.1101/2020.10.14.331199

Cited by 5 publications

(7 citation statements)

References 99 publications

(95 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The overall steps for the data preparation were detailed in our previous work ( Wu et al, 2020 ; Besson et al, 2021 ). All T1-weighted MRI were processed with Freesurfer (v6.0 2 ) using Northwestern University’s High Performance Computing Cluster (QUEST 3 ).…”

Section: Methodsmentioning

confidence: 99%

“…This, however, comes at the expense of using special convolutional operators capable of handling data mapped on graphs instead of traditional regular grids as it is the case with 2D or 3D images. For this purpose, and similarly to our previous work ( Wu et al, 2020 ; Besson et al, 2021 ), we used the graph convolutional layers introduced in Defferrard et al (2016) . In brief, this approach allows convolution filters to be learnt on unstructured data such as graphs using finite support recursive Chebychev filters applied to underlying Laplacian matrix.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Geometric deep learning reveals a structuro-temporal understanding of healthy and pathologic brain aging

Besson

Rogalskı

Gill

et al. 2022

Front. Aging Neurosci.

Self Cite

View full text Add to dashboard Cite

BackgroundBrain age has historically been investigated primarily at the whole brain level. The ability to deconstruct the brain into its composite parts and explore brain age at the sub-structure level offers unique advantages. These include the exploration of dynamic and interconnected relationships between different brain structures in healthy and pathologic aging. To achieve this, individual brain structures can be rendered as surface representations on which morphologic analysis is carried out. Combining the advantages of deep learning with the strengths of surface analysis, we investigate the aging process at the individual structure level with the hypothesis being that pathologic aging does not uniformly affect the aging process of individual structures.MethodsMRI data, age at scan time and diagnosis of dementia were collected from seven publicly available data repositories. The data from 17,440 unique subjects were collected, representing a total of 26,276 T1-weighted MRI accounting for longitudinal acquisitions. Surfaces were extracted for the cortex and seven subcortical structures. Deep learning networks were trained to estimate a subject’s age either using several structures together or a single structure. We conducted a cross-sectional analysis to assess the difference between the predicted and actual ages for all structures between healthy subjects, individuals with mild cognitive impairment (MCI) or Alzheimer’s disease dementia (ADD). We then performed a longitudinal analysis to assess the difference in the aging pace for each structure between stable healthy controls and healthy controls converting to either MCI or ADD.FindingsUsing an independent cohort of healthy subjects, age was well estimated for all structures. Cross-sectional analysis identified significantly larger predicted age for all structures in patients with either MCI and ADD compared to healthy subjects. Longitudinal analysis revealed varying degrees of involvement of individual subcortical structures for both age difference across groups and aging pace across time. These findings were most notable in the whole brain, cortex, hippocampus and amygdala.ConclusionAlthough similar patterns of abnormal aging were found related to MCI and ADD, the involvement of individual subcortical structures varied greatly and was consistently more pronounced in ADD patients compared to MCI patients.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Geometric deep learning reveals a structuro-temporal understanding of healthy and pathologic brain aging

Besson

Rogalskı

Gill

et al. 2022

Front. Aging Neurosci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Spectral methods, on the other hand, define (approximated) convolutions through the generalised Graph Laplacian, allowing them to describe both structured and unstructured data, and adapt to many different surface shapes. Accordingly, graph spectral methods have again found applications in cortical segmentation (Gopinath et al, 2019;He et al, 2020;Cucurull (Azcona et al, 2020), and for mapping cognitive function (Wu et al, 2020;Ribeiro et al, 2020;Liu et al, 2020). In other work, the rotationequivariant and expressive spectral filters of S2CNN (Cohen et al, 2018) were used successfully to classify Alzheimer's disease from mild cognitive impairment, using downsampled spherical cortical-imaging data (Barbaroux et al, 2020).…”

Section: Background and Related Workmentioning

confidence: 99%

Benchmarking Geometric Deep Learning for Cortical Segmentation and Neurodevelopmental Phenotype Prediction

Fawaz

Williams

Alansary

et al. 2021

Preprint

View full text Add to dashboard Cite

The emerging field of geometric deep learning extends the application of convolutional neural networks to irregular domains such as graphs, meshes and surfaces. Several recent studies have explored the potential for using these techniques to analyse and segment the cortical surface. However, there has been no comprehensive comparison of these approaches to one another, nor to existing Euclidean methods, to date. This paper benchmarks a collection of geometric and traditional deep learning models on phenotype prediction and segmentation of sphericalised neonatal cortical surface data, from the publicly available Developing Human Connectome Project (dHCP). Tasks include prediction of postmenstrual age at scan, gestational age at birth and segmentation of the cortical surface into anatomical regions defined by the M-CRIB-S atlas. Performance was assessed not only in terms of model precision, but also in terms of network dependence on image registration, and model interpretation via occlusion. Networks were trained both on sphericalised and anatomical cortical meshes. Findings suggest that the utility of geometric deep learning over traditional deep learning is highly task-specific, which has implications for the design of future deep learning models on the cortical surface. The code, and instructions for data access, are available from https://github.com/Abdulah-Fawaz/Benchmarking-Surface-DL.

show abstract

“…The shared features on the vertices of the input meshes to our models are the corresponding x, y, z coordinates (3 features total) of each vertex in the corresponding subject's native 3D space. The meshes used in our study are all registered to a common mesh template of the subcortical structures utilized in our prior work (Azcona et al, 2020;Besson et al, 2020;Wu et al, 2020a). The meshes in this study use a shared topology (same number of vertices/edges).…”

Section: Mesh Notationmentioning

confidence: 99%

“…In our prior work, class activation maps (CAMs) were extracted from our discriminative model to highlight areas, directly onto surfaces, that led to true positive (TP) predictions in AD binary classification. Wu et al (2020a) also use this mesh adaptation of Grad-CAM to highlight the areas of the cortex and subcortical structures that were most indicative for predicting fluid intelligence in children and adults.…”

Section: Grad-cam Mesh Adaptationmentioning

confidence: 99%

Analyzing Brain Morphology in Alzheimer’s Disease Using Discriminative and Generative Spiral Networks

Azcona

Besson

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Several patterns of atrophy have been identified and strongly related to Alzheimer's disease (AD) pathology and its progression. Morphological changes in brain shape have been identified up to ten years before clinical diagnoses of AD, making its early diagnosis more desirable. We propose novel geometric deep learning frameworks for the analysis of brain shape in the context of neurodegeneration caused by AD. Our deep neural networks learn low-dimensional shape descriptors of multiple neuroanatomical structures, instead of handcrafted features for each structure. A discriminative network using spiral convolution on 3D meshes is constructed for the in-vivo binary classification of AD from healthy controls (HCs) using a fast and efficient "spiral" convolution operator on 3D triangular mesh surfaces of human brain subcortical structures extracted from T1-weighted magnetic resonance imaging (MRI). Our network architecture consists of modular learning blocks using residual connections to improve overall classifier performance. In this work: (1) a discriminative network is used to analyze the efficacy of disease classification using input data from multiple brain structures and compared to using a single hemisphere or a single structure. It also outperforms prior work using spectral graph convolution on the same the same tasks, as well as alternative methods that operate on intermediate point cloud representations of 3D shapes. (2) Additionally, visual interpretations for regions on the surface of brain structures that are associated to true positive AD predictions are generated and fall in accordance with the current reports on the structural localization of pathological changes associated to AD. (3) A conditional generative network is also implemented to analyze the effects of phenotypic priors given to the model (i.e. AD diagnosis) in generating subcortical structures. The generated surface meshes by our model indicate learned morphological differences in the presence of AD that agrees with the current literature on patterns of atrophy associated to the disease. In particular, our inference results demonstrate an overall reduction in subcortical mesh volume and surface area in the presence of AD, especially in the hippocampus. The low-dimensional shape descriptors obtained by our generative model are also evaluated in our discriminative baseline comparisons versus our discriminative network and the alternative shape-based approaches.

show abstract

Novel age-dependent cortico-subcortical morphologic interactions predict fluid intelligence: A multi-cohort geometric deep learning study

Cited by 5 publications

References 99 publications

Geometric deep learning reveals a structuro-temporal understanding of healthy and pathologic brain aging

Geometric deep learning reveals a structuro-temporal understanding of healthy and pathologic brain aging

Benchmarking Geometric Deep Learning for Cortical Segmentation and Neurodevelopmental Phenotype Prediction

Analyzing Brain Morphology in Alzheimer’s Disease Using Discriminative and Generative Spiral Networks

Contact Info

Product

Resources

About