High-order Connectomic Manifold Learning for Autistic Brain State Identification

Brain Imaging and Behavior

2019

Self Cite

Decoding how intelligence is engrained in the human brain construct is vital in the understanding of particular neurological disorders. While the majority of existing studies focus on characterizing intelligence in neurotypical (NT) brains, investigating how neural correlates of intelligence scores are altered by atypical neurodevelopmental disorders, such as Autism Spectrum Disorders (ASD), is almost absent. To help fill this gap, we use a connectome-based predictive model (CPM) to predict intelligence scores from functional connectome data, derived from resting-state functional magnetic resonance imaging (rsfMRI). The utilized model learns how to select the most significant positive and negative brain connections, independently, to predict the target intelligence scores in NT and ASD populations, respectively. In the first step, using leave-one-out cross-validation we train a linear regressor robust to outliers to identify functional brain connections that best predict the target intelligence score (p − value < 0.01). Next, for each training subject, positive (respectively negative) connections are summed to produce single-subject positive (respectively negative) summary values. These are then paired with the target training scores to train two linear regressors: (a) a positive model which maps each positive summary value to the subject score, and (b) a negative model which maps each negative summary value to the target score. In the testing stage, by selecting the same connections for the left-out testing subject, we compute their positive and negative summary values, which are then fed to the trained negative and positive models for predicting the target score. This framework was applied to NT and ASD populations independently to identify significant functional connections coding for full-scale and verbal intelligence quotients in the brain.

Section: Discussionmentioning

confidence: 99%

Predicting full-scale and verbal intelligence scores from functional Connectomic data in individuals with autism Spectrum disorder

Dryburgh

McKenna

Brain Imaging and Behavior

2019

Self Cite

“…Representing a shape as a multi-directional varifold has several appealing properties. First, it captures complex shape patterns through exploring its local neighborhood in comparison to conventional cortical measures (e.g., cortical thickness) used in [4, 5]. Second, it allows to perform shape matching without the need to establish point-to-point correspondence between two shapes.…”

Section: Estimation Of Shape and Growth Brain Network Atlasesmentioning

confidence: 99%

“…However, while providing tools for quantifying connections between different neuroanatomical structures, conventional connectomics tools are not particularly designed to investigate the morphology (or shape) of the brain and its dynamic changes with time (e.g., cortical growth and cortical atrophy). Specifically, ‘shape-to-shape’ connections, where we measure the morphological similarity between the brain structure shape and another brain structure shape, are rarely explored in the neuroscience state-of-the-art –with the exception of recent works introducing cortical morphological networks for brain disorder diagnosis [4, 5]. Previous works showed that brain morphology can be affected by different psychiatric disorders.…”

Section: Introductionmentioning

confidence: 99%

Estimation of shape and growth brain network atlases for connectomic brain mapping in developing infants

2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)

Lin

et al. 2018

Self Cite

In vivo brain connectomics have heavily relied on using functional and diffusion Magnetic Resonance Imaging (MRI) modalities to examine functional and structural relationships between pairs of anatomical regions in the brain. However, research work on brain morphological (i.e., shape-to-shape) connections, which can be derived from T1-w and T2-w MR images, in both typical and atypical development or ageing is very scarce. Furthermore, the brain cannot be only regarded as a static shape, since it is a dynamic complex system that changes at functional, structural and morphological levels. Hence, examining the ‘connection’ between brain shape and its changes with time (e.g., growth) may help advance our understanding of connectomic brain dynamics as well as disorders that may affect it. To address these limitations, we unprecedentedly introduce two population-based shape and growth connectivity analysis tools that further extend the field of connectomics to brain morphology and dynamics: the morphome and the kinectome. Specifically, for a population of anatomically labelled shapes, the morphome identifies a network of anatomical shape regions that are connected when morphologically similar at a single timepoint, whereas the kinectome identifies anatomical shape regions that elicit similar evolution dynamics across successive timepoints. These proposed generic tools can be easily invested to examine how a baseline shape influences its deformation trajectory at later timepoints using any longitudinal shape data. We evaluated these tools on 23 infants, with right and left cortical surfaces reconstructed at birth, 3, 6, 9 and 12 months of age. Investigating the relationship between the neonatal morphome and the postnatal kinectome (from birth to 1 year of age) gave insights into brain connectivity at birth and how it develops over time.

“…However, existing works mainly used functional networks (derived from rsfMRI) and structural networks (derived from dMRI). These exclude the recent landmark works [4][5][6], which devised morphological brain networks (MBN) for mapping morphological 'connections' in the cortex. Basically, an MBN is generated by measuring the difference in morphology between two cortical regions based on a specific cortical attribute (e.g., sulcal depth).…”

Section: Introductionmentioning

confidence: 99%

Joint Correlational and Discriminative Ensemble Classifier Learning for Dementia Stratification Using Shallow Brain Multiplexes

Raeper

Lisowska

Lecture Notes in Computer Science

2018

Self Cite

The demented brain wiring undergoes several changes with dementia progression. However, in early dementia stages, particularly early mild cognitive impairment (eMCI), these remain challenging to spot. Hence, developing accurate diagnostic techniques for eMCI identification is critical for early intervention to prevent the onset of Alzheimer's Disease (AD). There is a large body of machine-learning based research developed for classifying different brain states (e.g., AD vs MCI). These works can be fundamentally grouped into two categories. The first uses correlational methods, such as canonical correlation analysis (CCA) and its variants, with the aim to identify most correlated features for diagnosis. The second includes discriminative methods, such as feature selection methods and linear discriminative analysis (LDA) and its variants to identify brain features that distinguish between two brain states. However, existing methods examine these correlational and discriminative brain data independently, which overlooks the complementary information provided by both techniques, which could prove to be useful in the classification of patients with dementia. On the other hand, how early dementia affects cortical brain connections in morphology remains largely unexplored. To address these limitations, we propose a joint correlational and discriminative ensemble learning framework for eMCI diagnosis that leverages a novel brain network representation, derived from the cortex. Specifically, we devise 'the shallow convolutional brain multiplex' (SCBM), which not only measures the similarity in morphology between pairs of brain regions, but also encodes the relationship between two morphological brain networks. Then, we represent each individual brain using a set of SCBMs, which are used to train joint ensemble CCA-SVM and LDA-based classifier. Our framework outperformed several state-of-the-art methods by 3-7% including independent correlational and discriminative methods.