Functional connectome fingerprinting using shallow feedforward neural networks.
Ontology highlight
ABSTRACT:
SUBMITTER: Sarar G
PROVIDER: S-EPMC8053937 | biostudies-literature | 2021 Apr
REPOSITORIES: biostudies-literature
Publications
Functional connectome fingerprinting using shallow feedforward neural networks.
Sarar Gokce G Rao Bhaskar B Liu Thomas T
Proceedings of the National Academy of Sciences of the United States of America 20210401 15
Although individual subjects can be identified with high accuracy using correlation matrices computed from resting-state functional MRI (rsfMRI) data, the performance significantly degrades as the scan duration is decreased. Recurrent neural networks can achieve high accuracy with short-duration (72 s) data segments but are designed to use temporal features not present in the correlation matrices. Here we show that shallow feedforward neural networks that rely solely on the information in rsfMRI ...[more]
Similar Datasets
Project description:IntroductionConnectome analysis of the human brain's structural and functional architecture provides a unique opportunity to understand the organization of the brain's functional architecture. In previous studies, connectome fingerprinting using brain functional connectivity profiles as an individualized trait was able to predict an individual's neurocognitive performance from the Human Connectome Project (HCP) neurocognitive datasets.Materials and methodsIn the present study, we extend connectome fingerprinting from functional connectivity (FC) to structural connectivity (SC), identifying multiple relationships between behavioral traits and brain connectivity. Higher-order neurocognitive tasks were found to have a weaker association with structural connectivity than its functional connectivity counterparts.ResultsNeurocognitive tasks with a higher sensory footprint were, however, found to have a stronger association with structural connectivity than their functional connectivity counterparts. Language behavioral measurements had a particularly stronger correlation, especially between performance on the picture language test (Pic Vocab) and both FC (r = .28, p < .003) and SC (r = 0.27, p < .00077).ConclusionsAt the neural level, we found that the pattern of structural brain connectivity related to high-level language performance is consistent with the language white matter regions identified in presurgical mapping. We illustrate how this approach can be used to generalize the connectome fingerprinting framework to structural connectivity and how this can help understand the connections between cognitive behavior and the white matter connectome of the brain.
| S-EPMC7303390 | biostudies-literature
Project description:Functional magnetic resonance imaging (fMRI) studies typically collapse data from many subjects, but brain functional organization varies between individuals. Here we establish that this individual variability is both robust and reliable, using data from the Human Connectome Project to demonstrate that functional connectivity profiles act as a 'fingerprint' that can accurately identify subjects from a large group. Identification was successful across scan sessions and even between task and rest conditions, indicating that an individual's connectivity profile is intrinsic, and can be used to distinguish that individual regardless of how the brain is engaged during imaging. Characteristic connectivity patterns were distributed throughout the brain, but the frontoparietal network emerged as most distinctive. Furthermore, we show that connectivity profiles predict levels of fluid intelligence: the same networks that were most discriminating of individuals were also most predictive of cognitive behavior. Results indicate the potential to draw inferences about single subjects on the basis of functional connectivity fMRI.
| S-EPMC5008686 | biostudies-literature
Project description:Systematic changes have been observed in the functional architecture of the human brain with advancing age. However, functional connectivity (FC) is also a powerful feature to detect unique "connectome fingerprints," allowing identification of individuals among their peers. Although fingerprinting has been robustly observed in samples of young adults, the reliability of this approach has not been demonstrated across the lifespan. We applied the fingerprinting framework to the Cambridge Centre for Ageing and Neuroscience cohort (n = 483 aged 18 to 89 years). We found that individuals are "fingerprintable" (i.e., identifiable) across independent functional MRI scans throughout the lifespan. We observed a U-shape distribution in the strength of "self-identifiability" (within-individual correlation across modalities), and "others-identifiability" (between-individual correlation across modalities), with a decrease from early adulthood into middle age, before improving in older age. FC edges contributing to self-identifiability were not restricted to specific brain networks and were different between individuals across the lifespan sample. Self-identifiability was additionally associated with regional brain volume. These findings indicate that individual participant-level identification is preserved across the lifespan despite the fact that its components are changing nonlinearly.
| S-EPMC10473304 | biostudies-literature
Project description:The ability to uniquely characterize individual subjects based on their functional connectome (FC) is a key requirement for progress toward precision psychiatry. FC fingerprinting is increasingly studied in the neuroimaging community for this purpose, where a variety of approaches have been developed for effective FC fingerprinting. Recent independent studies showed that fingerprinting accuracy suffers at large sample sizes and when coarser parcellations are used for computing the FC. Quantifying this problem and understanding the reasons these factors impact fingerprinting accuracy is crucial to develop more accurate fingerprinting methods for large sample sizes. Part of the challenge in fingerprinting is that FC captures both generic and subject-specific information. A systematic approach for identifying subject-specific FC information is crucial for making progress in addressing the fingerprinting problem. In this study, we addressed three gaps in our understanding of the FC fingerprinting problem. First, we studied the joint effect of sample size and parcellation granularity. Second, we explained the reason for reduced fingerprinting accuracy with increased sample size and reduced parcellation granularity. To this end, we used a clustering quality metric from the data mining community. Third, we developed a general feature selection framework for systematically identifying resting-state functional connectivity (RSFC) elements that capture information to uniquely identify subjects. In sum, we evaluated six different approaches from this framework by quantifying both subject-specific fingerprinting accuracy and the decrease in accuracy with an increase in sample size to identify which approach improved quality metrics the most.
| S-EPMC8288098 | biostudies-literature
Project description:Background: Functional connectivity quantifies the statistical dependencies between the activity of brain regions, measured using neuroimaging data such as functional magnetic resonance imaging (fMRI) blood-oxygenation-level dependent time series. The network representation of functional connectivity, called a functional connectome (FC), has been shown to contain an individual fingerprint allowing participants identification across consecutive testing sessions. Recently, researchers have focused on the extraction of these fingerprints, with potential applications in personalized medicine. Materials and Methods: In this study, we show that a mathematical operation denominated degree-normalization can improve the extraction of FC fingerprints. Degree-normalization has the effect of reducing the excessive influence of strongly connected brain areas in the whole-brain network. We adopt the differential identifiability framework and apply it to both original and degree-normalized FCs of 409 individuals from the Human Connectome Project, in resting-state and 7 fMRI tasks. Results: Our results indicate that degree-normalization systematically improves three fingerprinting metrics, namely differential identifiability, identification rate, and matching rate. Moreover, the results related to the matching rate metric suggest that individual fingerprints are embedded in a low-dimensional space. Discussion: The results suggest that low-dimensional functional fingerprints lie in part in weakly connected subnetworks of the brain and that degree-normalization helps uncovering them. This work introduces a simple mathematical operation that could lead to significant improvements in future FC fingerprinting studies.
| S-EPMC8978572 | biostudies-literature
Project description:A better characterization of how an individual's brain is functionally organized will likely bring dramatic advances to many fields of study. Here we show a model-based approach toward characterizing resting state functional connectivity MRI (rs-fcMRI) that is capable of identifying a so-called "connectotype", or functional fingerprint in individual participants. The approach rests on a simple linear model that proposes the activity of a given brain region can be described by the weighted sum of its functional neighboring regions. The resulting coefficients correspond to a personalized model-based connectivity matrix that is capable of predicting the timeseries of each subject. Importantly, the model itself is subject specific and has the ability to predict an individual at a later date using a limited number of non-sequential frames. While we show that there is a significant amount of shared variance between models across subjects, the model's ability to discriminate an individual is driven by unique connections in higher order control regions in frontal and parietal cortices. Furthermore, we show that the connectotype is present in non-human primates as well, highlighting the translational potential of the approach.
| S-EPMC4227655 | biostudies-literature
Project description:BackgroundFunctional connectome fingerprinting can identify individuals based on their functional connectome. Previous studies relied mostly on short intervals between fMRI acquisitions.ObjectiveThis cohort study aimed to determine the stability of connectome-based identification and their underlying signatures in patients with multiple sclerosis and healthy individuals with long follow-up intervals.MethodsWe acquired resting-state fMRI in 70 patients with multiple sclerosis and 273 healthy individuals with long follow-up times (up to 4 and 9 years, respectively). Using functional connectome fingerprinting, we examined the stability of the connectome and additionally investigated which regions, connections and networks supported individual identification. Finally, we predicted cognitive and behavioural outcome based on functional connectivity.ResultsMultiple sclerosis patients showed connectome stability and identification accuracies similar to healthy individuals, with longer time delays between imaging sessions being associated with accuracies dropping from 89% to 76%. Lesion load, brain atrophy or cognitive impairment did not affect identification accuracies within the range of disease severity studied. Connections from the fronto-parietal and default mode network were consistently most distinctive, i.e., informative of identity. The functional connectivity also allowed the prediction of individual cognitive performances.ConclusionOur results demonstrate that discriminatory signatures in the functional connectome are stable over extended periods of time in multiple sclerosis, resulting in similar identification accuracies and distinctive long-lasting functional connectome fingerprinting signatures in patients and healthy individuals.
| S-EPMC10460476 | biostudies-literature
Project description:BackgroundAdvanced stages of different renal diseases feature glomerular sclerosis at a histological level which is observed by light microscopy on tissue samples obtained by performing a kidney biopsy. Computer-aided diagnosis (CAD) systems leverage the potential of artificial intelligence (AI) in healthcare to support physicians in the diagnostic process.MethodsWe propose a novel CAD system that processes histological images and discriminates between sclerotic and non-sclerotic glomeruli. To this goal, we designed, tested, and compared two artificial neural network (ANN) classifiers. The former implements a shallow ANN classifying hand-crafted features extracted from Regions of Interest (ROIs) by means of image-processing procedures. The latter, instead, employs the IBM Watson Visual Recognition System, which uses a deep artificial neural network making decisions taking the images as input, without the need to design any procedure for describing images with features. The input dataset consisted of 428 sclerotic glomeruli and 2344 non-sclerotic glomeruli derived from images of kidney biopsies scanned by the Aperio ScanScope System.ResultsBoth AI approaches allowed to very accurately distinguish (mean MCC 0.95 and mean Accuracy 0.99) between sclerotic and non-sclerotic glomeruli. Although the systems may seem interchangeable, the approach based on feature extraction and classification would allow clinicians to gain information on the most discriminating features. In fact, further procedures could explain the classifier's decision by analysing which subset of features impacted the most on the final decision.ConclusionsWe developed a customizable support system that can facilitate the work of renal pathologists both in clinical and research settings.
| S-EPMC8765108 | biostudies-literature
Project description:The specificity and sensitivity of resting state functional MRI (rs-fMRI) measurements depend on preprocessing choices, such as the parcellation scheme used to define regions of interest (ROIs). In this study, we critically evaluate the effect of brain parcellations on machine learning models applied to rs-fMRI data. Our experiments reveal an intriguing trend: On average, models with stochastic parcellations consistently perform as well as models with widely used atlases at the same spatial scale. We thus propose an ensemble learning strategy to combine the predictions from models trained on connectivity data extracted using different (e.g., stochastic) parcellations. We further present an implementation of our ensemble learning strategy with a novel 3D Convolutional Neural Network (CNN) approach. The proposed CNN approach takes advantage of the full-resolution 3D spatial structure of rs-fMRI data and fits non-linear predictive models. Our ensemble CNN framework overcomes the limitations of traditional machine learning models for connectomes that often rely on region-based summary statistics and/or linear models. We showcase our approach on a classification (autism patients versus healthy controls) and a regression problem (prediction of subject's age), and report promising results.
| S-EPMC6777738 | biostudies-literature
Project description:BackgroundShort mindfulness-based interventions have gained traction in research due to their positive impact on well-being, cognition, and clinical symptoms across various settings. However, these short-term trainings are viewed as preliminary steps within a more extensive transformative path, presumably leading to long-lasting trait changes. Despite this, little is still known about the brain correlates of these meditation traits.MethodsTo address this gap, we investigated the neural correlates of meditation expertise in long-term Buddhist practitioners, comparing the large-scale brain functional connectivity of 28 expert meditators with 47 matched novices. Our hypothesis posited that meditation expertise would be associated with specific and enduring patterns of functional connectivity present during both meditative (open monitoring/open presence and loving-kindness and compassion meditations) and nonmeditative resting states, as measured by connectivity gradients.ResultsApplying a support vector classifier to states not included in training, we successfully decoded expertise as a trait, demonstrating its non-state-dependent nature. The signature of expertise was further characterized by an increased integration of large-scale brain networks, including the dorsal and ventral attention, limbic, frontoparietal, and somatomotor networks. The latter correlated with a higher ability to create psychological distance from thoughts and emotions.ConclusionsSuch heightened integration of bodily maps with affective and attentional networks in meditation experts could point toward a signature of the embodied cognition cultivated in these contemplative practices.
| S-EPMC11414651 | biostudies-literature