Neuropsychiatric disease classification using functional connectomics - results of the connectomics in neuroimaging transfer learning challenge

Schirmer, Markus D.; Venkataraman, Archana; Rekik, Islem; Kim, Minjeong; Mostofsky, Stewart H.; Nebel, Mary Beth; Rosch, Keri S.; Seymour, Karen E.; Crocetti, Deana; Irzan, Hassna; Hütel, Michael; Ourselin, Sébastien; Marlow, Neil; Melbourne, Andrew; Levchenko, Egor; Zhou, Shuo; Kunda, Mwiza; Lu, Haiping; Dvornek, Nicha C.; Zhuang, Jian; Pinto, Gideon; Samal, Sandip; Zhang, Jennings; Bernal-Rusiel, Jorge L.; Pienaar, Rudolph; Chung, Ai Wern

doi:10.1016/j.media.2021.101972

Cited by 23 publications

(15 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Out of the three parcellation schemes, the AAL atlas provides slightly better classification accuracy, which illustrates that the parcellation scheme can have an effect on the manner in which the network features are associated with the clinical outcome. We note that the classification accuracy reported in our analysis are comparable, and often higher compared to the results reported for the same dataset in [26]. In addition, the classification performance using dynamic connectivity based on sliding windows appear less accurate compared to the static connectivity results, that is consistent with our findings from the HCP study (not reported due to space constraints).…”

Section: Analysis Results For Cni Datasupporting

confidence: 88%

“…In our analysis we used 120 ADHD children after excluding those with the ASD diagnosis, and matched this cohort with another 120 NC subjects. The data was downloaded from the Github repositories that are provided in [26]. The acquisition protocol for the rs-fMRI data used a single shot, partially parallel gradient-recalled EPI sequence with TR/TE 2500/30ms, flip angle 70 degrees, and voxel resolution 3.05 × 3.15 × 3mm 3 on a Philips 3T Achieva scanner.…”

Section: Methodsmentioning

confidence: 99%

“…We evaluate the performance of different approaches using out-of-sample classification accuracy of the test samples, via several metrics popularly used in literature [26]. In particular, we report the mis-classification rate defined as the ratio of incorrect classification labels over the total number of instances evaluated, the F 1 score that is computed as the harmonic mean between recall (sensitivity) and precision, i.e.…”

Section: B Comparison Methods and Metricsmentioning

confidence: 99%

See 2 more Smart Citations

Flexible Bayesian Support Vector Machines for Brain Network-based Classification

Jin¹,

Kundu²

2022

Preprint

View full text Add to dashboard Cite

Objective: Brain networks have gained increasing recognition as potential biomarkers in mental health studies, but there are limited approaches that can leverage complex brain networks for accurate classification. Our goal is to develop a novel Bayesian Support Vector Machine (SVM) approach that incorporates highdimensional networks as covariates and is able to overcome limitations of existing penalized methods. Methods: We develop a novel Dirichlet process mixture of double exponential priors on the coefficients in the Bayesian SVM model that is able to perform feature selection and uncertainty quantification, by pooling information across edges to determine differential sparsity levels in an unsupervised manner. We develop different versions of the model that incorporates static and dynamic connectivity features, as well as an integrative analysis that jointly includes features from multiple scanning sessions. We perform classification of intelligence levels using resting state fMRI data from the Human Connectome Project (HCP), and a second Attention Deficiency Hyperactivity Disorder (ADHD) classification task. Results: Our results clearly reveal the considerable greater classification accuracy under the proposed approach over state-of-the-art methods. The multi-session analysis results in the highest classification accuracy in the HCP data analysis. Conclusion: We provide concrete evidence that the novel Bayesian SVMs provides an unsupervised and automated approach for network-based classification, that results in considerable improvements over penalized methods and parametric Bayesian approaches. Significance: Our work is one of the first to conclusively demonstrate the advantages of a Bayesian SVM in networkbased classification of mental health outcomes, and the importance of multi-session network analysis.

show abstract

Section: Analysis Results For Cni Datasupporting

confidence: 88%

Section: Methodsmentioning

confidence: 99%

Section: B Comparison Methods and Metricsmentioning

confidence: 99%

See 1 more Smart Citation

Flexible Bayesian Support Vector Machines for Brain Network-based Classification

Jin¹,

Kundu²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…This study used open‐source 240 resting‐state fMRI (rsfMRI) time series data from the Connectomics in NeuroImaging Transfer Learning Challenge (CNI‐TLC) [ 54 ] for connectome analysis and diagnosis of ADHD. First, each rsfMRI time series was averaged according to the AAL [ 55 ] standard parcellation atlas.…”

Section: Methodsmentioning

confidence: 99%

Graph Analysis with Multifunctional Self‐Rectifying Memristive Crossbar Array

et al. 2023

View full text Add to dashboard Cite

Many big data have interconnected and dynamic graph structures growing over time. Analyzing these graphical data requires the hidden relationship between the nodes in the graphs to be identified, which has conventionally been achieved by finding the effective similarity. However, graphs are generally non‐Euclidean, which does not allow finding it. In this study, the non‐Euclidean graphs are mapped to a specific crossbar array (CBA) composed of self‐rectifying memristors and metal cells at the diagonal positions. The sneak current, an intrinsic physical property in the CBA, allows for the identification of the similarity function. The sneak‐current‐based similarity function indicates the distance between the nodes, which can be used to predict the probability that unconnected nodes will be connected in the future, connectivity between communities, and neural connections in a brain. When all bit lines of the CBA are connected to the ground, the sneak current is suppressed, and the CBA can be used to search for adjacent nodes. This work demonstrates the physical calculation methods applied to various graphical problems using the CBA composed of the self‐rectifying memristor based on the HfO2 switching layer. Moreover, such applications suffer less from the memristors’ inherent issues related to their stochastic nature.

show abstract

“…Dynamic learning can help AI models to be continuously augmented based on expert’s feedback. Transfer learning is where a model trained with another dataset is fine-tuned over the target dataset [ 41 , 42 ]. Transfer learning is an effective approach when sufficient data is not available to train the model from scratch.…”

Section: Machine Learningmentioning

confidence: 99%

Radiomics, machine learning, and artificial intelligence—what the neuroradiologist needs to know

Wagner

Namdar²,

Biswas

et al. 2021

Neuroradiology

View full text Add to dashboard Cite

Purpose Artificial intelligence (AI) is playing an ever-increasing role in Neuroradiology. Methods When designing AI-based research in neuroradiology and appreciating the literature, it is important to understand the fundamental principles of AI. Training, validation, and test datasets must be defined and set apart as priorities. External validation and testing datasets are preferable, when feasible. The specific type of learning process (supervised vs. unsupervised) and the machine learning model also require definition. Deep learning (DL) is an AI-based approach that is modelled on the structure of neurons of the brain; convolutional neural networks (CNN) are a commonly used example in neuroradiology. Results Radiomics is a frequently used approach in which a multitude of imaging features are extracted from a region of interest and subsequently reduced and selected to convey diagnostic or prognostic information. Deep radiomics uses CNNs to directly extract features and obviate the need for predefined features. Conclusion Common limitations and pitfalls in AI-based research in neuroradiology are limited sample sizes (“small-n-large-p problem”), selection bias, as well as overfitting and underfitting.

show abstract

Neuropsychiatric disease classification using functional connectomics - results of the connectomics in neuroimaging transfer learning challenge

Abstract: This is a repository copy of Neuropsychiatric disease classification using functional connectomics -results of the connectomics in neuroImaging transfer learning challenge.

Cited by 23 publications

References 69 publications

Flexible Bayesian Support Vector Machines for Brain Network-based Classification

Flexible Bayesian Support Vector Machines for Brain Network-based Classification

Graph Analysis with Multifunctional Self‐Rectifying Memristive Crossbar Array

Radiomics, machine learning, and artificial intelligence—what the neuroradiologist needs to know

Contact Info

Product

Resources

About