Article
Neurosciences
Lennart Frahm, Theodore D. Satterthwaite, Peter T. Fox, Robert Langner, Simon B. Eickhoff
Summary: This study evaluated the performance of the Activation Likelihood Estimation (ALE) meta-analysis algorithm in structural neuroimaging data. The results showed that the sensitivity and specificity of the ALE algorithm in structural data were similar to those observed in functional data. To prevent significant clusters from being driven by single experiments, it is recommended to include at least 23 experiments in a VBM ALE dataset.
Article
Computer Science, Artificial Intelligence
Chunzhi Gu, Xuequan Lu, Chao Zhang
Summary: This paper proposes a probability-based approach for color transfer and treats it as a parameter estimation problem. By utilizing Gaussian Mixture Model and Expectation-Maximization algorithm for optimization, multiple high-quality color transfer results can be generated.
PATTERN RECOGNITION
(2022)
Article
Neurosciences
Qiu-Yu Lu, Jonathan M. Towne, Matthew Lock, Chao Jiang, Zhi-Xiang Cheng, Mohamad Habes, Xi-Nian Zuo, Yu-Feng Zang
Summary: Characterizing the involvement of specific brain regions in cognitive function has always been a challenge in cognitive neuroscience. Functional magnetic resonance imaging (fMRI) techniques have provided solutions for mapping functional neural networks. However, the complex nature of structure-function correspondence makes it difficult to fully capture higher-order cognitive functions in task design. Other research practices, such as brain-behavior association or between-group comparisons, are widely used to explore cognitive correlations with specific brain regions. However, the interpretation of results from specific brain regions in relation to their underlying cognitive functions has been too general in publications.
Review
Computer Science, Artificial Intelligence
Jinhe Dong, Jun Shi, Yue Gao, Shihui Ying
Summary: This paper proposes a robust noise model that incorporates a mixture of Gaussian noise modeling strategy into a baseline classification model. The number of mixture components is automatically selected using the penalized likelihood method. The proposed model defines hyperparameters from the error representation and achieves the best performance compared to conventional classification methods.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Geochemistry & Geophysics
Weiqiang Zhu, Ian W. McBrearty, S. Mostafa Mousavi, William L. Ellsworth, Gregory C. Beroza
Summary: Earthquake phase association algorithms play a crucial role in earthquake monitoring and research, but can be challenging for densely clustered earthquake sequences. This study presents a novel method, Gaussian Mixture Model Association (GaMMA), which effectively associates seismic phases and provides accurate estimates of earthquake location and magnitude.
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
(2022)
Article
Psychology, Mathematical
Jordi Manuello, Donato Liloia, Annachiara Crocetta, Franco Cauda, Tommaso Costa
Summary: Coordinate-based meta-analysis (CBMA) is a powerful technique in human brain imaging research, and the Coordinate-Based Meta-Analyses Toolbox (CBMAT) provides a user-friendly and automated MATLAB (R) functions for data preparation and post hoc analyses. This paper describes the code and provides an example of using CBMAT on a dataset. CBMAT can significantly improve the way data are handled in CBMAs. The code can be downloaded online.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Computer Science, Artificial Intelligence
Younghwan Jeon, Ganguk Hwang
Summary: This paper addresses the data association problem and proposes a Bayesian approach based on a mixture of Gaussian Processes (GPs) to adapt to changing observations. Experimental results and theoretical analysis demonstrate the effectiveness of the proposed method.
PATTERN RECOGNITION
(2022)
Article
Chemistry, Multidisciplinary
Alistair Forbes
Summary: This paper presents a method for evaluating the uncertainties associated with Gaussian-associated features using the GUM methodology. By calculating sensitivity matrices and estimating variance matrices associated with feature parameters, the uncertainties of fitted parameters can be estimated prior to measurement strategies. The calculations involved are direct and do not require optimization or Monte Carlo sampling, and can be implemented in spreadsheet software.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Artificial Intelligence
Tomoharu Iwata
Summary: Appropriate representations are critical for better clustering performance. Existing neural network-based clustering methods do not directly train neural networks to improve clustering performance. We propose a method that meta-learns clustering knowledge from labeled data and applies it to cluster unseen unlabeled data. The method trains neural networks to obtain representations that improve clustering performance through variational Bayesian (VB) inference and infinite Gaussian mixture models.
Article
Engineering, Civil
Xiaoxu Chen, Chengyuan Zhang, Zhanhong Cheng, Yuang Hou, Lijun Sun
Summary: Car-following models are essential for microscopic traffic simulation. This study presents a data-driven model based on a Bayesian Gaussian mixture model for probabilistic forecasting of human car-following behaviors. The model captures the temporal dynamics of human car-following behaviors and provides accurate predictions with quantified uncertainty. The results suggest that this model is promising for modeling and forecasting car-following behaviors.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Tao Li, Jinwen Ma
Summary: This paper introduces a powerful model called the mixture of Gaussian processes (MGP). Conventional MGPs cannot effectively handle the case where the input variable lies on a general manifold or a graph. Based on the attention mechanism, the paper proposes two novel mixture models of Gaussian processes that overcome the limitations of conventional MGPs. Experimental results demonstrate the effectiveness of these methods.
PATTERN RECOGNITION LETTERS
(2022)
Article
Geochemistry & Geophysics
Jiahui Qu, Qian Du, Yunsong Li, Long Tian, Haoming Xia
Summary: This article proposes a novel Gaussian mixture model-based anomaly detection method for hyperspectral images, with main contributions being a new extraction approach for anomaly pixels and a weighting approach for fusing the results.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Engineering, Multidisciplinary
Wei Huang, Hongpo Fu, Yu Li, Weiguo Zhang
Summary: This paper investigates the state estimation problems of nonlinear systems with outlier-corrupted measurements. A new Gaussian-multivariate Laplacian mixture (GMLM) distribution is proposed to model the non-Gaussian noises caused by randomly occurring measurement outliers, and its distribution characteristics are analyzed. The measurement likelihood probability density function (PDF) is formulated as the GMLM distribution and further expressed as a hierarchical Gaussian expression. A robust cubature Kalman filter, GMLMRCKF, is derived using the variational Bayesian (VB) method. Simulation test and real data evaluation show that the proposed GMLMRCKF has better estimation accuracy and consistency in the case of non-stationary heavy-tailed noises than existing robust filters, with comparable performance to the standard CKF in the absence of outliers and better robust performance in the presence of unknown outliers.
Review
Anatomy & Morphology
Ole J. J. Boeken, Edna C. C. Cieslik, Robert Langner, Sebastian Markett
Summary: The human thalamus plays a crucial role in relaying sensory signals, sensorimotor functions and various cognitive functions. However, its full functional repertoire and internal anatomical structure are still not fully understood. In this study, we developed a novel systems-level decoding approach that combines traditional neuroinformatics methods to investigate the functional profile of the thalamus. By decoding the functional repertoire of thalamic subregions in conjunction with functionally connected cortical regions, we identified key structure-function relationships and discovered associations with language, memory, and locomotion.
BRAIN STRUCTURE & FUNCTION
(2023)
Article
Cell Biology
LiQin Sheng, PanWen Zhao, HaiRong Ma, Joaquim Radua, ZhongQuan Yi, YuanYuan Shi, JianGuo Zhong, ZhenYu Dai, PingLei Pan
Summary: A comprehensive meta-analysis of numerous studies on PD patients found a lack of consistent CTh alterations, indicating CTh is not a reliable neuroimaging marker for PD. This inconsistency may be attributed to heterogeneous clinical populations, variations in imaging methods, and underpowered sample sizes. These results emphasize the importance of controlling potential confounding factors to produce robust and reproducible CTh results in PD.
Article
Psychology, Clinical
Debo Dong, Dezhong Yao, Yulin Wang, Seok-Jun Hong, Sarah Genon, Fei Xin, Kyesam Jung, Hui He, Xuebin Chang, Mingjun Duan, Boris C. Bernhardt, Daniel S. Margulies, Jorge Sepulcre, Simon B. Eickhoff, Cheng Luo
Summary: This study investigated the pathological interaction of sensory and cognitive function in schizophrenia and its relationship to system-level imbalance. The results revealed a compression of the cortical hierarchy organization, leading to a diminished separation between sensory and cognitive systems. Furthermore, the analysis showed reduced connectivity within unimodal regions and increased connectivity between unimodal regions and other areas. These findings suggest that disruptions in the somatosensory-motor system and inefficient integration of sensory information contribute to high-level cognitive deficits in schizophrenia.
PSYCHOLOGICAL MEDICINE
(2023)
Article
Anatomy & Morphology
Shanshan Zhen, Zachary A. Yaple, Simon B. Eickhoff, Rongjun Yu
Summary: Individuals not only take actions to gain immediate rewards, but also explore to gather information for future decisions. Research shows that exploration is associated with brain activity related to risk, cognitive control, and motor processing, highlighting the importance of the prefrontal-insular-motor cortical network in decision-making processes.
BRAIN STRUCTURE & FUNCTION
(2022)
Article
Psychology, Experimental
Lya K. Paas Oliveros, Aleks Pieczykolan, Rachel N. Plaeschke, Simon B. Eickhoff, Robert Langner
Summary: Difficulties in performing two tasks at once increase with age. Conflicting response codes can lead to interference, and this interference is more pronounced in older adults, suggesting deficits in multiple-action control.
PSYCHOLOGICAL RESEARCH-PSYCHOLOGISCHE FORSCHUNG
(2023)
Article
Clinical Neurology
Lukas Hensel, Fabian Lange, Caroline Tscherpel, Shivakumar Viswanathan, Jana Freytag, Lukas J. Volz, Simon B. Eickhoff, Gereon R. Fink, Christian Grefkes
Summary: This study assessed the contributions of the ipsilesional and contralesional anterior intraparietal cortex (aIPS) for hand motor function in stroke patients and found increased resting-state connectivity in patients with good motor outcome. Interhemispheric connectivity was also found to be correlated with better motor performance.
Article
Neurosciences
Tong He, Lijun An, Pansheng Chen, Jianzhong Chen, Jiashi Feng, Danilo Bzdok, Avram J. Holmes, Simon B. Eickhoff, B. T. Thomas Yeo
Summary: This paper presents a simple yet powerful approach to translate predictive models from large-scale datasets to small-scale studies, improving the prediction capability. The results demonstrate that meta-matching can greatly enhance predictions of new phenotypes in small independent datasets in various scenarios.
NATURE NEUROSCIENCE
(2022)
Review
Anesthesiology
Alina T. Henn, Bart Larsen, Lennart Frahm, Anna Xu, Azeez Adebimpe, J. Cobb Scott, Sophia Linguiti, Vaishnavi Sharma, Allan Basbaum, Gregory Corder, Robert H. Dworkin, Robert R. Edwards, Clifford J. Woolf, Ute Habel, Simon B. Eickhoff, Claudia R. Eickhoff, Lisa Wagels, Theodore D. Satterthwaite
Summary: Neuroimaging is a powerful tool for studying the relationship between chronic pain and brain structure. A meta-analysis of structural magnetic imaging studies found subtle but widespread alterations in brain structure associated with chronic pain. These alterations primarily occurred in brain regions involved in pain processing.
Review
Neurosciences
Ji Chen, Kaustubh R. Patil, B. T. Thomas Yeo, Simon B. Eickhoff
Summary: Much attention is being paid to developing diagnostic classifiers for mental disorders. In addition, machine learning is highlighted as a potential tool for gaining biological insights into the psychopathology and nosology of mental disorders. Brain imaging data, obtained noninvasively from large cohorts, has been used in studies to reveal intermediate phenotypes and refine the taxonomy of mental illness. Machine learning models' accuracy can identify pathophysiology-related features, addressing the dimensional and overlapping symptomatology of psychiatric illness. A multiview perspective combining molecular and system-level data and efforts toward data-driven definition of subtypes or disease entities through unsupervised and semisupervised approaches have also been emphasized.
BIOLOGICAL PSYCHIATRY
(2023)
Article
Neurosciences
Mengmeng Wang, Shunmin Zhang, Tao Suo, Tianxin Mao, Fenghua Wang, Yao Deng, Simon Eickhoff, Yu Pan, Caihong Jiang, Hengyi Rao
Summary: The Balloon Analog Risk Task (BART) has been widely used to assess risk-taking behavior and brain function. This study used activation likelihood estimation (ALE) meta-analysis and functional connectivity (FC) analysis to synthesize brain networks involved in risk-taking during the BART and compared differences between adults and adolescents. The results showed that reward, salience, and executive control networks play important roles in risk-taking during the BART, and adolescents exhibit greater activation compared to adults.
HUMAN BRAIN MAPPING
(2022)
Article
Neurosciences
Marisa K. Heckner, Edna C. Cieslik, Kaustubh R. Patil, Martin Gell, Simon B. Eickhoff, Felix Hoffstaedter, Robert Langner
Summary: Healthy aging is associated with changes in executive functioning (EF) and resting-state functional connectivity (RSFC) within brain networks. However, it is unclear how RSFC in EF-associated networks predicts individual EF performance. This study found low prediction accuracies and a lack of specificity regarding neurobiological networks for predicting EF abilities, suggesting the need for future research with different task states, brain modalities, larger samples, and more comprehensive measures.
Article
Neurosciences
Andy Wai Kan Yeung, Michaela Robertson, Angela Uecker, Peter T. Fox, Simon B. Eickhoff
Summary: The literature of neuroimaging meta-analysis, particularly the activation likelihood estimation (ALE) approach, has been thriving for over a decade. A meta-evaluation of these meta-analyses was performed to evaluate their design and reporting standards. The study found that the use of cluster-level family-wise error (FWE) correction method has become dominant, and there has been slight improvement in reporting on data redundancy elimination and providing input data. However, there is still room for improvement in terms of data and code availability statements and data submission to BrainMap.
HUMAN BRAIN MAPPING
(2023)
Article
Neurosciences
Di Wang, Nicolas Honnorat, Peter T. Fox, Kerstin Ritter, Simon B. Eickhoff, Sudha Seshadri, Mohamad Habes
Summary: We compared three heatmap methods derived from deep neural networks and SVM activation patterns to analyze structural MRI scans of subjects with Alzheimer's disease. Our results showed that all three heatmap methods were able to capture brain regions more accurately than SVM activation patterns, and the Integrated Gradients method had the best overlap with the independent meta-analysis.
Review
Behavioral Sciences
Arianna Sala, Aldana Lizarraga, Silvia Paola Caminiti, Vince D. Calhoun, Simon B. Eickhoff, Christian Habeck, Sharna D. Jamadar, Daniela Perani, Joana B. Pereira, Mattia Veronese, Igor Yakushev
Summary: Brain connectomics has become a major concept in neuroscience, and molecular imaging provides unique information that is inaccessible to MRI-based and electrophysiological techniques. Therefore, we encourage an integrative approach to better understand the brain connectome by combining MRI, electrophysiological techniques, and molecular imaging.
TRENDS IN COGNITIVE SCIENCES
(2023)
Article
Medicine, General & Internal
Bastian Cheng, Ji Chen, Alina Koenigsberg, Carola Mayer, Leander Rimmele, Kaustubh R. Patil, Christian Gerloff, Gotz Thomalla, Simon B. Eickhoff
Summary: This study used advanced machine learning techniques to analyze the dimensional structure of NIHSS and identified a five-dimensional representation, including left motor deficits, right motor deficits, dysarthria and facial palsy, language, and deficits in spatial attention and gaze. The study also validated the neurobiological basis of these dimensions through neuroanatomical and functional analysis, providing a valuable anatomical map for individualized stroke treatment and rehabilitation.
Article
Clinical Neurology
Jan Kasper, Simon B. Eickhoff, Svenja Caspers, Jessica Peter, Imis Dogan, Robert Christian Wolf, Kathrin Reetz, Juergen Dukart, Michael Orth
Summary: Kasper et al. found that in Huntington's disease, the functional integrity of the dopamine receptor-rich caudate nucleus plays a crucial role in maintaining network function. Loss of caudate functional integrity leads to motor signs independent of atrophy. This finding may have implications for other neurodegenerative diseases.
Article
Neurosciences
Shammi More, Georgios Antonopoulos, Felix Hoffstaedter, Julian Caspers, Simon B. Eickhoff, Kaustubh R. Patil
Summary: The difference between predicted age based on brain scans and chronological age can be used as a proxy for atypical aging. Different data representations and machine learning algorithms have different effects on performance criteria such as accuracy, generalizability, reliability, and consistency. The choice of feature representation and machine learning algorithm both affect performance, and further evaluation and improvements are needed for real-world application.
