Review
Chemistry, Analytical
L. Ortiz-Herrero, M. I. Maguregui, L. Bartolome
Summary: This review emphasizes the powerful applicability of multivariate regression methods in forensic dating, providing insights into the benefits and drawbacks of using these methods and proposing suggestions on how to make optimal use of them.
TRAC-TRENDS IN ANALYTICAL CHEMISTRY
(2021)
Article
Geosciences, Multidisciplinary
Xi Chen, Wei Chen
Summary: This study proposed landslide susceptibility evaluation models based on bivariate statistical correlation analysis and optimization of different kernel functions in Zichang City, China. By analyzing spatial patterns, three different models were compared to establish landslide susceptibility maps for the region, providing valuable information for natural hazards prevention efforts in the future.
Article
Neurosciences
Ruben Sanchez-Romero, Michael W. Cole
Summary: Cognition and behavior are influenced by interactions within brain networks, emphasizing the importance of causal interactions in studying brain function. Traditional bivariate methods for functional connectivity analysis lack consideration of confounders, leading to false positives. A new combined FC method (CombinedFC) was proposed to incorporate both simple bivariate and partial correlation measures, providing more valid causal inferences and improving upon existing methods.
JOURNAL OF COGNITIVE NEUROSCIENCE
(2021)
Review
Neurosciences
Laurentius Huber (Renzo), Emily S. Finn, Yuhui Chai, Rainer Goebel, Ruediger Stirnberg, Tony Stoecker, Sean Marrett, Kamil Uludag, Seong-Gi Kim, SoHyun Han, Peter A. Bandettini, Benedikt A. Poser
Summary: Recent advances in fMRI technology have enabled researchers to study information processing in the cortical layers more effectively, particularly in terms of connectivity. However, layer-fMRI still faces challenges that require more flexible and precise methods to address. This article describes newly developed acquisition methodologies that can provide more comprehensive data for investigating brain network connections.
PROGRESS IN NEUROBIOLOGY
(2021)
Review
Ecology
Abelardo Montesinos-Lopez, Osval Antonio Montesinos-Lopez, Jose Cricelio Montesinos-Lopez, Carlos Alberto Flores-Cortes, Roberto de la Rosa, Jose Crossa
Summary: This paper provides a guide on implementing Bayesian generalized kernel regression methods for genomic prediction in R, showcasing the construction process of seven kernel methods and providing specific examples. The strengths, limitations, and contributions of kernel methods are discussed.
Article
Biochemical Research Methods
Bingxing An, Mang Liang, Tianpeng Chang, Xinghai Duan, Lili Du, Lingyang Xu, Lupei Zhang, Xue Gao, Junya Li, Huijiang Gao
Summary: The study introduced a novel cosine kernel-based KRR model, KCRR, for genomic prediction (GP) in breeding programs. KCRR showed stable performance across multiple species, suggesting its potential for diverse genetic architectures. Additionally, a modified genomic similarity matrix called Cosine similarity matrix (CS matrix) was defined, which significantly improved computing efficiency without compromising prediction accuracy when compared to traditional methods like GBLUP. This research presents a promising strategy for future genomic prediction.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Neurosciences
Xiaoyu Wang, Katharina Zwosta, Uta Wolfensteller, Hannes Ruge
Summary: This study utilized advanced regression and predictive modeling techniques to demonstrate that learning-related functional connectivity alterations across the whole brain can predict individual habit strength. The functional connectivity changes involving the sensorimotor network and the cingulo-opercular network play a pivotal role in predicting individual habit strength.
HUMAN BRAIN MAPPING
(2023)
Article
Environmental Sciences
Mir Jafar Sadegh Safari, Shervin Rahimzadeh Arashloo, Babak Vaheddoost
Summary: Multiple kernel fusion (MKF) combines multiple sources of information to improve performance. This study applies MKF in hydrological modeling to simulate lake water depth and demonstrates improved predictive performance compared to other models.
ENVIRONMENTAL RESEARCH
(2023)
Article
Computer Science, Interdisciplinary Applications
Pau Batlle, Matthieu Darcy, Bamdad Hosseini, Houman Owhadi
Summary: We present a general kernel-based framework for learning operators between Banach spaces. Our approach is competitive in terms of cost-accuracy trade-off and matches or beats the performance of popular neural net methods on most benchmarks. The framework inherits advantages from kernel methods, such as simplicity, interpretability, convergence guarantees, a priori error estimates, and Bayesian uncertainty quantification.
JOURNAL OF COMPUTATIONAL PHYSICS
(2024)
Article
Computer Science, Artificial Intelligence
Juan Pablo Karmy, Julio Lopez, Sebastian Maldonado
Summary: This paper presents a novel method that extends kernel-based support vector regression to hierarchical time series analysis. The method constructs multiple predictors in a single optimization problem, pooling information across different levels. In addition to the traditional objectives of model fit and regularization, a third objective for data pooling is included. Experimental results demonstrate the advantages of this method over other strategies for handling hierarchical time series forecasting.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Interdisciplinary Applications
J-L Akian, L. Bonnet, H. Owhadi, E. Savin
Summary: This paper introduces algorithms for selecting/designing kernels in Gaussian process regression/kriging surrogate modeling techniques. It presents two classes of algorithms: kernel flow, which selects the best kernel by minimizing the loss of accuracy caused by removing a portion of the dataset, and spectral kernel ridge regression, which selects the best kernel by minimizing the norm of the function to be approximated in the associated RKHS. The effectiveness of both approaches is demonstrated through numerical examples.
JOURNAL OF COMPUTATIONAL PHYSICS
(2022)
Article
Computer Science, Artificial Intelligence
Ozan Ozyegen, Igor Ilic, Mucahit Cevik
Summary: This study introduces two novel evaluation metrics for time series forecasting, aiming to measure the local fidelity of local explanation methods. These metrics allow for a comprehensive comparison of various local explanation methods and provide an intuitive interpretation of model predictions. The research expands the theoretical foundation through experimental results and offers heuristic reasoning through extensive numerical studies.
APPLIED INTELLIGENCE
(2022)
Article
Chemistry, Multidisciplinary
Mingxin Tang, Libo Huang, Wei Chen
Summary: This study proposes a rapid and accurate PPA prediction method for template-based chip design, utilizing multivariate linear regression and Amdahl's law to fit and improve the accuracy of PPA estimation.
APPLIED SCIENCES-BASEL
(2022)
Article
Environmental Sciences
Xiaodong Zang, Wengang Qin, Yingying Xiong, Anlan Xu, Hesuyuan Huang, Tao Fang, Xiaowei Zang, Mingwu Chen
Summary: This study aimed to assess the association between aldehyde exposure and markers of inflammation and oxidative stress. The results showed significant associations between aldehyde compounds and markers such as serum iron levels and lymphocyte count. The weighted quantile sum and Bayesian kernel machine regression analyses confirmed the overall impact of aldehyde compounds on oxidative stress. This study has important guiding value for understanding the impact of environmental pollutants on population health.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Green & Sustainable Science & Technology
Qing Ye, Yi Li, Wenzhe Shen, Zhaoze Xuan
Summary: This study utilizes a combined method of kernel density estimation, spatial autocorrelation analysis, and multivariate logistic regression analysis to identify accident-prone areas and analyze the factors affecting the distribution of traffic accidents near highway ramps. Through data collection and analysis, the clustering characteristics of traffic accidents in the diversion and merging areas are identified, and four levels of accident-prone areas are classified based on accident rates. The spatial distribution of accidents is influenced by factors such as temperature, accident lane, weather conditions, and time of day, providing valuable insights for highway accident prevention management.
Article
Psychology
Shruti Ullas, Elia Formisano, Frank Eisner, Anne Cutler
ATTENTION PERCEPTION & PSYCHOPHYSICS
(2020)
Article
Mathematical & Computational Biology
Isma Zulfiqar, Michelle Moerel, Elia Formisano
FRONTIERS IN COMPUTATIONAL NEUROSCIENCE
(2020)
Article
Multidisciplinary Sciences
Miriam Heynckes, Peter De Weerd, Giancarlo Valente, Elia Formisano, Federico De Martino
Article
Neurosciences
Lars Hausfeld, Martha Shiell, Elia Formisano, Lars Riecke
Summary: Selective attention is crucial for processing auditory scenes with multiple speakers, as it involves separating relevant speech from irrelevant speech. This study found that increasing perceptual demand may reduce cortical processing of distractor speech and decrease their perceptual segregation.
Article
Neurosciences
Giancarlo Valente, Agustin Lage Castellanos, Lars Hausfeld, Federico De Martino, Elia Formisano
Summary: The study focuses on the false positive control of different permutation strategies and the statistical power of different cross-validation schemes. Simulations demonstrate that estimating the entire cross-validation error on each permuted dataset is the only statistically valid permutation strategy. Among different cross-validation schemes, a repeated split-half cross-validation is the most powerful.
Article
Neurosciences
Isma Zulfiqar, Michelle Moerel, Agustin Lage-Castellanos, Elia Formisano, Peter De Weerd
Summary: Recent studies have highlighted the possible contributions of direct connectivity between early sensory cortices to audiovisual integration. Anatomical connections between the early auditory and visual cortices are concentrated in visual sites representing the peripheral field of view. The study found evidence for an influence of barely-detectable visual stimuli on the response times for auditory stimuli, but not for the reverse effect, suggesting an asymmetry between the auditory influence on visual processing and the visual influence on auditory processing.
FRONTIERS IN HUMAN NEUROSCIENCE
(2021)
Article
Neurosciences
Isma Zulfiqar, Martin Havlicek, Michelle Moerel, Elia Formisano
Summary: Recent fMRI studies have shown differences in responses to natural sounds along the rostral-caudal axis of the human superior temporal gyrus. A forward model combining neuronal and hemodynamic BOLD response modeling was used to link neuronal response properties with fMRI data, revealing complementary neural information processing along this axis.
Article
Neurosciences
Lars Hausfeld, Niels R. Disbergen, Giancarlo Valente, Robert J. Zatorre, Elia Formisano
Summary: Neuroimaging studies show that the auditory cortex tracks ongoing speech and enhances tracking of the attended speaker in multi-speaker environments. While multi-instrument music can be appreciated by focusing on individual instruments (segregation) and multiple instruments simultaneously (integration), attentional modulation is observed during segregation tasks but not during integration tasks.
FRONTIERS IN NEUROSCIENCE
(2021)
Article
Psychology, Multidisciplinary
Bruno L. Giordano, Ricardo de Miranda Azevedo, Yenisel Plasencia-Calana, Elia Formisano, Michel Dumontier
Summary: This study surveyed the development of taxonomies and ontologies for everyday sounds and found a lack of comprehensive ontology that covers all semantic aspects of sound relations.
FRONTIERS IN PSYCHOLOGY
(2022)
Article
Neurosciences
Bruno L. Giordano, Michele Esposito, Giancarlo Valente, Elia Formisano
Summary: The authors compare three classes of models (acoustic, semantic, and sound-to-event deep neural network) to determine which can best link specific features of auditory stimuli to predicted functional magnetic resonance imaging responses in auditory cortical regions. Sound recognition involves the transformation of input waveforms into semantic representations in the brain. They found that sound-to-event deep neural networks outperform acoustic and semantic models in predicting both auditory cortex responses and perceived sound dissimilarity. These findings suggest the presence of intermediate acoustic-to-semantic sound representations in the superior temporal gyrus that cannot be accounted for by acoustic or semantic models.
NATURE NEUROSCIENCE
(2023)
Article
Nutrition & Dietetics
Antonietta Canna, Elena Cantone, Anne Roefs, Sieske Franssen, Anna Prinster, Elia Formisano, Francesco Di Salle, Fabrizio Esposito
Summary: In this study, ultra-high field fMRI was used to investigate the neural signals in the Nucleus Tractus Solitarius (NTS) in response to different taste stimuli in the human brainstem. The results demonstrated a significant blood oxygen level-dependent (BOLD) response in the predicted location of the NTS for all basic taste stimuli. This study suggests the potential of using a similar experimental strategy to explore the central nervous system involvement in eating disorders.
FRONTIERS IN NUTRITION
(2023)
Review
Radiology, Nuclear Medicine & Medical Imaging
Dimo Ivanov, Federico De Martino, Elia Formisano, Francisco J. Fritz, Rainer Goebel, Laurentius Huber, Sriranga Kashyap, Valentin G. Kemper, Denizhan Kurban, Alard Roebroeck, Shubharthi Sengupta, Bettina Sorger, Desmond H. Y. Tse, Kamil Uludag, Christopher J. Wiggins, Benedikt A. Poser
Summary: This article reviews the 9.4 T work done in Maastricht, including functional and anatomical imaging experiments. By utilizing specific techniques and optimized coils, the researchers were able to obtain high-quality imaging results and highlight the technical challenges and practical issues associated with ultra-high field MRI.
MAGNETIC RESONANCE MATERIALS IN PHYSICS BIOLOGY AND MEDICINE
(2023)
Review
Radiology, Nuclear Medicine & Medical Imaging
Steve Bates, Serge O. Dumoulin, Paul J. M. Folkers, Elia Formisano, Rainer Goebel, Aidin Haghnejad, Rick C. Helmich, Dennis Klomp, Anja G. van der Kolk, Yi Li, Aart Nederveen, David G. Norris, Natalia Petridou, Stefan Roell, Tom W. J. Scheenen, Menno M. Schoonheim, Ingmar Voogt, Andrew Webb
Summary: We propose a vision for a 14 Tesla MR system, which includes a novel whole-body magnet design using high temperature superconductor, a console and associated electronic equipment, an optimized radiofrequency coil setup for proton measurement in the brain, and a high-performance gradient set. This system has significant applications in neuroscience and medical research, allowing for fine-grained observation of neural activity and structural abnormalities.
MAGNETIC RESONANCE MATERIALS IN PHYSICS BIOLOGY AND MEDICINE
(2023)
Article
Psychology, Mathematical
Shruti Ullas, Elia Formisano, Frank Eisner, Anne Cutler
PSYCHONOMIC BULLETIN & REVIEW
(2020)
Article
Neuroimaging
Eva Berlot, Remo Arts, Jasper Smit, Erwin George, Omer Faruk Gulban, Michelle Moerel, Robert Stokroos, Elia Formisano, Federico De Martino
NEUROIMAGE-CLINICAL
(2020)