Euler Elastica Regularized Logistic Regression for Whole-Brain Decoding of fMRI Data

Objective: Multivariate pattern analysis methods have been widely applied to functional magnetic resonance imaging (fMRI) data to decode brain states. Due to the “high features, low samples” in fMRI data, machine learning methods have been widely regularized using various regularizations to avoid overfitting. Both total variation (TV) using the gradients of images and Euler's elastica (EE) using the gradient and the curvature of images are the two popular regulations with spatial structures. In contrast to TV, EE regulation is able to overcome the disadvantage of TV regulation that favored piecewise constant images over piecewise smooth images. In this study, we introduced EE to fMRI-based decoding for the first time and proposed the EE regularized multinomial logistic regression (EELR) algorithm for multi-class classification. Methods: We performed experimental tests on both simulated and real fMRI data to investigate the feasibility and robustness of EELR. The performance of EELR was compared with sparse logistic regression (SLR) and TV regularized LR (TVLR). Results: The results showed that EELR was more robustness to noises and showed significantly higher classification performance than TVLR and SLR. Moreover, the forward models and weights patterns revealed that EELR detected larger brain regions that were discriminative to each task and activated by each task than TVLR. Conclusion: The results suggest that EELR not only performs well in brain decoding but also reveals meaningful discriminative and activation patterns. Significance: This study demonstrated that EELR showed promising potential in brain decoding and discriminative/activation pattern detection.

[1]  Bertrand Thirion,et al.  Multi-scale Mining of fMRI Data with Hierarchical Structured Sparsity , 2011, 2011 International Workshop on Pattern Recognition in NeuroImaging.

[2]  David D. Cox,et al.  Functional magnetic resonance imaging (fMRI) “brain reading”: detecting and classifying distributed patterns of fMRI activity in human visual cortex , 2003, NeuroImage.

[3]  Vince D. Calhoun,et al.  Task-specific feature extraction and classification of fMRI volumes using a deep neural network initialized with a deep belief network: Evaluation using sensorimotor tasks , 2017, NeuroImage.

[4]  Jia Liu,et al.  The Hierarchical Brain Network for Face Recognition , 2013, PloS one.

[5]  D. Mumford.  Elastica and Computer Vision , 1994 .

[6]  Jin Fan,et al.  Involvement of the anterior cingulate and frontoinsular cortices in rapid processing of salient facial emotional information , 2011, NeuroImage.

[7]  Jonathan E. Taylor,et al.  Interpretable whole-brain prediction analysis with GraphNet , 2013, NeuroImage.

[8]  David M. Groppe,et al.  Seeing Scenes: Topographic Visual Hallucinations Evoked by Direct Electrical Stimulation of the Parahippocampal Place Area , 2014, The Journal of Neuroscience.

[9]  Bertrand Thirion,et al.  Adaptive multi-class Bayesian sparse regression - An application to brain activity classification , 2009 .

[10]  John Shawe-Taylor,et al.  SCoRS—A Method Based on Stability for Feature Selection and Mapping in Neuroimaging , 2014, IEEE Transactions on Medical Imaging.

[11]  Yutaka Hatakeyama,et al.  Decoding analysis for fMRI based on Deep Brief Network , 2014, 2014 World Automation Congress (WAC).

[12]  Tony F. Chan,et al.  Euler's Elastica and Curvature-Based Inpainting , 2003, SIAM J. Appl. Math..

[13]  R. Schultz.  Developmental deficits in social perception in autism: the role of the amygdala and fusiform face area , 2005, International Journal of Developmental Neuroscience.

[14]  Lars Kai Hansen,et al.  Model sparsity and brain pattern interpretation of classification models in neuroimaging , 2012, Pattern Recognit..

[15]  N. Kanwisher,et al.  The Fusiform Face Area: A Module in Human Extrastriate Cortex Specialized for Face Perception , 1997, The Journal of Neuroscience.

[16]  Jody Tanabe,et al.  See Blockindiscussions, Blockinstats, Blockinand Blockinauthor Blockinprofiles Blockinfor Blockinthis Blockinpublication Comparison Blockinof Blockindetrending Blockinmethods Blockinfor Optimal Blockinfmri Blockinpreprocessing , 2022 .

[17]  L. Ambrosio,et al.  On a Variational Problem Arising in Image Reconstruction , 2003 .

[18]  Brian A. Wandell,et al.  Population receptive field estimates in human visual cortex , 2008, NeuroImage.

[19]  Berthold K. P. Horn.  The Curve of Least Energy , 1983, TOMS.

[20]  Gaël Varoquaux,et al.  Total Variation Regularization for fMRI-Based Prediction of Behavior , 2011, IEEE Transactions on Medical Imaging.

[21]  Anthony J. Yezzi,et al.  Curve evolution implementation of the Mumford-Shah functional for image segmentation, denoising, interpolation, and magnification , 2001, IEEE Trans. Image Process..

[22]  A. Love.  A treatise on the mathematical theory of elasticity , 1892 .

[23]  Hongbin Zha,et al.  Total Variation and Euler's Elastica for Supervised Learning , 2012, ICML.

[24]  Jack L. Gallant,et al.  Encoding and decoding in fMRI , 2011, NeuroImage.

[25]  Russell A. Epstein,et al.  The Parahippocampal Place Area Recognition, Navigation, or Encoding? , 1999, Neuron.

[26]  Shin Ishii,et al.  Deep learning of fMRI big data: a novel approach to subject-transfer decoding , 2015, ArXiv.

[27]  Masa-aki Sato,et al.  Sparse estimation automatically selects voxels relevant for the decoding of fMRI activity patterns , 2008, NeuroImage.

[28]  Jean-Michel Morel,et al.  Level lines based disocclusion , 1998, Proceedings 1998 International Conference on Image Processing. ICIP98 (Cat. No.98CB36269).

[29]  Synho Do,et al.  How much data is needed to train a medical image deep learning system to achieve necessary high accuracy , 2015, 1511.06348.

[30]  Juan Shi,et al.  Graph Cuts for Curvature Based Image Denoising , 2011, IEEE Transactions on Image Processing.

[31]  Semir Zeki,et al.  Neural Correlates of Hate , 2008, PloS one.

[32]  Kaustubh Supekar,et al.  Sparse logistic regression for whole-brain classification of fMRI data , 2010, NeuroImage.

[33]  Michael Eickenberg,et al.  Grouping Total Variation and Sparsity: Statistical Learning with Segmenting Penalties , 2015, MICCAI.

[34]  J. Sergent,et al.  Functional neuroanatomy of face and object processing. A positron emission tomography study. , 1992, Brain : a journal of neurology.

[35]  Lawrence Carin,et al.  Sparse multinomial logistic regression: fast algorithms and generalization bounds , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[36]  G. M.,et al.  A Treatise on the Mathematical Theory of Elasticity , 1906, Nature.

[37]  Amy L. Thomas,et al.  The neural response to facial attractiveness. , 2009, Neuropsychology.

[38]  Zhiying Long,et al.  Improved sparse decomposition based on a smoothed L0 norm using a Laplacian kernel to select features from fMRI data , 2015, Journal of Neuroscience Methods.

[39]  Lisa M Shin,et al.  Exaggerated amygdala response to masked facial stimuli in posttraumatic stress disorder: a functional MRI study , 2000, Biological Psychiatry.

[40]  C. Fraser,et al.  Mathematical Technique and Physical Conception in Euler's Investigation of the Elastica , 1991 .

[41]  Francisco Pereira,et al.  Information mapping with pattern classifiers: A comparative study , 2011, NeuroImage.

[42]  Luca Baldassarre,et al.  Structured Sparsity Models for Brain Decoding from fMRI Data , 2012, 2012 Second International Workshop on Pattern Recognition in NeuroImaging.

[43]  Gaël Varoquaux,et al.  Identifying Predictive Regions from fMRI with TV-L1 Prior , 2013, 2013 International Workshop on Pattern Recognition in Neuroimaging.

[44]  Stefan Haufe,et al.  On the interpretation of weight vectors of linear models in multivariate neuroimaging , 2014, NeuroImage.

[45]  Tom Heskes,et al.  Efficient Bayesian multivariate fMRI analysis using a sparsifying spatio-temporal prior , 2010, NeuroImage.

[46]  Xue-Cheng Tai,et al.  A Fast Algorithm for Euler's Elastica Model Using Augmented Lagrangian Method , 2011, SIAM J. Imaging Sci..

[47]  Sean M. Polyn,et al.  Beyond mind-reading: multi-voxel pattern analysis of fMRI data , 2006, Trends in Cognitive Sciences.

[48]  David Mumford,et al.  Filtering, Segmentation and Depth , 1993, Lecture Notes in Computer Science.