Deep Representational Similarity Learning for Analyzing Neural Signatures in Task-based fMRI Dataset
- PDF / 1,599,005 Bytes
- 15 Pages / 595.224 x 790.955 pts Page_size
- 8 Downloads / 159 Views
ORIGINAL ARTICLE
Deep Representational Similarity Learning for Analyzing Neural Signatures in Task-based fMRI Dataset Muhammad Yousefnezhad1,2
· Jeffrey Sawalha2 · Alessandro Selvitella3 · Daoqiang Zhang1
Accepted: 21 September 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Similarity analysis is one of the crucial steps in most fMRI studies. Representational Similarity Analysis (RSA) can measure similarities of neural signatures generated by different cognitive states. This paper develops Deep Representational Similarity Learning (DRSL), a deep extension of RSA that is appropriate for analyzing similarities between various cognitive tasks in fMRI datasets with a large number of subjects, and high-dimensionality — such as whole-brain images. Unlike the previous methods, DRSL is not limited by a linear transformation or a restricted fixed nonlinear kernel function — such as Gaussian kernel. DRSL utilizes a multi-layer neural network for mapping neural responses to linear space, where this network can implement a customized nonlinear transformation for each subject separately. Furthermore, utilizing a gradientbased optimization in DRSL can significantly reduce runtime of analysis on large datasets because it uses a batch of samples in each iteration rather than all neural responses to find an optimal solution. Empirical studies on multi-subject fMRI datasets with various tasks — including visual stimuli, decision making, flavor, and working memory — confirm that the proposed method achieves superior performance to other state-of-the-art RSA algorithms. Keywords fMRI analysis · Representational similarity analysis · Deep representational similarity learning
Introduction One of the most significant challenges in both neuroscience and machine learning is comprehending how the human brain works (Kriegeskorte et al. 2006; Khaligh-Razavi and Daoqiang Zhang
[email protected] Muhammad Yousefnezhad [email protected] Jeffrey Sawalha [email protected] Alessandro Selvitella [email protected] 1
College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing, 211106, China
2
Department of Computing Science and The Department of Psychiatry, University of Alberta, Edmonton T6G 2R3, AB, Canada
3
Department of Mathematical Sciences, Purdue University Fort Wayne, 2101 E Coliseum Blvd, Fort Wayne, IN 46805, USA
Kriegeskorte 2014). Indeed, we have long been fascinated by the process of conscious thought, which translates to an interest in better understanding human brains. We anticipate this will offer methods for diagnosing and treating mental health disorders, which could have tremendous benefits (Haxby et al. 2014). The neural activities can be analyzed at different levels, but a crucial step is knowing what the similarities (or differences) between distinctive cognitive tasks are Haxby et al. (2014); Yousefnezhad and Zhang (2017a, c). It is like a spotlight that allows us to facilitate other areas of brain studies. Since task-based fu
Data Loading...
