Convolution-GRU Based on Independent Component Analysis for fMRI Analysis with Small and Imbalanced Samples

Abstract

Functional magnetic resonance imaging (fMRI) is a commonly used method of brain research. However, due to the complexity and particularity of the fMRI task, it is difficult to find enough subjects, resulting in a small and, often, imbalanced dataset. A dataset with small samples causes overfitting of the learning model, and the imbalance will make the model insensitive to the minority class, which has been a problem in classification. It is of great significance to classify fMRI data with small and imbalanced samples. In the present study, we propose a 3-step method on a small and imbalanced fMRI dataset from a word-scene memory task. The steps of the method are as follows: (1) An independent component analysis is performed to reduce the dimension of data; (2) The synthetic minority oversampling technique is used to generate new samples of the minority class to balance data; (3) A convolution-Gated Recurrent Unit (GRU) network is used to classify the independent component signals, indicating whether the subjects are performing episodic memory tasks. The accuracy of the proposed method is 72.2%, which improves the classification performance compared with traditional classifiers such as support vector machines (SVM), logistic regression (LGR), linear discriminant analysis (LDA) and k-nearest neighbor (KNN), and this study gives a biomarker for evaluating the reactivation of episodic memory.

The Application of Machine Learning ICA-VMD in an Intelligent Diagnosis System in a Low SNR Environment

“…x 1 (t) = a 11 s 1 (t) + a 12 s 2 (t) + n(t) x 2 (t) = a 21 s 1 (t) + a 22 s 2 (t) + n(t) (30) where a ij (i = 1,2, j = 1, . .…”

“…It has been found that with the development of over-or under-complete representations [25,26], not all raw data can be separated by a small number of sensors. There are many documents on the application of ICA in different fields, and it has many contributions [27][28][29][30][31][32].…”

Understanding and predicting future relapse in depression from resting state functional connectivity and self-referential processing

“…Future studies should replicate these findings in and more balanced datasets for building more reliable and generalizable classifiers. Though SMOTE has been used successfully for this purpose in previous fMRI studies (Eslami and Saeed, 2019;Kawahara et al, 2017;Wang et al, 2020;Yuan et al, 2019), recently concerns have been raised about its efficacy (94). Hence, we also investigated other oversampling techniques (including borderline SMOTE, k-means SMOTE, SVM SMOTE, and ADAptive SYNthetic (ADASYN) sampling) to explore stability of results (See Supplement 2).…”

Machine Learning Methods with Noisy, Incomplete or Small Datasets

“…The proposed algorithm solves the small-data problem by using permutation-variable importance (PVI) and persistent entropy of topological imprints; as well as applying a support vector machine (SVM) classifier to achieve the severity classification of Parkinson disease patients. In [3], Wang et al (China) addressed the problem of small and unbalanced datasets in functional magnetic resonance imaging (fMRI) for neuroscience studies. Their technique combines Independent Component Analysis (ICA) for dimensionality reduction, data augmentation to balance data and a convolution-gated recurrent unit (GRU) network.…”