Integrated space\u2013frequency\u2013time domain feature extraction for MEG-based Alzheimer\u2019s disease classification

Su Yang,Ricardo Bruña Fernandez,Jose Miguel Sanchez Bornot,Farzin Deravi,Girijesh Prasad,Kongfatt Wong-Lin

doi:10.1186/s40708-021-00145-1

Abstract

Magnetoencephalography (MEG) has been combined with machine learning techniques, to recognize the Alzheimer’s disease (AD), one of the most common forms of dementia. However, most of the previous studies are limited to binary classification and do not fully utilize the two available MEG modalities (extracted using magnetometer and gradiometer sensors). AD consists of several stages of progression, this study addresses this limitation by using both magnetometer and gradiometer data to discriminate between participants with AD, AD-related mild cognitive impairment (MCI), and healthy control (HC) participants in the form of a three-class classification problem. A series of wavelet-based biomarkers are developed and evaluated, which concurrently leverage the spatial, frequency and time domain characteristics of the signal. A bimodal recognition system based on an improved score-level fusion approach is proposed to reinforce interpretation of the brain activity captured by magnetometers and gradiometers. In this preliminary study, it was found that the markers derived from gradiometer tend to outperform the magnetometer-based markers. Interestingly, out of the total 10 regions of interest, left-frontal lobe demonstrates about 8% higher mean recognition rate than the second-best performing region (left temporal lobe) for AD/MCI/HC classification. Among the four types of markers proposed in this work, the spatial marker developed using wavelet coefficients provided the best recognition performance for the three-way classification. Overall, the proposed approach provides promising results for the potential of AD/MCI/HC three-way classification utilizing the bimodal MEG data.

Highlights

There has been a rapid development of machine learning (ML) techniques applied to electroencephalography (EEG) and magnetoencephalography (MEG) data for the diagnosis of dementia, and its most common form, the Alzheimer’s disease (AD) [1]
5 Evaluation and discussion In this work, we proposed a hybrid method to combine the advantages of leave-one-out cross-validation (LOOCV) and Monte Carlo random sampling cross-validation (MCRSCV)
6 Conclusion and future work In this work, we presented a bimodal recognition algorithm to explore the effectiveness of three-class (AD vs. mild cognitive impairment (MCI) vs. healthy control (HC)) classification based on MEG modalities

Summary

Introduction

There has been a rapid development of machine learning (ML) techniques applied to electroencephalography (EEG) and magnetoencephalography (MEG) data for the diagnosis of dementia, and its most common form, the Alzheimer’s disease (AD) [1]. Measures of 25 labelled cortical regions were used to generate two separate feature vectors (one per hemisphere), and a tenfold cross-validation was used to evaluate the algorithm for parameter tuning, using 80% of the data In this scenario, a high recognition rate of 96.54% was reported for the classification between AD and HC, while the performance degraded to 90.26% for MCI vs HC. The major focus of this work is to address the three-class classification problem using a series of newly developed wavelet-based biomarkers (features) to leverage the unique advantages of MEG-related signals. The images were fed into a series of two-dimensional wavelet packet decomposition (WPD) filter banks, where they were used for the preliminary feature extraction [17] The details of this approach are presented in Sect.

MN m n

The optimal features and performances are highlighted in bold

Actual HC

Findings

Magnetometer Gradiometer Score fusion

Conclusion and future work