Classification of ADHD with fMRI data and multi-objective optimization
Graphical abstract
Introduction
Attention deficit hyperactivity disorder (ADHD) is a common childhood disorders which lasts to adulthood in most cases. It is defined as a neurodevelopmental disorder in the fifth edition of diagnostic and statistical manual of mental disorders, mainly characterized by attention deficits, excessive activity and behavioral impulses [1]. The worldwide pooled prevalence of ADHD is reported to be 3.4% in children and adolescents [2]. So far, the etiology and pathogenesis of ADHD is still not clear [3], and the diagnosis of ADHD currently relies mainly on the subjective experience of doctors. Therefore objective diagnosis and effective treatment of ADHD is one of the significant topics in the field of neuroscience.
Medical imaging technologies such as electroencephalography, functional near-infrared spectroscopy, magnetic resonance imaging (MRI), and functional magnetic resonance imaging (fMRI) have been used for computer-aided diagnosis of ADHD. see, e.g., [4], [5], [6], [7], [8], [9]. As a method of fMRI, resting-state fMRI (rs-fMRI) has shown prominent advantages in the pathological analysis of psychiatric diseases. Various feature extraction, selection and classification methods have been used in rs-fMRI based disease diagnosis. Castellanos et al. [10] found functional connectivity (FC) information of fMRI can be a prominent feature for ADHD diagnosis. Du et al. [11] used graph kernel principal components analysis (PCA) to extract features and proposed a discriminative subnetwork to classify ADHD. Miao and Zhang [12] discussed the classification of ADHD with fMRI data, and used relief algorithm to obtain a subset of fractional amplitude of low-frequency fluctuation features. Itani et al. [13] proposed a multi-level approach based on decision trees for ADHD classification. Qureshi et al. [14] computed the global connectivity maps of fMRI and used hierarchical extreme learning machine to classify ADHD. Riaz et al. [15] integrated non-imaging and imaging data and used a machine learning framework to study alterations of functional connectivity between ADHD and normal control (NC) subjects. Considering the data imbalanced property, they generated synthetic minority class samples using synthetic minority over-sampling technique (SMOTE) [16]. It needs to be noted that recently multi-objective evolutionary computation algorithms, such as multi-objective particle swarm algorithm [17] and multi-objective self-adaptive particle swarm algorithms [18] etc. have shown some success in feature selection (see [19] for a survey) and can also be used for rs-fMRI feature selection.
Most of the classification algorithms mentioned above make the assumption of well-balanced training datasets and equal misclassification costs. However, dataset imbalance is a critical problem in ADHD rs-fMRI datasets. Due to imbalanced learning, imbalanced datasets may lead to overfocus on the majority class, hence degrade the performance of a classifier. There are many approaches proposed to handle the imbalanced dataset problem (also called the class imbalance problem), see [20], [21], [22], [23]. These approaches basically fall into two major categories: data level approaches and algorithm level approaches. The main idea of data level approaches to handle the imbalance is to resample the training set, whereas algorithm level approaches normally adopt the idea of introducing unequal misclassification costs in decision making process. In ADHD classification, data level approaches such as SMOTE have been used to handle the dataset imbalance. However, by performing random oversampling the minority or under-sampling the majority class, these strategies for creating balanced training datasets may lead to suboptimal performance [24].
Considering the multi-objective nature of classification problems, Shao et al. [25] have proposed a bi-objective classification method for the classification of ADHD. However, dataset imbalance is not considered. In this work, incorporating a decision maker's preference, we propose a novel reference point based three objective optimization classification scheme. The main contributions of our work are as follows.
- •
We propose using a cost sensitive three objective classification model which is based on SVM to handle the ADHD dataset imbalance problem.
- •
From a practical viewpoint, an interactive multi-objective optimization method incorporating decision maker's preference information is proposed. A preferred subset of pareto optimal classifiers can be obtained, thus a classifier with the best performance can be selected.
The rest of the paper is organized as follows. In Section 2, we first introduce the acquisition and preprocessing of data. Then the three objective classification scheme for ADHD diagnosis is proposed. As a main part of the scheme, a three objective classification model based on L1-norm SVM is introduced and a reference point based multi-objective optimization method is proposed. Section 3 shows some computational experiments. An interactive multi-objective decision making example and some comparison results of the three objective classification scheme and some other methods are given. In Section 4, we give some further discussions about the results. Finally we draw the conclusion in Section 5.
Section snippets
Data and preprocessing
In this study, the rs-fMRI datasets are downloaded from the Neuro Bureau ADHD-200 consortium (http://fcon1000projectsnitrc.org/) [26]. Datasets are from three sites, they are Kennedy Krieger Institute (KKI), New York University Medical Center (NYU) and Peking University (Peking), respectively. Five datasets, namely, KKI, NYU and Peking-1, Peking-2 and Peking-joint are used in our experiment, where Peking-joint consists of three datasets Peking-1, Peking-2 and Peking-3. There are four kinds of
Results
We use our proposed reference point based multi-objective classification scheme to classify the five datasets from ADHD-200 consortium. Each dataset was randomly stratified into three datasets namely training set, validation set, and testing set. The ratio 6:2:2 is used, i.e., 60% of each dataset is used to train the model, 20% is used as the validation set to select the classifier and 20% is used for testing.
We take Peking-1 dataset as an example to describe the classification process in
Discussion
In our experiment, KKI and Peking-1 datasets in the ADHD-200 consortium are small and highly imbalanced. For KKI, the total number of samples is 83, and only 22 samples are ADHD samples; while for Peking-1 dataset, there are 85 samples in total, and only 24 samples are positive.
Among the four traditional machine learning methods RF, ELM, L1SVM and L2SVM, RF method obtained higher average accuracy values than ELM, L1SVM and L2SVM methods on the KKI, NYU, Peking-1, and Peking-joint datasets.
Conclusion
In this paper we have proposed a reference point based multi-objective classification scheme to classify ADHD. Our scheme uses a three objective SVM formulation based on L1-norm. It considers the empirical errors for positive and negative samples separately, thus the class imbalance problem can be handled effectively from algorithm level. Furthermore, considering a decision making process, normally a decision maker has her/his own preferences. Therefore, we adopted an interactive
Declaration of Competing Interest
The authors do not have financial and personal relationships with other people or organizations that could inappropriately influence (bias) their work.
Acknowledgments
This work was supported by the Scientific and Technological Innovation Foundation of Shunde Graduate School, University of Science and Technology Beijing (No. BK19CE017), and the National Environmental Corrosion Platform of China. We also would like to thank the associate editor and the anonymous reviewers for their insightful and detailed comments, which have greatly improved the quality of our article.
References (40)
- et al.
Attributed graph distance measure for automatic detection of attention deficit hyperactive disordered subjects
Front. Neural Circuits
(2014) - et al.
Classification of ADHD and non-ADHD subjects using a universal background model
Biomed. Signal Process. Control
(2018) - et al.
Cingulate-precuneus interactions: a new locus of dysfunction in adult attention-deficit/hyperactivity disorder
Biol. Psychiatry
(2008) - et al.
Network-based classification of ADHD patients using discriminative subnetwork selection and graph kernel PCA
Comput. Med. Imaging Graph.
(2016) - et al.
A multi-level classification frame-work for multi-site medical data: application to the ADHD-200 collection
Exp. Syst. Appl.
(2018) - et al.
Fusion of fMRI and non-imaging data for ADHD classification
Comput. Med. Imaging Graph.
(2018) - et al.
An improved sup- port vector machine-based diabetic readmission prediction
Comput. Methods Progr. Biomed.
(2018) - et al.
Feature rearrangement based deep learning system for predicting heart failure mortality
Comput. Methods Progr. Biomed.
(2020) - et al.
Classification of ADHD with bi-objective optimization
J. Biomed. Inform.
(2018) - et al.
The neuro bureau ADHD-200 preprocessed repository
Neuroimage
(2017)
Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain
Neuroimage
Principal components analysis and track quality index: a machine learning approach
Transp. Res. C: Emerg. Technol.
Exploring the trade-off between generalization and empirical errors in a one-norm SVM
Eur. J. Oper. Res.
Discrete representation of non-dominated sets in multi-objective linear programming
Eur. J. Oper. Res.
Diagnostic and Statistical Manual of Mental Disorders: DSM- 5
Annual research review: a meta-analysis of the worldwide prevalence of mental disorders in children and adolescents
J. Child Psychol. Psychiatry
Is the Theta/Beta EEG marker for ADHD inherently flawed?
J. Attent. Disord.
ADHD classification by a texture analysis of anatomical brain MRI data
Front. Syst. Neuro- Sci.
Extreme learning machine-based classification of ADHD using brain structural MRI data
PLOS ONE
Design of multichannel functional near-infrared spectroscopy system with application to propofol and sevoflurane anesthesia monitoring
Neurophotonics
Cited by (16)
Automated detection of ADHD: Current trends and future perspective
2022, Computers in Biology and MedicineA review of ADHD detection studies with machine learning methods using rsfMRI data
2024, NMR in BiomedicineMultiple measurement analysis of resting-state fMRI for ADHD classification in adolescent brain from the ABCD study
2023, Translational Psychiatry