Multiple-Kernel Support Vector Machine for Predicting Internet Gaming Disorder Using Multimodal Fusion of PET, EEG, and Clinical Features

Boram Jeong; Jiyoon Lee; Heejung Kim; Seungyeon Gwak; Yu Kyeong Kim; So Young Yoo; Donghwan Lee; Jung-Seok Choi

doi:10.3389/fnins.2022.856510

Multiple-Kernel Support Vector Machine for Predicting Internet Gaming Disorder Using Multimodal Fusion of PET, EEG, and Clinical Features

Front Neurosci. 2022 Jun 30:16:856510. doi: 10.3389/fnins.2022.856510. eCollection 2022.

Authors

Boram Jeong¹, Jiyoon Lee², Heejung Kim^{3

4}, Seungyeon Gwak¹, Yu Kyeong Kim³, So Young Yoo⁵, Donghwan Lee¹, Jung-Seok Choi²

Affiliations

¹ Department of Statistics, Ewha Womans University, Seoul, South Korea.
² Department of Psychiatry, Samsung Medical Center, Seoul, South Korea.
³ Department of Nuclear Medicine, SMG-SNU Boramae Medical Center, Seoul, South Korea.
⁴ Institute of Radiation Medicine, Medical Research Center, Seoul National University, Seoul, South Korea.
⁵ Department of Psychiatry, SMG-SNU Boramae Medical Center, Seoul, South Korea.

Abstract

Internet gaming disorder (IGD) has become an important social and psychiatric issue in recent years. To prevent IGD and provide the appropriate intervention, an accurate prediction method for identifying IGD is necessary. In this study, we investigated machine learning methods of multimodal neuroimaging data including Positron Emission Tomography (PET), Electroencephalography (EEG), and clinical features to enhance prediction accuracy. Unlike the conventional methods which usually concatenate all features into one feature vector, we adopted a multiple-kernel support vector machine (MK-SVM) to classify IGD. We compared the prediction performance of standard machine learning methods such as SVM, random forest, and boosting with the proposed method in patients with IGD (N = 28) and healthy controls (N = 24). We showed that the prediction accuracy of the optimal MK-SVM using three kinds of modalities was much higher than other conventional machine learning methods, with the highest accuracy being 86.5%, the sensitivity 89.3%, and the specificity 83.3%. Furthermore, we deduced that clinical variables had the highest contribution to the optimal IGD prediction model and that the other two modalities were also indispensable. We found that more efficient integration of multimodal data through kernel combination could contribute to better performance of the prediction model. This study is a novel attempt to integrate each method from different sources and suggests that integrating each method, such as self-administrated reports, PET, and EEG, improves the prediction of IGD.

Keywords: Positron Emission Tomography; electroencephalography; integrative analysis; internet gaming disorder; kernel support vector machine; multimodal.