A study on the combination of functional connection features and Riemannian manifold in EEG emotion recognition
IntroductionAffective computing is the core for Human-computer interface (HCI) to be more intelligent, where electroencephalogram (EEG) based emotion recognition is one of the primary research orientations. Besides, in the field of brain-computer interface, Riemannian manifold is a highly robust and...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-01-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2023.1345770/full |
| _version_ | 1797355303272972288 |
|---|---|
| author | Minchao Wu Minchao Wu Rui Ouyang Chang Zhou Zitong Sun Fan Li Ping Li |
| author_facet | Minchao Wu Minchao Wu Rui Ouyang Chang Zhou Zitong Sun Fan Li Ping Li |
| author_sort | Minchao Wu |
| collection | DOAJ |
| description | IntroductionAffective computing is the core for Human-computer interface (HCI) to be more intelligent, where electroencephalogram (EEG) based emotion recognition is one of the primary research orientations. Besides, in the field of brain-computer interface, Riemannian manifold is a highly robust and effective method. However, the symmetric positive definiteness (SPD) of the features limits its application.MethodsIn the present work, we introduced the Laplace matrix to transform the functional connection features, i.e., phase locking value (PLV), Pearson correlation coefficient (PCC), spectral coherent (COH), and mutual information (MI), to into semi-positive, and the max operator to ensure the transformed feature be positive. Then the SPD network is employed to extract the deep spatial information and a fully connected layer is employed to validate the effectiveness of the extracted features. Particularly, the decision layer fusion strategy is utilized to achieve more accurate and stable recognition results, and the differences of classification performance of different feature combinations are studied. What's more, the optimal threshold value applied to the functional connection feature is also studied.ResultsThe public emotional dataset, SEED, is adopted to test the proposed method with subject dependent cross-validation strategy. The result of average accuracies for the four features indicate that PCC outperform others three features. The proposed model achieve best accuracy of 91.05% for the fusion of PLV, PCC, and COH, followed by the fusion of all four features with the accuracy of 90.16%.DiscussionThe experimental results demonstrate that the optimal thresholds for the four functional connection features always kept relatively stable within a fixed interval. In conclusion, the experimental results demonstrated the effectiveness of the proposed method. |
| first_indexed | 2024-03-08T14:09:12Z |
| format | Article |
| id | doaj.art-9b7c1d1c41564a2683bcd85435b05362 |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-03-08T14:09:12Z |
| publishDate | 2024-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-9b7c1d1c41564a2683bcd85435b053622024-01-15T04:20:38ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2024-01-011710.3389/fnins.2023.13457701345770A study on the combination of functional connection features and Riemannian manifold in EEG emotion recognitionMinchao Wu0Minchao Wu1Rui Ouyang2Chang Zhou3Zitong Sun4Fan Li5Ping Li6Anhui Province Key Laboratory of Multimodal Cognitive Computation, School of Computer Science and Technology, Anhui University, Hefei, ChinaKey Laboratory of Flight Techniques and Flight Safety, Civil Aviation Flight University of China, Guanghan, ChinaAnhui Province Key Laboratory of Multimodal Cognitive Computation, School of Computer Science and Technology, Anhui University, Hefei, ChinaAnhui Province Key Laboratory of Multimodal Cognitive Computation, School of Computer Science and Technology, Anhui University, Hefei, ChinaAnhui Province Key Laboratory of Multimodal Cognitive Computation, School of Computer Science and Technology, Anhui University, Hefei, ChinaKey Laboratory of Flight Techniques and Flight Safety, Civil Aviation Flight University of China, Guanghan, ChinaAnhui Province Key Laboratory of Multimodal Cognitive Computation, School of Computer Science and Technology, Anhui University, Hefei, ChinaIntroductionAffective computing is the core for Human-computer interface (HCI) to be more intelligent, where electroencephalogram (EEG) based emotion recognition is one of the primary research orientations. Besides, in the field of brain-computer interface, Riemannian manifold is a highly robust and effective method. However, the symmetric positive definiteness (SPD) of the features limits its application.MethodsIn the present work, we introduced the Laplace matrix to transform the functional connection features, i.e., phase locking value (PLV), Pearson correlation coefficient (PCC), spectral coherent (COH), and mutual information (MI), to into semi-positive, and the max operator to ensure the transformed feature be positive. Then the SPD network is employed to extract the deep spatial information and a fully connected layer is employed to validate the effectiveness of the extracted features. Particularly, the decision layer fusion strategy is utilized to achieve more accurate and stable recognition results, and the differences of classification performance of different feature combinations are studied. What's more, the optimal threshold value applied to the functional connection feature is also studied.ResultsThe public emotional dataset, SEED, is adopted to test the proposed method with subject dependent cross-validation strategy. The result of average accuracies for the four features indicate that PCC outperform others three features. The proposed model achieve best accuracy of 91.05% for the fusion of PLV, PCC, and COH, followed by the fusion of all four features with the accuracy of 90.16%.DiscussionThe experimental results demonstrate that the optimal thresholds for the four functional connection features always kept relatively stable within a fixed interval. In conclusion, the experimental results demonstrated the effectiveness of the proposed method.https://www.frontiersin.org/articles/10.3389/fnins.2023.1345770/fullemotion recognitionhuman-computer interface (HCI)electroencephalogram (EEG)functional connection featureRiemannian manifolddecision fusion |
| spellingShingle | Minchao Wu Minchao Wu Rui Ouyang Chang Zhou Zitong Sun Fan Li Ping Li A study on the combination of functional connection features and Riemannian manifold in EEG emotion recognition Frontiers in Neuroscience emotion recognition human-computer interface (HCI) electroencephalogram (EEG) functional connection feature Riemannian manifold decision fusion |
| title | A study on the combination of functional connection features and Riemannian manifold in EEG emotion recognition |
| title_full | A study on the combination of functional connection features and Riemannian manifold in EEG emotion recognition |
| title_fullStr | A study on the combination of functional connection features and Riemannian manifold in EEG emotion recognition |
| title_full_unstemmed | A study on the combination of functional connection features and Riemannian manifold in EEG emotion recognition |
| title_short | A study on the combination of functional connection features and Riemannian manifold in EEG emotion recognition |
| title_sort | study on the combination of functional connection features and riemannian manifold in eeg emotion recognition |
| topic | emotion recognition human-computer interface (HCI) electroencephalogram (EEG) functional connection feature Riemannian manifold decision fusion |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2023.1345770/full |
| work_keys_str_mv | AT minchaowu astudyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT minchaowu astudyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT ruiouyang astudyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT changzhou astudyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT zitongsun astudyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT fanli astudyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT pingli astudyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT minchaowu studyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT minchaowu studyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT ruiouyang studyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT changzhou studyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT zitongsun studyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT fanli studyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition AT pingli studyonthecombinationoffunctionalconnectionfeaturesandriemannianmanifoldineegemotionrecognition |