A Novel Seizure Detection Method Based on the Feature Fusion of Multimodal Physiological Signals

Seizure detection is traditionally done using video/electroencephalography monitoring, but for out-of-hospital patients, this method is costly. In recent years, portable device to detect seizures gains attention. In this article, multimodal signals collected by portable devices are studied, and a seizure detection algorithm is proposed based on adaptive multibit local differential ternary pattern (MLDTP). This algorithm is used for detecting seizure period and interseizure period. Traditional local binary pattern has certain limitations in describing 1-D time-series signals. It can only describe two types of structures in signals: 1) rising structure and 2) falling structure, making the signal patterns overly monotonous and not conducive to classification tasks. To address this issue, this article introduces two additional structures, slowly rising structure and slowly falling structure, into the signal description using the MLDTP method. This method constructs multibit neighboring relationships of the signals and adaptively selects the optimal MLDTP parameters for different modalities using the Archimedes optimization algorithm (AOA). Additionally, this article extensively discusses a multimodal signal fusion strategy, mapping features of different modal signals to the same feature space through the MLDTP algorithm to achieve information complementarity. Long-term recorded data from 18 patients were collected using the wearable device Biovital P1, with 13 cases from the Children's Hospital affiliated with Children's Hospital, Zhejiang University School of Medicine, and five cases from the fourth Affiliated Hospital of Anhui Medical University. The data set underwent fivefold cross-validation, resulting in average accuracy, precision, sensitivity, and F1 score of 96.81%, 98.55%, 95.24%, and 96.87%, respectively.