Epileptic seizure prediction system using wearable HRV sensor

2016 Fiscal Year Final Research Report

• Project/Area Number: 25282175
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Rehabilitation science/Welfare engineering
• Research Institution: Tokyo Medical and Dental University
• Principal Investigator: Miho Miyajima 東京医科歯科大学, 医歯(薬)学総合研究科, 助教 (70616177)
• Co-Investigator(Kenkyū-buntansha): 藤原 幸一 京都大学, 情報学研究科, 助教 (10642514) ; 山川 俊貴 熊本大学, 学内共同利用施設等, 助教 (60510419)
• Co-Investigator(Renkei-kenkyūsha): KANO Manabu 京都大学, 情報学研究科, 教授 (60510419) ; MATSUURA Masato 東京医科歯科大学, 名誉教授 (60134673) ; MAEHARA Taketoshi 東京医科歯科大学, 医歯学総合研究科, 教授 (40211560)
• Research Collaborator: INAJI Motoki 東京医科歯科大学, 医学部附属病院, 講師 (00422486) ; SASANO Tetsuo 東京医科歯科大学, 大学院保健衛生学研究科, 准教授 (00466898) ; WATANABE Satsuki 国立精神・神経医療研究センター病院, 精神科, 医師 (30796016) ; SAKUMA Taeko 東京医科大学, 医学部, 助教 (70419026) ; JIN Kazutaka 東北大学, 医学系研究科, 教授 (20436091) ; NAKAZATO Nobukazu 東北大学, 医学系研究科, 教授 (80207753)
• Project Period (FY): 2013-04-01 – 2017-03-31
• Keywords: てんかん発作兆候検知 / 心拍変動解析（HRV） / ウェアラブルセンサ / 多変量統計的プロセス管理（MSPC） / 生活の質（QOL）

Outline of Final Research Achievements

Patients with intractable epilepsy suffer from accidents and injuries associated with epileptic seizures. To prevent the seizure-associated accidents and improve quality of life (QOL) of epileptic patients, the present work proposes a new real-time epileptic seizure prediction and alert system employing a wearable heart rate variability (HRV) telemeter and a smartphone. The R-R interval (RRI) data is stored into a smartphone via a Bluetooth wireless transmission. The smartphone application for epileptic seizure prediction detect peri-ictal status based on multivariate statistical process control (MSPC) for the HRV data and alert the patients and the caregivers to the seizure. The HRV-based seizure prediction algorithm demonstrated sensitivity of 91% for partial seizures, that is, competitive performance with electroencephalography (EEG)-based methods. The possibility of realizing a HRV-based epileptic seizure prediction system with high wearability was shown.

Free Research Field

臨床てんかん学