A multimodal seizure detection artificial intelligence-based smart wear

基于多模态癫痫检测人工智能的智能穿戴

• 批准号: 538852-2019
• 负责人: Nguyen, DangKhoa
• 金额: $ 19.92万
• 依托单位: Université de Montréal
• 依托单位国家: 加拿大
• 项目类别: Collaborative Health Research Projects
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=697785
• 关键词: multimodal; seizure; detection; artificial; intelligence

Epilepsy is a chronic neurological condition that affects as many as 1 in every 100 Canadians. With the ageing of the population, the prevalence and incidence of epilepsy among the elderly is greater than any other age group due to the increasing number brain insults. The first line of treatment consists of long-term drug therapy but more than a third of patients are said to be drug-resistant and continue to suffer from disabling seizures. Due to their unpredictable nature, uncontrolled seizures represent a major personal handicap and source of worriment for patients (risks of injury and death). In addition, persistent seizures constitute a considerable public health burden due to high use of health care resources, high number of disability days or unemployment, and low annual income. For these patients, accurate and rapid detection of seizures could significantly improve their care. Indeed, a system capable of detecting seizures could alert family members, caretakers or medical personnel to intervene in a timely fashion to limit the risk of injuries and death. Hence, the main objective of this project is to develop a user-friendly artificial intelligence-based seizure detection system based on non-invasive multimodal physiological signals obtained through smart wearable devices. Social, ethical, environmental, economic, and legal implications will be taken into account to ensure a smooth and rapid translation to patients with epilepsy. Potential applications are numerous for the benefit of patients (reducing anxiety and enhancing selfconfidence), caretakers (evaluation and surveillance), and the healthcare system (reducing associated costs).

癫痫是一种慢性神经系统疾病，每100名加拿大人中就有1人患有癫痫。随着人口的老龄化，老年人癫痫的患病率和发病率高于其他年龄组，这是由于脑损伤的数量不断增加。一线治疗包括长期药物治疗，但据说超过三分之一的患者具有抗药性，并继续遭受致残性癫痫发作。由于其不可预测的性质，不受控制的癫痫发作代表了患者的主要个人障碍和担忧来源（受伤和死亡的风险）。此外，由于大量使用卫生保健资源，大量的药物滥用， 残疾日或失业，以及低年收入。对于这些患者，准确和快速地检测癫痫发作可以显着改善他们的护理。事实上，一个能够检测癫痫发作的系统可以提醒家庭成员、看护人或医务人员及时进行干预，以限制受伤和死亡的风险。因此，本报告的主要目的 该项目是开发一种用户友好的基于人工智能的癫痫发作检测系统，该系统基于通过智能可穿戴设备获得的非侵入性多模态生理信号。将考虑社会、伦理、环境、经济和法律的影响，以确保顺利、快速地向癫痫患者转化。潜在的应用有很多，有利于患者（减少焦虑和增强自信），护理人员（评估和监测）和医疗保健系统（降低相关成本）。