Personalized Contingent Neurostimulation for Epilepsy by Machine Learning in Organic Brain Interfaces

通过有机脑接口中的机器学习对癫痫进行个性化偶然神经刺激

• 批准号: 549625-2020
• 负责人: Genov, Roman
• 金额: $ 18.39万
• 依托单位: University of Toronto
• 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=699978
• 关键词: Personalized; Contingent; Neurostimulation; Epilepsy; Machine

Epilepsy is a chronic, neurological condition characterized by unpredictable, disabling seizures. This devastating disorder increases mortality by 4 times and depression by 2 times population rates. Unfortunately, epilepsy is extremely common, affecting 1 in 100 people, and in 33% of cases, medications do not prevent seizures. An emerging technology to aid such patients with drug-resistant epilepsy is contingent neurostimulation devices. These devices are implanted in a patient so that they can detect epileptiform activity and electrically stimulate the nervous system to prevent/abort seizures. Recently, two commercial contingent neurostimulators have been approved for treating epilepsy. These devices have proven successful at reducing seizure activity, however their efficacy is low. In particular, they are very unlikely to produce seizure freedom, which is necessary to restore patient independence and quality of life. The primary reasons why existing neurostimulation devices have had such limited success are that: (a) they use rudimentary (sparse and large) electrodes that poorly interface with the brain and (b) they are extremely simplistic in their seizure detection and abortion capabilities due to the use of basic electronics as constrained by energy limitations of an implant. We propose to combine (a) the far superior brain-interfacing electrode technology from our partner Panaxium and (b) our energy-efficient state-of-the-art implantable electronic circuits that employ artificial intelligence for optimal seizure prediction and prevention - to greatly improve the seizure-freedom efficacy of the neuromodulation therapy for intractable epilepsy. Our team is uniquely well qualified for this project. It is led by an expert in low-power biomedical and computational electronics (Genov), an epilepsy neurosurgeon and an expert in brain neurophysiology (Valiante), and Panaxiums Technology Commercialization Leader and a 20-year industry veteran (Weinroth).

癫痫是一种慢性神经系统疾病，其特征是不可预测的，致残的， 癫痫发作。这种毁灭性的疾病使死亡率增加了4倍，使抑郁症增加了2倍 人口比率。不幸的是，癫痫是非常常见的，影响1在100人， 在33%的病例中，药物不能预防癫痫发作。一种新兴的技术， 耐药性癫痫患者的神经刺激装置是偶然的。这些设备 被植入病人体内，这样它们就可以检测癫痫样活动， 神经系统以防止/中止癫痫发作。最近，两个商业团队 神经刺激器已被批准用于治疗癫痫。这些设备已经证明 成功地减少癫痫发作活动，但它们的功效很低。特别是， 不太可能产生癫痫发作自由，这是恢复患者独立性所必需的 和生活质量。 现有神经刺激器械成功有限的主要原因 是：（a）他们使用基本的（稀疏和大）电极， 大脑和（B）它们在癫痫发作检测和流产能力方面极其简单 这是由于受植入物的能量限制的约束而使用基本电子器件。我们 建议将（a）我们的远为上级的脑接口电极技术与联合收割机相结合 合作伙伴Panaxium和（B）我们的节能最先进的植入式电子电路 采用人工智能进行最佳癫痫预测和预防-大大 提高了神经调节治疗难治性癫痫的无并发症疗效。 我们的团队非常适合这个项目。它由一位低功耗领域的专家领导， 生物医学和计算电子学（Genov），癫痫神经外科医生和专家 大脑神经生理学（Valiante）和Panaxiums技术商业化领导者 以及拥有20年行业经验的Weinroth。