首页> 外文会议>Asilomar Conference on Signals, Systems and Computers >On developing an FPGA based system for real time seizure prediction
【24h】

On developing an FPGA based system for real time seizure prediction

机译:开发基于FPGA的实时癫痫发作预测系统

获取原文

摘要

For the purposes of designing a neurostimulator system meant to predict and suppress seizures in patients with epilepsy, we have developed a one-class support vector machine (SVM) learning algorithm that accurately predicts upcoming seizures. Trained on patient intracranial electroencephalography (iEEG) data, our algorithm has a high seizure prediction sensitivity while maintaining an acceptably low number of false positives. In order to transition this algorithm to an implantable clinical device, the hardware solution must draw minimal power and conform to a small form factor. With the end goal of designing and fabricating an application specific integrated circuit (ASIC), we have implemented our machine learning algorithmic pipeline on a field programmable gate array (FPGA). This FPGA implementation runs on real-time patient data and maintains all performance specifications.
机译:为了设计旨在预测和抑制癫痫患者癫痫发作的神经刺激器系统,我们开发了一种一类支持向量机(SVM)学习算法,可以准确地预测即将发生的癫痫发作。经过对患者颅内脑电图(iEEG)数据的训练,我们的算法具有较高的癫痫发作预测敏感性,同时保持了可接受的少量误报。为了将该算法过渡到可植入的临床设备,硬件解决方案必须消耗最小的功率并符合较小的尺寸要求。为了设计和制造专用集成电路(ASIC)的最终目标,我们已经在现场可编程门阵列(FPGA)上实现了机器学习算法流水线。该FPGA实现基于实时患者数据运行,并保持所有性能指标。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号