Individualized Assessment of Cerebral Autoregulation Based on Transfer Function Analysis and Machine Learning Techniques
文献类型：会议论文
| 作者 | Lixuan Lin; Quanli Qiu; Pandeng Zhang; Aqiang Zhang; Jia Liu |
| 出版日期 | 2017 |
| 会议日期 | 2017 |
| 会议地点 | Shenzhen |
| 英文摘要 | Transfer function analysis (TFA) is currently the most widely applied approach to assess cerebral autoregulation (CA). It however can only differentiate patients in groups but not in individuals. In this paper, we proposed a combined method using TFA and a simple machine learning method to quantify cerebral autoregulation for individuals. Firstly, TFA was applied to estimate autoregulatory parameters between arterial blood pressure (ABP) and cerebral blood flow velocity (CBFV). Secondly, we trained a neural network classifier with data in two categories ( healthy controls and symptomatic patients with severe stenosis (>70% and <99%) in cerebral artery, respectively, assuming that autoregulation is intact in the former category and abolished in the latter category, respectively. Finally, we tested data in healthy subject and patients with mild (<50%), moderate (<70%), severe stenosis (<99%) in cerebral artery in order to investigate the individual autoregulation. Our training and testing process achieves 84.1% and 87.3% accuracy, respectively. Comparing to the results from a standalone TFA method, the proposed method shows that autoregulation is more likely scattered within groups, indicating that the degree of stenosis does not necessarily reflects the underlying alteration of hemodynamics. |
| 语种 | 英语 |
| 源URL | [http://ir.siat.ac.cn:8080/handle/172644/12693]  |
| 专题 | 深圳先进技术研究院_数字所 |
| 作者单位 | 2017 |
| 推荐引用方式 GB/T 7714 | Lixuan Lin,Quanli Qiu,Pandeng Zhang,et al. Individualized Assessment of Cerebral Autoregulation Based on Transfer Function Analysis and Machine Learning Techniques[C]. 见:. Shenzhen. 2017. |
入库方式： OAI收割
来源：深圳先进技术研究院
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。