《厦门大学学报（自然科学版）》

有研究表明禅修冥想有益于现代人的身心健康,为了研发能有助于禅修的冥想神经反馈系统,采集乐易心法七日禅学员脑电图信号,提取15维脑电图特征,以禅修导师给学员的评分作为标记,对数据进行个性化校准,测试多种分类和回归算法对学员水平的分类性能.实验结果表明,个性化校准方案可以有效解决脑电图研究中的个体差异问题,15维脑电图特征数据经校准后可以使随机森林等分类算法以93%以上的准确率识别出高水平禅修者,为更加智能的冥想神经反馈系统的研发提供了支持.

Meditation benefits the health of modern people.In order to develop smart neurofeedback systems which help the Zen practice,several electroencephalograph indexes and three advanced analyzing methods are used to classify good or bad new Zen practitioners based on the evaluation made by Zen master.Individualized calibration processes are conducted.Results indicate that 15-dimensional EEG feature and individualized calibration process can effectively solve problems of individual difference of EEG and the accuracy of the classification algorithms can achieve higher than 93% for identifying good meditation practice.Results support the development of smarter meditation neurofeedback systems.

引言
1 实验方法
2 数据处理
3 数据挖掘及结果
4 讨论
5 结论

图1 实验流程图<br/>Fig.1 The flowchart of experiment

图1 实验流程图
Fig.1 The flowchart of experiment

图2 3类冥想提示材料<br/>Fig.2 Instructions of 3 types of meditation

图2 3类冥想提示材料
Fig.2 Instructions of 3 types of meditation

表1 禅修水平得分分布表<br/>Tab.1 Zen training score distribution

表1 禅修水平得分分布表
Tab.1 Zen training score distribution

图3 电极位置
Fig.3 Electrode positions

表2 未经校准过程的数据挖掘结果<br/>Tab.2 Data mining result without calibration

表2 未经校准过程的数据挖掘结果
Tab.2 Data mining result without calibration

表3 相对校准后数据挖掘结果<br/>Tab.3 Data mining result with relative calibration

表3 相对校准后数据挖掘结果
Tab.3 Data mining result with relative calibration

表4 绝对校准后数据挖掘结果<br/>Tab.4 Data mining result with absolute calibration

表4 绝对校准后数据挖掘结果
Tab.4 Data mining result with absolute calibration

表5 绝对校准前后二分类准确率结果<br/>Tab.5 Accuracy result without/with absolute calibration

表5 绝对校准前后二分类准确率结果
Tab.5 Accuracy result without/with absolute calibration

[1] 余国良,董妍.我国心理健康研究的现状、热点与发展趋势[J].教育研究,2012,33(6):97-102.
[2] 周昌乐.博学切问[M].厦门:厦门大学出版社,2015:209-212.
[3] TANG Y Y,HOLZEL K,POSNER M.The neuroscience of mindfulness meditation[J].Nature Reviews Neuroscience,2015,16(4):213-225.
[4] ZHOU R L,LIU L.Eight-week mindfulness training enhances left frontal EEG asymmetry during emotional challenge:a randomized controlled trial[J].Mindfulness,2017,1(8):181-189.
[5] 赵建强,周昌乐.短期禅修效果的脑电图多尺度排列熵分析[J].厦门大学学报(自然科学版),2016,55(3):420-425.
[6] EVANS R J.Handbook of neurofeedback[M].Binghamton:The Haworth Medical Press,2007:3-10.
[7] 吴金华,张艳秋,唐毅.数据挖掘在生物信息学中的应用文献计量学视角[J].生物信息学,2016,14(4):249-253.
[8] LIU Y S,SOURINA O,HOU X Y.Neurofeedback games to improve cognitive abilities[C]∥2014 International Conference on Cyberworlds.Santander:IEEE,2014:161-168.
[9] BRANDMEYER T,DELORME A.Meditation and neurofeedback[J].Frontiers in Psychology,2013,4(1):688670.
[10] SAS C,CHOPRA R.Medit aid:a wearable adaptive neurofeedback-based system for training mindfulness state[J].Pers Ubiquit Comput,2015,7(19):1169-1182.
[11] HASHEMI A,PINO J L,MOFFAT G,et al.Characterizing population EEG dynamics throughout adulthood[J].eNeuro,2016,3(6):1-13
[12] 周昌乐.从当代脑科学看禅定状态达成的可能性及其意义[J].杭州师范大学学报(社会科学版),2015(3):17-23.
[13] Emotiv Inc.Emotiv epoc & testbenchTM specifications[EB/OL].[2017-10-07].https:∥www.emotiv.com/files/Emotiv-EPOC-Product-Sheet-2014.pdf.
[14] FELL J,AXMACHER N,HAUPT S.From alpha to gamma:electrophysiological correlates of meditation-related states of consciousness[J].Medical Hypotheses,2010,75(2):218-224.
[15] 汪云九.神经信息学神经系统的理论和模型[M].北京:高等教育出版社,2006:403-407.
[16] 佩特根,于尔根斯,绍柏.混沌与分形[M].田逢喜,译.北京:国防工业出版社,2008:300-305.
[17] WEISSTEIN W E.Minkowski-bouligand dimension[EB/OL].[2017-12-18].http:∥mathworld.wolfram.com/Minkowski-BouligandDimension.html.
[18] SHANNON C.A mathematical theory of communication[J].Bell Systems Technical Journal,1948,3(27):379-423.
[19] TAKENS F.Detecting strange attractors in turbulence[C]∥Dynamical Systems and Turbulence.Heidelberg:Springer,1981:366-381.
[20] 李颖洁,邱意弘,朱贻盛.脑电信号分析方法及其应用[M].北京:科学出版社,2009:1-3.
[21] FRANK E,HALL A M,WITTEN H I.The WEKA workbench[M].San Francisco:Morgan Kaufmann,2016:1-8.
[22] FRANK E,WITTEN H I.Generating accurate rule sets without global optimization[C]∥ICML'98 Proceedings of the Fifteenth International Conference on Machine Learning.San Francisco:Morgan Kaufmann,1998:144-151.
[23] 周志华.机器学习[M].北京:清华大学出版社,2016:30-35.
[24] CHIESA A,CALATI R,SERRETTI A.Does mindfulness training improve cognitive abilities? A systematic review of neuropsychological findings[J].Clinical Psychology Review,2011,31(3):449-464.

备注

引言

1 实验方法

2 数据处理

3 数据挖掘及结果

4 讨论

5 结论

学报简介

备注

引言