基于可穿戴設(shè)備的日常壓力狀態(tài)評估研究
doi: 10.11999/JEIT170120
-
2.
(中國科學(xué)院電子學(xué)研究所 北京 100190) ②(中國科學(xué)院大學(xué) 北京 100049) ③(中國科學(xué)院心理研究所 北京 100101)
基金項(xiàng)目:
國家自然科學(xué)基金(61302033)，北京市自然科學(xué)基金(Z160003)，國家重點(diǎn)研發(fā)計(jì)劃(2016YFC1304302)
Research on Daily Stress Detection Based on Wearable Device
-
2.
(Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China)
Funds:
The National Natural Science Foundation of China (61302033), The Key Project of Beijing Municipal Natural Science Foundation (Z160003), The National Key Research and Development Project (2016YFC1304302, 2016YFC0206502, 2016YFC1303900)
-
摘要: 現(xiàn)代生活普遍壓力較大，容易引起消極痛苦的應(yīng)激，導(dǎo)致不良情緒甚至滋生各類慢性病。心理專家需要了解個體的壓力狀態(tài)，從而開展對應(yīng)性心理疏導(dǎo)和治療。傳統(tǒng)心理學(xué)自評法存在一定的主觀性；基于生理多導(dǎo)儀的壓力狀態(tài)評估法，受設(shè)備體積所限無法用于日常壓力狀態(tài)評估。針對上述問題，該文采用可穿戴式傳感設(shè)備實(shí)時采集個體生理信號，利用心理和生理的伴生關(guān)系，對個體的心理壓力進(jìn)行長期實(shí)時評估。同時通過蒙特利爾影像應(yīng)激實(shí)驗(yàn)(MIST)誘發(fā)出被試平靜、輕微及高度壓力3種壓力狀態(tài)，此實(shí)驗(yàn)范式同時包含認(rèn)知負(fù)荷精神壓力因素與社會評價(jià)心理壓力因素，與日常真實(shí)生活更為接近。該文共采集39名健康被試的實(shí)驗(yàn)數(shù)據(jù)，通過對數(shù)據(jù)的特征值提取等預(yù)處理，結(jié)合隨機(jī)森林算法對最優(yōu)特征子集進(jìn)行選擇，采用支持向量機(jī)(SVM)分類算法對3種壓力狀態(tài)進(jìn)行分類預(yù)測。實(shí)驗(yàn)結(jié)果表明，通過隨機(jī)森林特征選擇優(yōu)化后的SVM分類，與通用的單一SVM分類算法相比，具有更好的分類識別效果，對3種壓力狀態(tài)的分類準(zhǔn)確率可從78%提高至84%。
-
關(guān)鍵詞:
- 壓力識別 /
- 可穿戴設(shè)備 /
- 心電 /
- 呼吸 /
- 支持向量機(jī)
Abstract: In modern life, high stress causes negative emotions and even leads to various chronic diseases. Psychologists need to understand the stress state of the individual in order to facilitate the corresponding psychological treatment. The traditional method of self-evaluation in psychology contains some subjectivity, while the method based on physiological polygraph can not be used in daily stress assessment because of the volume of equipment. For these reasons, a wearable device is used to collect the physiological signals and an assessment of the individuals stress is made according to the associated relationship between the psychological and physiological. The Montreal Imaging Stress Task (MIST) is used to induce three states of no, moderate and high stress. The MIST includs both mental and psychosocial stress factors, which is more closing to a real-life condition. The experimental data are collected from 39 healthy subjects. Features are extracted from the data and the random forest is used to select the optimal stress-related feature combination, which is used to train and test the Support Vector Machine (SVM) classifier. Finally, the results show that the combination of random forest feature selection and SVM achieves a better performance. The accuracy is improved from 78% to 84% in the three states detection.-
Key words:
- Stress recognition /
- Wearable device /
- Electrocardiogram /
- Respiration /
- Support Vector Machine (SVM)
-
CACIOPPO J T and TASSINARY L G. Principles of Psychophysiology: Physical, Social, and Inferential Elements [M]. New York, NY, US, Cambridge University Press, 1990: 10-12. KIRSCHBAUM C, PRUSSNER J C, STONE A A, et al. Persistent high cortisol responses to repeated psychological stress in a subpopulation of healthy men[J]. Psychosomatic Medicine, 1995, 57(5): 468-474. MCEWEN B S and STELLAR E. Stress and the individual: Mechanisms leading to disease[J]. Archives of Internal Medicine, 1993, 153(18): 2093-2101. doi: 10.1001/archinte. 1993.00410180039004. LUTCHYN Y, JOHNS P, CZERWINSKI M, et al. Stress is in the eye of the beholder[C]. IEEE International Conference on Affective Computing and Intelligent Interaction (ACII), Xian, China, 2015: 119-124. doi: 10.1109/ACII.2015. 7344560. HAAG A, GORONZY S, SCHAICH P, et al. Emotion recognition using bio-sensors: First steps towards an automatic system[C]. Tutorial and Research Workshop on Affective Dialogue Systems, Kloster Irsee, Germany, 2004: 36-48. doi: 10.1007/978-3-540-24842-2_4. SANO A, PHILLIPS A J, AMY Z Y, et al. Recognizing academic performance, sleep quality, stress level, and mental health using personality traits, wearable sensors and mobile phones[C]. IEEE 12th International Conference on Wearable and Implantable Body Sensor Networks (BSN), Cambridge, MA, USA, 2015: 1-6. doi: 10.1109/BSN.2015.7299420. CHOI J and GUTIERREZ-OSUNA R. Using heart rate monitors to detect mental stress[C]. IEEE Sixth International Workshop on Wearable and Implantable Body Sensor Networks, Berkeley, CA, USA, 2009: 219-223. doi: 10.1109/ BSN.2009.13. WIJSMAN J, GRUNDLEHNER B, LIU H, et al. Towards mental stress detection using wearable physiological sensors [C]. IEEE 33th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, MA, USA, 2011: 1798-1801. doi: 10.1109/IEMBS.2011.6090512. MCDUFF D, GONTAREK S, and PICARD R. Remote measurement of cognitive stress via heart rate variability[C]. IEEE 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Chicago, IL, USA, 2014: 2957-2960. doi: 10.1109/EMBC.2014.6944243. MCDUFF D J, HERNANDEZ J, GONTAREK S, et al. Cogcam: Contact-free measurement of cognitive stress during computer tasks with a digital camera[C]. Proceedings of the 2016 ACM CHI Conference on Human Factors in Computing Systems, San Jose, CA, USA, 2016: 4000-4004. doi: 10.1145/ 2858036.2858247. DEDOVIC K, RENWICK R, MAHANI N K, et al. The montreal imaging stress task: Using functional imaging to investigate the effects of perceiving and processing psychosocial stress in the human brain[J]. Journal of Psychiatry Neuroscience, 2005, 30(5): 319-325. PAN J and TOMPKINS W J. A real-time QRS detection algorithm[J]. IEEE Transactions on Biomedical Engineering, 1985, 32(3): 230-236. doi: 10.1109/TBME.1985.325532. BERNTSON G G, QUIGLEY K S, JANG J F, et al. An approach to artifact identification: Application to heart period data[J]. Psychophysiology, 1990, 27(5): 586-598. doi: 10.1111/j.1469-8986.1990.tb01982.x. LU W, NYSTROM M M, PARIKH P J, et al. A semi- automatic method for peak and valley detection in free-breathing respiratory waveforms[J]. Medical Physics, 2006, 33(10): 3634-3636. doi: 10.1118/1.2348764. STEPHENS C L, CHRISTIE I C, and FRIEDMAN B H. Autonomic specificity of basic emotions: Evidence from pattern classification and cluster analysis[J]. Biological Psychology, 2010, 84(3): 463-473. doi: 10.1016/j.biopsycho. 2010.03.014. 劉袁緣, 陳靚影, 俞侃, 等. 基于樹結(jié)構(gòu)分層隨機(jī)森林在非約束環(huán)境下的頭部姿態(tài)估計(jì)[J]. 電子與信息學(xué)報(bào), 2015, 37(3): 543-551. doi: 10.11999/JEIT140433. LIU Yuanyuan, CHEN Jingying, YU Kan, et al. Head pose estimation based on tree-structure cascaded random forests in unconstrained environment[J]. Journal of Electronics Information Technology, 2015, 37(3): 543-551. doi: 10.11999/ JEIT140433. HOVSEPIAN K, AL,ABSI M, ERTIN E, et al. CStress: Towards a gold standard for continuous stress assessment in the mobile environment[C]. Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, Osaka, Japan, 2015: 493-504. doi: 10.1145/ 2750858.2807526. 高發(fā)榮, 王佳佳, 席旭剛, 等. 基于粒子群優(yōu)化-支持向量機(jī)方法的下肢肌電信號步態(tài)識別[J]. 電子與信息學(xué)報(bào), 2015, 37(5): 1154-1159. doi: 10.11999/JEIT141083. GAO Farong, WANG Jiajia, XI Xugang, et al. Gait recognition for lower extremity electromyographic signals based on PSO-SVM method[J]. Journal of Electronics Information Technology, 2015, 37(5): 1154-1159. doi: 10.11999/JEIT141083. 陳素根, 吳小俊. 基于特征值分解的中心支持向量機(jī)算法[J]. 電子與信息學(xué)報(bào), 2016, 38(3): 557-564. doi: 10.11999/ JEIT150693. CHEN Sugen and WU Xiaojun. Eigenvalue proximal support vector machine algorithm based on eigenvalue decoposition[J]. Journal of Electronics Information Technology, 2016, 38(3): 557-564. doi: 10.11999/JEIT150693. SETZ C, ARNRICH B, SCHUMM J, et al. Discriminating stress from cognitive load using a wearable EDA device[J]. IEEE Transactions on Information Technology in Biomedicine, 2010, 14(2): 410-417. doi: 10.1109/TITB.2009. 2036164. -
計(jì)量
- 文章訪問數(shù): 1515
- HTML全文瀏覽量: 203
- PDF下載量: 303
- 被引次數(shù): 0
下載:
