Background: The International Association for the Study of Pain (IASP), describes the pain as an unpleasant sensory and emotional experience associated with actual or potential tissue damage. Pain is a system of defense mechanism that prompts the body to avoid the source of noxious stimulation and a potential area of tissue damage. Electrocardiogram (ECG) is the electrical activity of the heart. ECG gives the transthoracic interpretation of the electrical activity of the heart over a period of time. Analysis of ECG signal provides information regarding the condition of heart at a given moment. This thesis is aimed to challenge the ability of ANI to detect noxious stimulation during a surgical operation.
Methods: ECG signals were collected from 30 patients aged between 20 to 80 years old that undergone a general anesthesia surgery at the National Taiwan University Hospital (NTUH). RR series was calculated from the ECG and then re-sampled at 8 Hz using a linear interpolation. RRi samples were isolated into a 64 sec moving window and normalized for patient comparability. ANI is calculated from the enveloped area between the local minima and local maxima of the RR series. Artificial neural network (ANN) models including recurrent neural network (RNN) and nonlinear autoregressive neural network (NARX) were applied to predict pain using ANI and Doctors assessment of pain during the surgery.
Results: The results show ANI mean±SD of 40.77±9.99 before the surgery began, 44.62±10.69 after the surgery, 39.52±8.36 during drug administration, 30.39±7.50 during intubation, 24.26±7.04 during electric knife incision and 34.47±8.62 during stitching. The mean and standard deviation for the period of the surgery is 154±62.64 min. Predictive results for training ANN, RNN and NARX gave a mean absolute error of 3.9564, 3.163 and 3.808 respectively.
Conclusion: This paper presents the application of ANI index to evaluate pain during a surgical operation. The clinical study presented in this thesis indicates that a more painful situation in the surgical operation is associated with a decreased ANI index, confirming that ANI monitoring as a tool capable of measuring a change in the level of pain during surgery. Predicting pain using artificial neural network models was not successful and we aim at applying deep learning in the future.

Background: The International Association for the Study of Pain (IASP), describes the pain as an unpleasant sensory and emotional experience associated with actual or potential tissue damage. Pain is a system of defense mechanism that prompts the body to avoid the source of noxious stimulation and a potential area of tissue damage. Electrocardiogram (ECG) is the electrical activity of the heart. ECG gives the transthoracic interpretation of the electrical activity of the heart over a period of time. Analysis of ECG signal provides information regarding the condition of heart at a given moment. This thesis is aimed to challenge the ability of ANI to detect noxious stimulation during a surgical operation.
Methods: ECG signals were collected from 30 patients aged between 20 to 80 years old that undergone a general anesthesia surgery at the National Taiwan University Hospital (NTUH). RR series was calculated from the ECG and then re-sampled at 8 Hz using a linear interpolation. RRi samples were isolated into a 64 sec moving window and normalized for patient comparability. ANI is calculated from the enveloped area between the local minima and local maxima of the RR series. Artificial neural network (ANN) models including recurrent neural network (RNN) and nonlinear autoregressive neural network (NARX) were applied to predict pain using ANI and Doctors assessment of pain during the surgery.
Results: The results show ANI mean±SD of 40.77±9.99 before the surgery began, 44.62±10.69 after the surgery, 39.52±8.36 during drug administration, 30.39±7.50 during intubation, 24.26±7.04 during electric knife incision and 34.47±8.62 during stitching. The mean and standard deviation for the period of the surgery is 154±62.64 min. Predictive results for training ANN, RNN and NARX gave a mean absolute error of 3.9564, 3.163 and 3.808 respectively.
Conclusion: This paper presents the application of ANI index to evaluate pain during a surgical operation. The clinical study presented in this thesis indicates that a more painful situation in the surgical operation is associated with a decreased ANI index, confirming that ANI monitoring as a tool capable of measuring a change in the level of pain during surgery. Predicting pain using artificial neural network models was not successful and we aim at applying deep learning in the future.

A Thesis i
Abstract ii
Acknowledgements iv
Contents v
List of tables vii
List of Figures viii
Abbreviation ix
Chapter 1: Introduction 1
1.1 Background 1
1.2 Motivation 3
1.3 Purpose 5
1.4 Brief Summary of Chapters 5
Chapter 2: General Anesthesia 6
2.1 General Anesthesia 6
2.2 Anesthesia Process 6
Chapter 3: Analysis Algorithm 8
3.1 ANI computation 8
3.2 RR samples, windowing, normalization and filtering 8
3.3 Area under the curve (AUC) parameter computation 9
3.4 Artificial neural network (ANN) 10
3.5 Recurrent neural network (RNN) 11
3.6 Nonlinear autoregressive neural network (NARX) 12
Chapter 4: Experiment Condition and Method 14
4.1 Data source 14
4.2 Expert Assessment of Pain Level 17
Chapter 5: Results 19
5.1 States of analgesia / nociception during general anaesthesia 19
5.2 ANI Analysis 19
5.3 Prediction results 21
Chapter 6: Conclusion and future work 27
References 28

1. R. D. Miller, editor. Miller's Anesthesia. Elsevier, 6th edition, 2005.

2. S. P. Desai, M. S. Desai, R. Maddi, and G. E. Battit. A tale of two paintings: Depictions of the first public demonstration of ether anesthesia. Anesthesiology, 2007.

3. F. D. Moore. John Collins warren and his act of conscience - a brief narrative of the trial and triumph of a great surgeon. Annals of Surgery, 1999.

4. J. Enderle, S. Blanchard, and J. Bronzino, editors. Introduction to Biomedical Engineering. Elsevier Academic Press, 2nd edition, 2005.

5. H. Merskey and N. Bogduk, editors. Part III: Pain Terms, a Current List with Definitions and Notes on Usage. Classification of Chronic Pain. IASP Task Force on Taxonomy. IASP Press, 2nd edition, 1994.

6. W. F. Gannong, editor. Review of Medical Physiology. Lange Medical Books/McGraw-Hill, 22nd edition, 2005.

7. R. A. Rhoades and D. R. Bell, editors. Medical Physiology Principles for Clinical Medicine. Lippincott Williams and Wilkins, 3rd edition, 2009.

8. A. Guyton and J. E. Hall, editors. Textbook of Medical Physiology. Elsevier Saunders, 11th edition, 2006.

9. L. Brunton, K. Parker, D. Blumenthal, I. Buxton. “Goodman and Gilman's Manual of Pharmacology and Therapeutics.” McGraw Hill Medical, 12th edition, 2008.

10. INFARMED. Folheto informativo. Remifentanil ultiva, 2006.

11. T. W. Latson, D. O'flaherty. "Effects of surgical stimulation on autonomic reflex function: assessment by changes in heart rate variability." British Journal of Anaesthesia BJA, vol. 70, no. 3, pp. 301-305, 1993.

12. V. Pichot, J. M. Gaspoz, S. Molliex, A. Antoniadis, T. Busso, F. Roche, F. Costes, L. Quintin, J. R. Lacour, J. C. Barthélémy. "Wavelet transform to quantify heart rate variability and to assess its instantaneous changes." Journal of applied physiology, vol. 86, no. 3, 1081-1091, 1999.

13. M. Jeanne, R. Logier, J. D. Jonckheere, B. Tavernier. “Heart rate variability during total intravenous anesthesia: effects of nociception and analgesia.” Autonomic Neuroscience, vol. 147, no. 1, pp. 91-96, 2009.

14. A. J. Camm, M. Malik, J. T. Bigger, G. Breithardt, S. Cerutti, R. J. Cohen, P. Coumel et al. "Heart rate variability: standards of measurement, physiological interpretation and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology." Circulation 93, no. 5, pp. 1043-1065, 1916.

15. M. L. Guen, M. Jeanne, K. Sievert, M. A. Moubarik, T. Chazot, P. A. Laloe, J. F. Dreyfus, M. Fischler. “The analgesia nociception index: a pilot study to evaluation of a new pain parameter during labor,” International Journal of Obstetric Anesthesia, vol. 21, no. 2, pp. 146-151, 2012.

16. M. Jeanne, C. Clement, J. D. Jonckheere, R. Logier, B. Tavernier. “Variations of the analgesia nociception index during general anaesthesia for laparoscopic abdominal surgery.” Journal of Clinical Monitoring Computing, vol. 26, no. 4, pp. 289-294, 2012.

17. R. Logier, J. Dejonckheere, A. Dassonneville. “An efficient algorithm for R–R interval series filtering.” Engineering in Medicine and Biology Society, Proceedings of the 26th annual international conference of the IEEE, vol. 2, pp. 3937-3940, 2004.

18. E. D. Ubeyli, “Recurrent neural networks employing Lyapunov exponents for analysis of ECG signals. Expert Systems with Applications,” vol. 37, pp. 1192–1199, 2010.

19. E. D. Ubeyli, “Combining recurrent neural networks with eigenvector methods for classification of ECG beats,” Digital Signal Processing, vol. 19, pp. 320–329, 2009.

20. R. Salloum, C. C. J. Kuo, “ECG-based biometrics using recurrent neural networks,” Acoustics, Speech and Signal Processing (ICASSP), IEEE International conference on, pp. 2062-2066, 2017.

21. M. Cheng, W. J. Sori, F. Jiang, A. Khan, S. Liu. “Recurrent Neural Network Based Classification of ECG Signal Features for Obstruction of Sleep Apnea Detection.” 2017 IEEE International Conference on Computational Science and Engineering (CSE) and IEEE International Conference on Embedded and Ubiquitous Computing (EUC), vol. 2, pp. 199-202.

22. A. Graves, A. R. Mohamed, G. Hinton. "Speech recognition with deep recurrent neural networks." In Acoustics, speech and signal processing (ICASSP), 2013 IEEE international conference, pp. 6645-6649, 2013.

23. K. Cho, B. V. Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, & Y. Bengio. “Learning phrase representations using RNN encoder-decoder for statistical machine translation.” arXiv preprint arXiv: 1406.1078, 2014.

24. A. M. Sullivan, H. Xia, J. C. McBride, X Zhao, “Reconstruction of Missing Physiological Signals Using Artificial Neural Networks,” Computing in Cardiology, IEEE, pp. 317−320, 2010.

25. J. P. M. Menezes, G. A. Barreto, “Long-term time series prediction with the NARX network: An empirical evaluation,” Neurocomputing, vol. 71, no. 16, pp. 3335–3343, 2008.

26. T. Ledowski, W. S. Tiong, C. Lee, B. Wong, T. Fiori, N. Parker. “Analgesia nociception index: evaluation as a new parameter for acute postoperative pain,” British Journal of Anaesthesia, vol. 111, pp. 627-629, 2013.

27. R. Logier, M. Jeanne, J. De jonckheere, A. Dassonneville, M. Delecroix, B. Tavernier – “PhysioDoloris: a monitoring device for Analgesia / Nociception balance evaluation using heart rate variability analysis.” Engineering Medicine Biology Society, Annual International Conference of IEEE, pp. 1194-1197, 2011.

28. M. Jeanne, R. Logier, J. De Jonckheere, B. Tavernier. “Validation of a graphic measurement of heart rate variability to assess analgesia/nociception balance during general anesthesia.” Annual International Conference of the IEEE, Engineering Medicine and Biological Society, pp. 1840-1843, 2009.

29. M. Delecroix, M. Jeanne, A. Keribedj, N. Couturier, and R. Logier. "Automated analgesic drugs delivery guided by vagal tone evaluation: Interest of the Analgesia Nociception Index (ANI)." In Engineering in Medicine and Biology Society (EMBC), 2013 35th Annual International Conference of the IEEE, pp. 1952-1955, 2013.

30. G. N. Fatma, E. D. Übeyli, I. Güler. "Recurrent neural networks employing Lyapunov exponents for EEG signals classification." Expert systems with applications, vol. 29, no. 3, pp. 506-514, 2005.

31. C. Z. Lipton, D. C. Kale, C. Elkan, R. Wetzel. “Learning to diagnose with LSTM recurrent neural networks.” Published as a conference paper at ICLR 2016.

32. W. Xie, A. Noble, A. Zisserman, “Layer recurrent neural networks,” ICLR 2017.

33. D. Howard, M. Beale, M. Hagan. "Neural network toolbox™ 6." User’s guide, pp. 37-55, 2008.

34. S. Chen, S. Billings, & P. Grant, “Non-linear system identification using neural networks,” International Journal of Control, vol. 51, no. 6, pp. 1191– 1214, 1990.

35. M. Jeanne, R. Logier, J. De Jonckheere, B. Tavernier. “Validation of a graphic measurement of heart rate variability to assess analgesia/ nociception balance during general anesthesia.” Engineering in Medicine and Biology Society. EMBC. Annual International Conference of IEEE, pp. 1840–1843, 2009.