癲癇是大腦神經元突發性異常放電，導致短暫的大腦功能障礙的一種慢性疾病。由於異常放電神經元所涉及的部位不同，發作的情形也有很大的差異。關於癲癇診斷，醫生會透過分析EEG資料以及觀察病人行為模式，找出腦部異常放電點和其擴散的模式。因此，癲癇發作的分析對於診斷治療是一個首要的步驟。這篇論文提出一個使用影像與聲音資料分析的結果，再加以合併進行判斷的癲癇發作偵測演算法。通常在癲癇發作時，病人會無法控制自己的行為，像是無意識的發出尖叫、呼喊，或是全身痙攣。根據這些在聲音與動作上的特徵，我們採用了Hidden Markov Model (HMM)來分類這些不同的聲音，並且利用image difference與corner point的數量變化來分析影像上的變化，找出影像上因癲癇發作導致劇烈變化的地方。最後再使用Dempster-Shafer Theory將影像與聲音兩方面的分析結果合併，根據兩方的結果共同判斷此時是否為癲癇發作。在本篇論文最後，針對了一些癲癇病患的EEG錄影影片作偵測，也得到了期望中的結果。

A seizure is an abnormal movement or behavior caused by unusual electrical activity in the brain, and patients with epilepsy may experience various types of seizures. Seizure analysis plays one of the major roles in epilepsy diagnosis and treatment. Therefore seizure detection is one prerequisite step for epilepsy treatment. In this thesis, the fusion of audio and video is presented for seizure detection. Normally when seizure occurs, the patients would lose their control and yell or scream unconsciously, accompanied with a great movement like jerking. Based on these characters, audio features are extracted and applied into with Hidden Markov Model (HMM) for classifying the screaming and other kinds of sounds. Then through a Dempster-Shafer Theory they are fused with the video features, which depict the strong movement, for detecting the video-audio segments where seizure occurs. Results have been tested by data obtained from several seizure patients and showed promising results.

[1] Taiwan Epilepsy Association, http://www.epilepsyorg.org.tw/
[2] 施茂雄, “癲癇發作型態的認識與診斷,” TEA Epilepsy News, vol. 1, no. 4, Nov. 1993.
[3] Rabiner, L.R., “A tutorial on hidden Markov models and selected applications in speech recognition,” Proceedings of the IEEE, vol. 77, issue 2, pages 257-286, Feb. 1989.
[4] Climent Nadeu, Dusan Macho and Javier Hernando, “Time and frequency filtering of filter-bank energies for robust HMM speech recognition,” Speech Communication, vol. 34, pages 93-114, 2001.
[5] Tong Zhang and C.-C. Jay Kuo, “Audio Content Analysis for Online Audiovisual Data Segmentation and Classification,” IEEE Transactions on Speech and Audio Processing, vol. 9, no. 4, pages 441-457, May 2001.
[6] Tong Zhang and C.-C. Jay Kuo, “Classification and Retrieval of Sound Effects in Audiovisual Data Management,” IEEE Transactions on Speech and Audio Processing, vol. 10, issue 5, pages 293-302, July 2002.
[7] Robin R. Murphy, “Dempster–Shafer Theory for Sensor Fusion in Autonomous Mobile Robots,” IEEE Transaction on Robotics and Automation, vol. 14, no. 2, pages 197-205, April 1998.
[8] Huadong Wu1, Mel Siegel, Rainer Stiefelhagen, JieYang, “Sensor Fusion Using Dempster-Shafer Theory,” IEEE Instrumentation and Measurement Technology Conference, Anchorage, AK, USA, 21-23 May 2002.
[9] Keane T.M., Taylor K.L., Penk W.E.and Bauer M, “Approximation Algorithms and Decision Making in the Dempster-Shafer Theory of Evidence--An Empirical Study,” International Journal of Approximate Reasoning, vol. 17, no. 2, pages 217-237, August 1997.
[10] Ivan Osorio, Mark G. Frei, and Steven B. Wilkinson, “Real-Time Automated Detection and Quantitative Analysis of Seizures and Short-Term Prediction of Clinical Onset,” Epilepsia, vol. 39, page 615, June 1998.
[11] Gabor A.J., Leach R.R. and Dowla F.U., “Automated seizure detection using a self-organizing neural network,” Electroenceph. clin. Neurophysiol., vol. 99, pages 257-266, Sep. 1996.
[12] Murro AM, King DW, Smith JR, Gallagher BB, Flanigin HF and Meador K, “Computerized seizure detection of complex partial seizures,” Electroenceph. clin. Neurophysiol., vol. 79, pages 330-333, Oct. 1999.
[13] Jerger, Kristin K., Netoff, Theoden I., Francis, Joseph T., Sauer, Timothy, Pecora, Louis, Weinstein, Steven L., and Schiff, Steven J, “Early Seizure Detection,” Journal of Clinical Neurophysiology, vol. 18, pages 259-268, May 2001.
[14] Hyoung-Gook Kim, Nicolas Moreau, and Thomas Sikora, “Audio Classification Based on MPEG-7 Spectral Basis Representations,” IEEE Transaction on Circuits And Systems for Video Technology, vol. 14, no. 5, May 2004.
[15] P.H.S. Torr, “A Structure and Motion Toolkit in Matlab - Interactive Advantures in S and M,” Microsoft Research Technical Report, June 2002.
[16] C. Harris and M. Stephens, “A combined corner and edge detector,” In Proc. Alvey Conf., pages 189-192, 1987.
[17] Goguen, J. A, “The logic of inexact concepts,” Syntheses vol. 19, pages 325-373, 1969.
[18] Zadeh, L. A, “Similarity relations and fuzzy orderings,” Information Science, vol. 3, pages 177-206, 1971.
[19] Karl Sentz and Scott Ferson, “Combination of Evidence in Demster-Shafer Theory,” SANDIA REPORT SAND2002-0835, Unlimited Release, April 2002.