在這篇論文中，我們將把擴散網路(diffusion networks)應用在生醫信號辨識上。擴散網路在1993年首次被提出[1]，J.Movellan在2002年發表了用來調變擴散網路參數的蒙地卡羅期望值最大化學習(Monte Carlo EM learning)演算法[2]。J.Movellan認為配合這個演算法後，擴散網路具有學習隨機序列分佈的潛力。但是未有後續關於它應用與性質的研究出現。因此，此篇論文以研究擴散網路的應用與性質為目標，特別是生醫信號辨識上的應用。
要探討擴散網路的應用，則必須對它的性質有更近一步的了解。此篇研究中使用各式人造信號去測試擴散網路在學習時的特性。對於幫助擴散網路學習上，此論文提出了選擇擴散網路增益值的方法。對於觀察擴散網路學習上，此論文提出了用來判斷學習狀況的學習指標。對於驗證擴散網路性質上，此論文證實J.Movellan所說的：「擴散網路具有學習隨機序列分佈的潛力」，但要注意執行蒙地卡羅取樣時信心水準的問題。
這篇論文中所有的模擬結果，無論是執行蒙地卡羅期望值最大化學習演算法或實現生醫信號辨識系統，都是用寫成MATLAB程式碼來完成。
在生醫信號辨識的具體成果上，此篇論文成功的使用擴散網路辨識了心電圖信號和神經突波信號。並且提出了一個系統化的流程，對於其他的生醫信號，也可以利用這個流程來建立自己的辨識系統。

[1] Kuvaeva O F , Levshina I P & Stefanov S B "Effect of noise stress on quantitative characteristics of rat brain synapses" Bulletin of Experimental Biology and Medicine , vol. 97 , no.6 (1984)
[2] Javier R Movellan "A monte Carlo EM Approach for Partially Observable Diffusion Process : Theory and Applications to Neural Network" Neural Computation 14, 1507-1544 (2002)
[3] Javier R Movellan & McClelland J L "Learning continuous probability distributions with symmetric diffusion networks" Cognitive Science 17, 463–496 (1993)
[4] Chen H & Murray A F "A Continuous Restricted Boltzmann Machine with an Implementable Training Algorithm" IEE Proceedings of Vision, Image and Signal Processing, vol.150, no.3, p.153-158 (2003)
[5] Ming-Ru Chu "Design of the Neuron Circuit for the Diffusion Network in VLSI" 清華大學電子所碩士論文 (2006)
[6] Glaser E M & Marks W B "On-line separation of interleaved neuronal pulse sequences Data Acquisition Process" Biol. Med. 5 137–56 (1968)
[7] Glaser E M "Separation of neuronal activity by waveform analysis" Advances in Biomedical Engineering vol 1,ed R M Kenedi (New York: Academic) pp 77–136 (1971)
[8] Gerstein G L, Bloom M J, Espinosa I E, Evanczuk S & Turner M R "Design of a laboratory for multineuron studies" IEEE Trans. Systems, Man Cybern. 13 668–76 (1983)
[9] Pearlmutter B A "Gradient calculations for dynamic recurrent neural networks: A survey" IEEE Transactions on Neural Networks, 6(5), 1212–1228. (1995)
[10] Ackley D H , Hinton G E & Sejnowski T J "A learning algorithm for Boltzmann machines" Cognitive Science, 9, 147–169(1985)
[11] Fishman G S "Monte Carlo sampling: Concepts, algorithms,and applications" New York: Springer-Verlag (1996)
[12] Dempster A P, Laird N M & Rubin D B "Maximum likelihood from incomplete data via the EM algorithm" J. Roy. Stat. Soc., 39, 1–38 (1977)
[13] Lewicki S M "A review of methods for spike sorting: the detection and classification of neural action potentials" Comput. Neural Syst. 9 R53-R78 (1998)
[14] Gerstein G L & Clark W A "Simultaneous studies of firing patterns in several neurons" Science 143 1325-7 (1964)
[15] Keehn D G "An iterative spike separation technique" IEEE Trans. Biomed. Eng. 13 19-28 (1966)
[16] Prochazka V J, Conrad B & Sindermann F "A neuronelectric signal recognition system" Electroenceph. Clin. Neurophysiol. 32 95-7 (1972)
[17] Aapo Hyvärinen & Erkki Oja "Independent Component Analysis:Algorithms and Applications" Neural Networks, 13(4-5):411-430 (2000)
[18] Wheeler B C & Heetderks W J "A comparison of techniques for classification of multiple neural signals" IEEE Trans. Biomed. Eng. 29 752-9 (1982)
[19] Schmidt E M "Computer separation of multi-unit neuroelectric data: a review" J. Neurosci. Meth. 12 95-111 (1984)
[20] Simon W "The real-time sorting of neuron-electric action potentials in multiple unit studies" Electroencepg. Clin. Neurophysiol. 18 192-5 (1965)
[21] Feldman J F & Roberge F A "Computer detection and analysis of neuronal spike sequences" Inform. 9 185-97 (1971)
[22] Dinning G J "Real-time classification of multiunit neural signals using reduced feature sets" IEEE Trans. Biomed. Eng. 28 804-12 (1981)
[23] Zhang K C , Sejnowski T J & McNaughton B L "Automatic separation of spike parameter clusters from tetrode recordings" Soc. Neurosci. Abstr. 23 504 (1997)
[24] Takahashi S , Sakurai Y , Tsukada M and Anzai Y "Classification of neuronal activities from tetrode recodings using independent component analysis" Neurocomputing, vol. 49, pp. 289-298 (2002)
[25] Takahashi S , Anzai Y & Sakurai Y "Automatic sorting for multineuronal activity recorded with tetrodes in the presence of overlapping spikes" J. Neurophysiol., vol. 89, pp. 2245-2258 (2003)
[26] Joon Hwan Choi , Dohoon Kim & Taejeong Kim "A New Overlapping Resolution Method for Multi-channel Spike Sorting" Neural Engineering, 2005. Conference Proceedings. 2nd International IEEE EMBS Conference Page(s):683 – 686 (2005)
[27] Kyung Hwan Kim & Sung June Kim "Method for unsupervised classification of multiunit neural signal recording under low signal-to-noise ratio" IEEE Trans. on Biomedical Engineering Vol. 50 , issue 4 , Pages(s):421 – 431 (2003)
[28] Rieke F , Warland D , R. De Ruyter van Steveninck & Bialek W "Spikes: Exploring the Neural Code" Cambridge MA: MIT Press (1996)
[29] Kyung Hwan Kim & Sung June Kim "A wavelet-based method for action potential detection from extracellular neural signal recording with low signal-to-noise ratio" IEEE Trans. on Biomedical Engineering Vol. 50 , issue 8 , Pages(s):999 – 1011 (2003)
[30] Kadambe S , Murray R & Boudreaux-Bartels G F "Wavelet transform-based QRS complex detector" IEEE Trans. Biomed. Eng. Vol. , 46 , Page(s) 838–847 (1999)
[31] Kohler B U , Hennig C & Orglmeister R "The principles of software QRS detection" IEEE Eng. Med. Biol. Mag, Vol. 21, Page(s). 42–57 (2002)
[32] Huber P J "Projection pursuit" Ann. Statist., Vol. 13, Page(s). 435–475 (1985)
[33] Dongho H, Yadunandana N R, Jose C P & Karl G "Real-Time PCA(Principa1 Component Analysis) implementation on DSP" Neural Networks, 2004. Proceedings. 2004 IEEE International Joint Conference on Volume 3, 25-29 (2004)
[34] Gomis P , Suarez K , Passariello G , Mendoza I , Caminal P & Lander P "Abnormal intra-QRS signals and late potentials in the highresolution ECG associated with chagasic myocarditis" IEEE Computers in Cardiology Volume , Issue , 8-11 , Page(s):633 – 636 (1996)
[35] Singh S "2D spiral recognition using possibilistic measures" Pattern Recognition Letters 19(2):141-147.(1998)