Some aspects of designing perceptron network to predict bladder-cancer patient-survival

• Autorzy: Kolasa M. , Wojtyna R. , Jóźwicki W.

Warianty tytułu

PL

Pewne aspekty projektowania sieci perceptronowej do przewidywania długości życia chorych na raka pęcherza moczowego

• Konferencja: Signal Processing Algorithms, Architectures, Arrangements, and Applications. 11th IEEE Signal Processing Workshop SPA 2007 ; 7.09.2007 ; Poznan, Poland
• Języki publikacji: EN

Abstrakty

EN

A problem of establishing an optimal number of neurons in a hidden layer of a perceptron network used to predict survival time of patients with bladder cancer has been discussed. Our considerations are important in postoperative treatments of this illness. The applied neural network is a three layer one with one hidden layer. Its designing and testing were performed in MATLAB environment. As the network teaching method, classical error back-propagation algorithm with a momentum factor was applied. For the assumed model of the problem, we have obtained a characteristic graph of the function describing false results of the survival predictions. We have utilized a representative training set and investigated the network for different number of neurons in the hidden layer. A distinct error minimum has been observed for 13 neurons in this layer. It is not out of the question that the character of the achieved curve is repeatable for different input/ output vectors and may be practicable for determining the number of neurons in networks dedicated to biological models.

PL

W pracy podjęto próbę wskazania metody doboru optymalnej liczby neuronów dla warstwy ukrytej sieci neuronowej, analizującej dane modelu przeżycia pooperacyjnego u pacjentów z rakiem pęcherza moczowego. Trójwarstwową sieć zaprojektowano w środowisku Matlab, z zastosowaniem modelu perceptronu wielowarstwowego. Jako metodę uczenia sieci zastosowano klasyczny algorytm uczenia metodą wstecznej propagacji błędu ze współczynnikiem momentum. Dla założonego modelu przewidywania przeżycia u chorych z rakiem pęcherza moczowego uzyskano charakterystyczny przebieg krzywej błędnych prognoz. W oparciu o stworzony zbiór uczący zbadano działanie sieci dla różnej liczby neuronów w warstwie ukrytej. Zaobserwowano wyraźne minimum błędu dla 13 neuronów w tej warstwie. Nie można wykluczyć, że przebieg krzywej ma charakter powtarzalny dla różnych wektorów wejścia/wyjścia i może być pomocny w typowaniu liczby neuronów w sieciach dedykowanych modelowi biologicznemu.

Słowa kluczowe

EN

artificial neural network; network hidden layer; bladder cancer; prognosis; survival analysis

PL

sztuczna sieć neuronowa; warstwa ukryta sieci; rak pęcherza moczowego; prognozowanie; analiza przeżycia

• Wydawca: Wydawnictwo SIGMA-NOT
• Czasopismo: Elektronika : konstrukcje, technologie, zastosowania
• Rocznik: 2008
• Tom: Vol. 49, nr 4
• Strony: 80--84
• Opis fizyczny: Bibliogr. 23 poz., rys., tab., wykr.

Twórcy

autor

Kolasa M.

autor

Wojtyna R.

autor

Jóźwicki W.

• University of Technology and Life Sciences, Faculty of Telecommunication & Electrical Engineering, Bydgoszcz

Bibliografia

• [1] Abbod M.F., Catto J.W.F., Linkens D.A., Wild P.J., Herr A., Wissmann C., Pilarsky C., Hartmann A.F.C.: Artificial Intelligence Technique for Gene Expression Profiling of Urinary Bladder Cancer. In: 3rd International IEEE Conference Intelligent Systems 2006, pp. 646-651.
• [2] Snow P.B., Rodvold D.M., Brandt J.M.: Artificial Neural Networks in Clinical Urology. Urology 1999, vol. 54 pp. 787-790.
• [3] Snow P.B., Smith D.S., Catalona W.J.: Artificial neural networks in the diagnosis and prognosis of prostate cancer: a pilot study. J Urol 1994, vol. 152 pp. 1923-1926.
• [4] Michaels E.K., Niederberger C.S., Golden R.M., et al.: Use of a Neural Network to Predict Stone Growth After Shock Wave Lithotripsy. Urology 1998, vol.51 pp. 335-338.
• [5] Catto J.W.F., Linkens D.A., Abbod M.F., Chen M., Burton J.L., Feeley K.M., Hamdy F.C.: Artificial Intelligence in Predicting Bladder Cancer Outcome: A Comparison of Neuro-Fuzzy Modeling and Artificial Neural Networks, Clinical Cancer Research 2003, vol. 9 pp. 4172-4177.
• [6] Naguib R.N.G., Qureshi K.N., Hamdy F.C., Neal D.E.: Neural Network analysis of Prognostic Markers in Bladder Cancer. In: 19th International Conference IEEE/EMBS Chicago 1997, vol.3, pp. 646-651.
• [7] Tkacz E.J., Kostka P., Jonderko K., Mika B.: Supervised and Unsupervised Learning Systems as a Part of Hybrid Structures Applied in EGG Signals Classifiers. IEEE Annual Conference Engineering in Medicine and Biology, Shanghai, China 2005.
• [8] Żurada J.: Introduction to Artificial Neural Networks, West Publishing Company, 1992.
• [9] Cauwenberghs G., Bayoumi M.: Learning on silicon, adaptive VLSI neural systems, Kluwer Academic Publishers, 1999.
• [10] Maass W., Bishop C.: Pulsed neural networks, Massachusetts Institute of Technology, The MIT Press, 1999. [11] Kohonen T. Self-organizing maps, 3-ed, Springer Verlag, Berlin Heidelberg, 2001.
• [12] Długosz R., Talaśka T, Wojtyna R.: New Binary-Tree-Based Winner-Takes-All Circuit for Learning on Silicon Kohonen's Networks, International Conference On Signals And Electronic Systems (ICSES), Łódź, Poland, 2006.
• [13] Talaśka T., Długosz R., Pedrycz W.: Adaptive Weight Change Mechanism for Kohonens's Neural Network Implemented in CMOS 0.18 um Technology, 11th European Symposium on Artificial Neural Networks, Bruges, Belgium, 2007.
• [14] Talaśka T., Długosz R., Wojtyna R.: Current mode analog Kohonen neural network, International conference Mixed Design of Integrated Circuits and Systems, Ciechocinek, Poland, 2007
• [15] Wei J.T., Zhang Z., Barnhill S.D., Madyastha K.R., Zhang H., Oesterling J.E.: Understanding artificial neural networks and exploring their potential applications for the practicing urologist Urology, vol. 52, no. 2, 1998.
• [16] Tewari A., Narayan P.: Nowel staging tool for localized prostate cancer: A pilot study using genetic adaptive neural networks Journal of Urology, vol. 160, no. 2, 1998.
• [17] Magnotta V.A., Heckel D., Andreasen N.C., Cizadlo T., Corson P.W., Ehrhardt J.C., Yu W.T.: Measurement of brain structures with artificial neural networks: Two- and three-dimensional i applications, Radiology, vol. 211, no. 3, 1999.
• [18] Tadeusiewicz R., Ogiela M. Medical image understanding technology: Artificial intelligence and soft-computing for image i understanding, Springer Verlag, Berlin Heidelberg New York, VIII 2004 (Studies in fuzziness & soft computing, vol. 156).
• [19] Lopez-Beltran A., Sauter G., Gasser T., et al.: Infiltrating urothelial carcinoma. In: Eble JN, et al, editors. WHO classification of tumours. Pathology and genetics. Tumours of the urinary system and male genital organs, Lyon, France7 IARC Press; M pp. 97-104.
• [20] Shariat S.R., Karakiewicz P.A., Palapattu G.S., Amiel G.E., Lotan Y., Rogers C.G., Vazina A., Bastian P.J., Gupta A., Salagowsky A.I., Schoenberg M., Lerner S.P.: Nomograms Provide Improved Accuracy for Predicting Survival after Radical Cystectomy, Clin Cancer Res, 12(22), November 15, 2006 pp. 6663-76.
• [21] Karakiewicz P.I., Shariat S.F., Palapattu G.S., Gilad A.E., Lotan Y., Rogers C.G., Vazina A., Gupta A., Bastian P.J., Perrotte P., Sagalowsky A.I., Schoenberg M., Lerner S.P.: Nomogram for predicting disease recurrence after radical cystectomy for Transitional Cell Carcinoma of the Bladder, J. Urol, October 2006, vol. 176, pp. 1354-1362.
• [22] Habuchi T., Marberger M., Droller M.J. et al.: Prognostic markers for bladder cancer: International Consensus Panel on Bladder Tumor Markers, Urology 2005, vol. 66 pp. 64-74.
• [23] Khashman A., Dimililer K.: Neural Network Arbitration for Optimum DCT Image Compression, Eurocon 2007, September 201 Warsaw.