以倒傳遞網路預測環孢靈在腎臟移植後患者之用量適當性

Using Back-Propagation Network to Predict Proper Cyclosporine Dosage in Patients After Kidney Transplantation

張俊雄(Tsun-Hsiung Chang)

指導教授 ： 黃有評

大同大學/電機資訊學院/資訊工程學系所/碩士(2008年)

摘要

人工智慧技術已經廣泛的應用在各種領域研究上，也常常應用於醫療決策輔助行為上。本研究主要是應用倒傳遞類神經網路，從腎臟移植病患的歷史病例中，建構出可用以預測免疫抑制劑環孢靈在腎臟移植病患上的作用結果之分類模式。實驗結果顯示，利用倒傳遞類神經網路可以提昇正確辨識率達到86.81%，比醫師以個人經驗判別的正確率66.29%高出許多。所以本研究可以有效協助醫療人員掌握環孢靈之血中濃度，提升環孢靈的用藥安全及用藥品質，且提高腎臟移植病患之存活率。

關鍵字

人工智慧 ； 倒傳遞類神經網路 ； 腎臟移植 ； 環孢靈

並列摘要

Artificial intelligence technology has been extensively used in various applications. It is also used as an auxiliary tool for medical policy decision making. The application of back-propagation network in this research builds the assortment model from the history of kidney transplant patients who took the cyclosporine. About 66.29% of patients can be correctly identified by doctor’s personal experience to differentiate the results from using cyclosporine, while 86.81% of correctness is achieved by the application of back-propagation neural network strategy. We hope the results could help the medical personnel master the effectiveness of cyclosporine and improve the drug safety, the quality of using medicine, and the survival rate of kidney transplant patients.

並列關鍵字

Artificial intelligence ； back-propagation neural network ； kidney transplant ； cyclosporine

參考文獻

[17] 胡秀蓉，使用Cyclosporine不同的監測方法（C2與C0）應用於孩童肝臟移植的研究，高雄醫學大學藥學研究所碩士論文，2003年。

[21] 林文綾，腎臟移植後病人生活品質及其相關因素之探討，國防醫學院護理研究所碩士論文，2000年。

[23] Durham University website, http://www.dur.ac.uk .

[24] M.J.A Berry and G.S. Gordon, Mastering Data Mining, Jon Wiley & Sons, INC., Canada, 2000.

[26] G. Camps-Valls, B. Porta-Oltra, E. Soria-Olivas, J.D. Martín-Guerrero, A.J. Serrano-López, J.J. Pérez-Ruixo and N.V. Jiménez-Torres, “Prediction of cyclosporine dosage in patients after kidney transplantation using neural Networks,” IEEE Trans. on Biomedical Engineering, vol. 50, no. 4, pp.442-448, April 2003.