在眾多的老年常見的疾病中,關節炎佔有相當高的罹患比率,而膝關節的退化多少都會影響老人的生活。行動不便不但會造成了生理上的障礙,對於老人心理的衝擊影響甚鉅。根據健保申報資料統計顯示,台灣地區接受膝關節置換術從1998年到2003年約成長了29.9%,其利用率高且醫療費用成本高,佔健保相當份量的支出比例。 本研究的目的乃是要建構一個輔助臨床的評估系統。就醫院觀點而言,初期住院日評估可以協助醫師事先建立診療標準流程,在有限的醫療資源與人力情況下如何合理分配與評估並可作為後續出院照顧計畫之安排;就病患與家屬觀點而言,可減少病患對自己住院期間長短不確定性與減輕病患家屬精神壓力的心理調適。 本研究分別應用三種人工智慧技術(決策樹、類神經網路與案例式推理)以作為探討醫療資源分配之工具。研究結果顯示,決策樹正確分類率為78.53%;類神經網路之均方根誤差為0.0436;案例式推理在誤差為一天時,準確率可達79.21%。總結而言,本研究結果已顯示此系統的實際可行性,並提供適切結果。
Arthritis is the most common disease among the old people in Taiwan. Arthrosis of the knee caused tremendous inconvenience to the elders’ life. In addition, the disability not only impact old people physically but also mentally. According to the statistics released by Bureau of National Health Insurance, people who undertook the surgery of total knee arthroplasty have grown 29.9% from 1998 to 2003. It is a huge portion of the medical budget. In this thesis, the computer aided clinical evaluation system is expected to help the doctors to make a better decision. A standard procedure can be built up as doctors’ reference. Furthermore, it can work as reference for planning of aftercare under the limited medical resource and human resource. Finally, the uncertainty and anxiety of the patient and their family can be reduced. In this work, we propose the following three different methods of artificial intelligence technology for assistance in medical resource planning. It contained Decision Tree, Neural Network and Case-Based Reasoning. Results of this study showed that the classification of Decision Tree is as large as 78.53%, the RMSE of Neural Network is at 0.0436, and the accuracy with absolute tolerance at one day of Case-Based Reasoning reached 79.21%. To conclude, this study have demonstrated that these new systems can be practically implemented and provide adequate results.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。