台灣人工生殖技術享譽國際醫學界,隨著現代人生活與工作壓力的增加,不孕夫妻的比率越來越高,求助醫療所之不孕症患者日益增加。進行人工生殖治療的全程步驟複雜,且試管嬰兒的療程費用是相當昂貴的也易引發一些副作用,受療者須忍受身心的考驗,故醫師若在眾多的變項中,能有一套判斷與評估時的參考依據,將有助於提升醫療診斷的正確率及速度,並能有效減輕不孕患者的醫療負擔。 然而,人工生殖成功懷孕之預測研究,通常使用啟發式演算法當研究工具,著重於獲得重要因素和預測效果,但往往不能得知因素交互關係對懷孕結果的影響,於是本研究試圖構造一個約略決策模型,考慮到資訊系統的複雜性,先利用多變量統計方法對指標集進行降維,再借助約略集合方法推導出決策規則並結合流網圖探索解讀之。 本研究使用國內某醫學單位所提供的IVF資料庫,將632筆資料分為兩類:懷孕成功與懷孕失敗,並將57項特徵值降維成19項。研究結果得到17條顯著(涵蓋率為6%以上)的決策規則,整體正確率達80.85%,顯示本研究模式非常適用預測懷孕結果以及提供決策者有用的訊息便於制定醫療方法。
With work pressure increasing, the ratio of infertile couples is also raising, so more and more patients with infertility go to hospital for treatment. The whole procedure of ART is complex and it also easily leads to some side effects, the sufferers have to endure the test of body and mind. So it is important for doctors to understand patient characteristics in order to improve the accuracy of medical diagnosis and speed. However, studies have not yet adequately introduced rules based on patient characteristics and IVF of original data. This study uses rough set theory, a rule-based decision-making technique, to extract rules related to the outcome of pregnancy; then uses a flow network graph, a path-dependent approach, to infer decision rules and variables; and finally presents the relationships between rules and different kinds of result. In this study, we collect 632 patients samples, equally divided into two classes: Success-ful pregnancy and failed pregnancy. The results obtained the number of significant (the rate of covering is over 6% ) decision rules is 17 and the overall accuracy rate is 80.85%. The results show that this combined model can fully predict the outcome of pregnancy and provide useful information for decision-makers in devising medical strategy.