本研究以診斷資料爲探勘的資料來源、及以某一病患爲探勘的目標,利用資料探勘的關聯規則分別從以下兩方面偵測病患的疾病診斷是否異常:一是設計一個快速探勘關聯規則的方法,並且關聯規則的前置項目組必須包含於此病患症狀中,根據關聯規則所顯示出的傾向特徵,可判斷此病患是否具有疾病診斷異常的傾向;二是設計一個快速探勘關聯規則的方法,並且關聯規則的前置項目組必須包含於此病患的診斷疾病中,根據關聯規則所顯示出的傾向特徵,可判斷此病患是否具有症狀問診異常的傾向。依據文中所提出之方法,我們設計與建置一個偵測疾病異常診斷的探勘系統。此探勘結果,對臨床經驗不足之醫療人員可以對其避免診斷的疏忽,可以提供非常有用的參考資訊。
This paper uses diagnostic data as the source of mining. We let a patient to be as the target of mining, and use association rules of data mining to detect careless diagnosis of the patient's diseases from two aspects: one is to propose a fast method to mine association rules whose antecedents are contained in the patient's symptoms, and we detect whether the diseases diagnosed is carelessness or not according to the characteristics of the association rules; the other one is to propose a fast method to mine association rules whose antecedents are contained in the patient's diseases diagnosed, and we detect whether the symptoms inquired is carelessness or not according to the characteristics of the association rules. A mining system is designed and constructed to detect careless diagnoses of diseases based on the both methods. The results of detecting can provide very useful information to avoid careless diagnoses of diseases for inexperience hospital staffs.
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。