視覺化展現大量及多維度疾病關聯以協助臨床診斷及醫學研究-以攝護腺癌為例

Presentation and visualization of large-scale/multi-dimension disease associations – making sense of diseasome to clinicians and researchers of Prostate Cancer

許權廣(Chun-Kung Hsu)

指導教授 ： 李友專

臺北醫學大學/醫學科技學院/醫學資訊研究所/碩士(2013年)

https://doi.org/10.6831/TMU.2013.00151

摘要

惡性腫瘤已經連續三十年成為我國十大死因之首，藉由早期發現、早期治療，將可大幅提高病患存活率。所以預防醫學十分重要，若有足夠的資料來計算疾病與疾病之間的共病關聯程度，將可大幅提高早期發現惡性腫瘤的機率。醫學研究的資料量非常龐大，本研究透過除錯、篩選、資料加權等步驟，建立一套方法學，依不同目的找出健保資料庫內疾病之間的多種共病關聯，並透過資料純化的處理使其成為臨床診斷及醫學研究上更有意義的資訊。然而，即使經過資料純化，還是有幾萬筆以上的內容需要瀏覽。因此，本研究將以視覺化介面展現大量及多維度疾病關聯圖表，利用動態圖像的方式，使任何人可以迅速發現隱藏在資訊內部的特徵和規律。針對各年齡層病患進行預防醫學的實現，甚至是找出以往所不了解的共病關係，提升臨床醫師的看診質量，也提供醫療研究人員更有意義的參考數據。

關鍵字

惡性腫瘤 ； 共病關聯 ； 健保資料庫 ； 視覺化 ； 預防醫學

並列摘要

Malignancy has become the first of the ten leading causes of death in Taiwan for last 30 years. It will be able to significantly improve patient survival by early detection and early treatment. Therefore, preventive medicine is very important. If there is enough information to calculate the degree of association between the disease and the disease comorbidity, we could significantly improve the chances of early detection of malignancy. The medical research information is very huge, In this study, through debugging, screening, data weighted and other steps to establish a methodology to find out depending on the purpose of the National Health Insurance database disease comorbidity combinations, and purified through information processing makes it more meaningful information on the clinical diagnosis and medical research. However, even after the information was purified, it still needs to browse the contents of more than tens of thousands of pen. Therefore, this study will be a visual interface to show a large and multi-dimensional disease associated chart with the way of the dynamic image. So that anyone can quickly discover the hidden information internal characteristics and laws to realization of preventive medicine for patients of all ages, even identify previously unknown comorbidity relationship enhance clinicians seeing patients quality, also provide reference data for medical researchers more meaningful.

並列關鍵字

Malignancy ； comorbidity combinations ； visualization ； relationship of disease and disease ； preventive medicine

參考文獻

方世榮，《統計學導論》(第五版)，臺北市： 華泰文化事業股份有限公司，2005。

楊志良（民89）。生物統計學新論。（2版）。 臺北市：藝軒。

沈宜靜，林建良，許惠恒。(2011)。糖尿病與癌症之關聯以及台灣現況探討，內科學誌，22，19-30。

Mei-Hua Chuang, Chin-lon Lin, Yuh-Feng Wang, Wen-Long Taso, Yuh-Chang Liang. (2003). Medication Errors in Health Care Institutions. Tzu Chi Med J, 15, 247-258.

Mei-Hua Chuang, Yuh-Jang Liang, Szu-Ching Shen, Chin-Lon Lin.( 2003).Computerized Medication Prescribing Error in Hospital Settings: Tzu Chi J, 15,321-326.