Model Builder在類流感的疫情地理資訊系統上之應用

The Application of Model Builder in Influenza-Like Illness with GIS

當禽流感爆發傳播整個世界的同時，為了日益增多人類的健康觀點來看，一個及時動態且具保護作用的系統，對疾病的預防監控上則具有其一定的重要性。台灣疾病管制中心－疾病管制局，其藉由定點通報醫師的監控系統，利用定點通報的方式，透過網際網路來回報許多傳染病的疑似門診病患數，並以此進行監控各傳染病的暴發情形。疾病管制局能透過使用這個監視系統，來監控類流感 (Influenza-like illness, ILI)的傳染動向，用以防止嚴重的流感大流行的流行性感冒和事件發生情況。這篇文章整合了空間統計方法及Model Builder模型建立技術，並以GIS方式說明類流感的時序性及空間性，視其動向分析，架構成為一能展示並預測監控疾病的重要性及趨勢性。空間內插法－克力金(Kriging)，其效用為被用來探索ILI暴發高峰的空間趨勢，作為一種疾病監視的方法。配合時空數學預測模型，可以藉由發展半自動化的技術，來改進使用的效率，和那些具獨特分析且皆具有適合空間關係的模型，是具有相同的效用。這次的發展內容，主要是介紹一種疾病監視及監控的工具，是種具新穎且兼具實用的分析工具。

In view of the growing human health concern as the avian flu outbreaks spread the entire world, a timely active protection system will be crucial to disease prevention. The sentinel surveillance system of Taiwan CDC, through sentinel physicians, reports the number of outpatients suspected of infection by numerous infectious diseases from Internet. CDC can prevent the pandemic flu and occurrence of severe influenza by using this surveillance system to monitor the move of influenza-like (ILI) illness. This paper illustrates the integration of GIS with spatial statistical methods and Model Builder technology onto the analysis of temporal and spatial moves of influenza-like illness, enables one to predict both the severity and trend of disease. Spatial interpolation - the Kriging method - was used to explore the utility of the spatial trend of ILI peaks, as a method of disease surveillance. The model improves efficiency through utilization of the semi-automation techniques developed, as well as by the model's unique analysis for spatial correlation. This development introduces a new and very useful analytical tool for disease surveillance and monitoring.