理解傳染病的空間擴散：地理計算的方法

對抗傳染性疾病有兩大面向：防止傳染病擴散以及準確評估各地區對於醫療資源可及性的高低。前者是避免疾病的地理影響範圍繼續擴大；後者則是力求保護疫區居民的健康、減少生命財產損失。而要有效達成這兩個面向的防疫工作，關鍵前提就是要對於傳染病擴散的時空特徵有足夠瞭解。從理解到防止再到保護，這三個面向的整合涉及了防疫過程中的大部分重要決策。然而，文獻上針對這三個面向的研究皆存在方法學上的不足。首先，過往已有文獻透過量性描述群聚的擴散型態來理解傳染病的擴散特徵，但是僅止於學理性的討論，缺乏系統性的分類來定義各種可能的群聚擴散型態；同時，也缺乏相對應的方式來自動剖析各種型態出現的時間和位置。第二，移動管制分區是最為嚴厲的防止疾病擴散的手段，但在面臨疾病威脅的極端狀況時（例如新冠肺炎），也是普遍被多數國家採用。實務上的移動管制分區大多以現行的行政區層級做為管制單位，忽略了人口移動行為的特徵，導致容易劃定地理範圍過大或過小的管制分區，致使防止擴散的效益不如預期。最後，在傳染病疫情中，醫療資源需求會隨著疫情嚴重程度起伏而變動，導致各地區的可及性出現時序性的變化。然而，目前並未有文獻考量到如此的變動特性，所以也缺乏合適的模型進行動態評估。這些方法學上的不足可能會產生有偏誤的分析結果或是無法揭露重要的疫情訊息，進而影響到防疫決策的判斷。因此，本論文旨在針對這三個面向的分析方法分別進行改良。研究成果包含了能自動判斷各種群聚演化擴散型態的MST-DBSCAN演算法、考量人口移動規律性進行區域劃分的HuMoRZ演算法、以及Epi-RA模式利用傳染病擴散模擬將資源需求的動態變化整合至空間可及性評估模型當中。本論文以實際案例應用的方式來證實這三個方法的實用性，相關成果也已透過國際學術期刊文章或專書專章的形式進行發表，進一步證實本論文的價值和貢獻。

關鍵字

傳染病擴散；時空過程； MST-DBSCAN ； HuMoRZ ； Epi-RA

並列摘要

Confronting infectious diseases contains two major aspects: preventing epidemic diffusion and evaluating spatial accessibility to medical resources. The former is to control the influenced area; the latter is to protect residents’ health and lives. Moreover, understanding space-time process of epidemic diffusion is critical for these two aspects. The three aspects are related to almost all decision makings during an epidemic; however, some methodological shortages exist. First, to understand epidemic diffusion, a systematical classification of evolution types of disease clusters and a method to automatically profile location and time of every type is still lack. Second, to prevent epidemic diffusion, containment zone is the most severe approach used in an extreme situation. A containment zone is usually formed by administrative districts, yet it is not effective due to the neglect of human movement properties. Finally, demand for medical resource may vary over time based on severity of an epidemic, which result in spatiotemporally dynamic accessibility, yet such dynamic is still neglected in the literature These methodological shortages may generate inappropriate results and thus negatively affect decision makings. Therefore, this thesis developed the following three new methods: (1) MST-DBSCAN, which can automatically profile various cluster evolution types, (2) HuMoRZ, which considers the regularity of human mobility to delineate containment zones, and (3) Epi-RA, which integrates evaluation of spatial accessibility with simulation of epidemic diffusion. These methods’ feasibilities are demonstrated through real case studies, and they have also been published in international academic journal or book, which further proves their contributions.

並列關鍵字

Epidemic diffusion ； Space-time process ； MST-DBSCAN ； HuMoRZ ； Epi-RA

參考文獻

Anselin, L., J. Cohen, D. Cook, W. Gorr, and G. Tita. 2000. Spatial Analyses of Crime. Criminal Justice 4 (2):213-262.

Google Scholar

Arango, C. 2020. Lessons learned from the coronavirus health crisis in Madrid, Spain: How COVID-19 has changed our lives in the last 2 weeks. Biological Psychiatry 88 (7):e33-e34.

Google Scholar

Auzan, A. A. 2020. The economy under the pandemic and afterwards. Population and Economics 4 (2):4-12.

Google Scholar

Bailey, T. C., and A. C. Gatrell. 1995. Interactive spatial data analysis. Essex: Longman Scientific Technical.

Google Scholar

Bajardi, P., C. Poletto, J. J. Ramasco, M. Tizzoni, V. Colizza, and A. Vespignani. 2011. Human mobility networks, travel restrictions, and the global spread of 2009 H1N1 pandemic. PloS one 6 (1):e16591.

Google Scholar

國際替代計量

理解傳染病的空間擴散：地理計算的方法

全文下載

主題瀏覽