Spatial pattern analysis and association of influenza morbidity with climatic factors using fuzzy AHP in Chiang Mai, Thailand | |
| Author | Supachai Nakapan |
| Call Number | AIT Diss. no.RS-13-01 |
| Subject(s) | Influenza--Geographic Information Systems--Thailand--Chiang Mai Climatic changes--Geographic Information Systems--Thailand--Chiang Mai |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Technical Science in Remote Sensing and Geographic Information Systems |
| Publisher | Asian Institute of Technology |
| Series Statement | Dissertation ; no. RS-13-01 |
| Abstract | The geographical and distribution of many seasonal infectious diseases are linked to the climate, therefore theusing of seasonal climate factors as predictive indicators to forecast is the most important. Apart from that, thecaused by the world’s climate change was convincing evidence influence to the anthropogenic and provided the addition of incentive to more knowledge of climate and disease interaction. The factors of climate variability, minimum andmaximum temperature, humidity and rainfall are major causes of seasonal disease that has impact onsocio-economic and public health. Analysis of climate factors was conductingusingboth meteorological data and disease case data from population at village and sub-districtleveland thesewere collected fromChiang Mai Provincial Public Health Office (CMPHO), Chiang Mai Province, Thailand. The relationship between climate and disease pattern are well established with many occurrenceduring seasonal or erupting from unseasonable, flood or draught condition.An innovativeconcept inGeographic Information Systems(GIS) should aim to visualize and analyzethe epidemiologic distribution through patterns, trends and relationship that would be more important to understand spatial and temporal diffusion of disease outbreak related toclimate factors. For this, Chiang Mai Province was chosen toa study area according toits high diseasesuch as influenza, pneumonia, diarrhea etc.incidences. Climate obviously has a high spatial variability, so it is important that the observed climate factors can be representative of the environmental conditions as experiencedby patients. In addition, Chiang Mai Province provides a high degree of variability of all climate factors, so this variability can provide a wide range of sample conditions for study. Thus, Chiang Mai Province has both a large enough population to provide consistent yearly disease incidences while the observed weather stations are sufficiently representative of environmental conditionscoveringthe area of study. The purpose of main objective of this researchwasfocus to applythe spatial pattern climatology and epidemiology approach for analyzing ofseasonal epidemic, influenzasurveillance in Chiang Mai Province. The specific objectives wereto (1) transform the disease pattern ofinfluenzaincidence rateby usingstatisticaltechniques to indicate disease mapping (2) analyze the relationship between climate factorsand influenza outbreak(3) verify new techniquesto integrate GIS using the fuzzy analysis for influenzarisk zone map.The first objective is to createadisease map and cluster detection, the input data based on epidemic data, climate data and demographic data at village level in year 2001-2009 from CMHPO. Epidemic data contained in the daily patients of influenza cases and climatic factors for 9 years including rainfall (mm.), temperature (oC) and humidity (%).The Empirical Bayes Estimators was use to smoothing the epidemic incidence rate. The method of kernel density interpolation was chosenfor spatial predictionto generate the empirical Bayesian estimates of epidemic incidence rate (EBEEIR) map. The kernels were generated using ArcView GIS spatial analyst extension. Generating all kernels with 100 m. uniform cell size is giving a basis for comparison. The selectingbandwidths for the kernel posed a challengeand various bandwidths were used for this analysis. A 5 km. radius was also used to generate kernel densities which generatedthe disease map. The spatial statistics analysishas beenapplied to create hotspots byuse ofthe “Hotspot Analysis (Getis-Ord*iG)” methodin the extension ofmapping cluster tools toillustratingthe hotspots of disease outbreaks in averaging monthly of year 2001-2009 (January-December). ivThe second objective istofocus on the usingof spatial analytical toolsto reveal and identify significant difference areason the map from the point data surroundings basedon the diseases incidence location. The spatial analysis and statisticsat a provincial scale has beenused to investigate the patterns of spatial diffusion of influenza incidence in Chiang Mai province, Thailand,in terms of itsspatial and geographical distributions. The sub-district (tambon) locationsdata, influenza’s patients, and population were collectedduring 2001-2008 to contributethe objective. Thestudy developed a multivariate model to employ anassessment ofthe relationship between climate factors and influenza outbreaks and validatedthe properof the modelindicatorstoforecasting theinfluenza outbreaks. A multiple regression techniqueshavebeen appliedto employthe statistical model. The GIS techniques and Inverse Distance Weighted (IDW) interpolation were used forspatial diffusion mapping of influenza risk zones. The results from the correlation between influenza outbreaks and climate factorsshow that there is significance for the studyarea. A statistical methodwas validated the model by comparison between results frommodel outputs and actual outbreaks.The last objective is to use Fuzzy analysis apply with GIS to develop a risk zonation map of influenzain Chiang Mai province. The risk zone is obtainedan Analytical Hierarchy Process (AHP) and Fuzzy Analytical Hierarchy Process (FAHP)method. The main understanding of AHP doesnothandle to the association of uncertainty to thejudgments because of that agreewith person’s knowledgeas a crisp number and their Eigen values between 1 and 9. In order to avoid this incompetence, FAHP is determining the betterproposed an alternative to relieve the uncertainness of its method used instead of AHP.Therefore, a number of criteria have been applied such as rainfall, temperature range, relative humidity and population density,etc. The spatial autocorrelation wasidentifiedthe risk areas of influenza incidence location in Chiang Mai during 2001 to 2008. . The results expressed are that high risk zone mapswere mostly considered useful information forclimate factors related to the influenza incidence. The consistency ratio (CR) of AHP and FAHP were found to be acceptable as 0.032 and 0.087, respectively which was less than 0.1. The results can be expressed that the influenza risk zone mapswere mostly considered by criteria which illustrated the risk areas delineated high value of coefficient of determination ( R2= 0.70, 0.76 for AHP and FAHP, respectively) and the relationship between influenza incidence and climate factors was also highly correlate. The results and applicationof the methodology shows that the most properly of identificationof risk area. Furthermore, regarding the expert knowledgecriteria,typesandthe numberto rank all factors made the methodology is highly flexible. Consequently, the understanding of temporal patternsandthe spatial analysisof climate and its affect to thehuman health, particular of the respiratory diseases outbreaks such as influenza,is mostlyimportance to planningthe transmission of the disease outbreak and strategy for the human healthcare system. Also the use of GIS in multi criteria decision support systems,such as fuzzy analysis,can enhance the awareness and warning system for the disease outbreak. It is felt that availability and analysis of daily climatic and disease data may reveal further insight and greater understanding in future for better healthcare and disease control. |
| Year | 2013 |
| Corresponding Series Added Entry | Asian Institute of Technology. Dissertation ; no. RS-13-01 |
| Type | Dissertation |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Remote Sensing (RS) |
| Chairperson(s) | Tripathi, Nitin K. |
| Examination Committee(s) | Souris, Marc ;Taravudh Tipdecho |
| Scholarship Donor(s) | Chiang Mai University, Thailand |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2013 |