كليدواژه :
آنفلوانزاي پرندگان , تحليل مكاني , رگرسيون تقويت شده , رگرسيون وزندار جغرافيايي
چكيده فارسي :
اﺑﺘﻼء ﭘﺮﻧﺪﮔﺎن ﺑﻪ ﺑﯿﻤﺎري آﻧﻔﻠﻮاﻧﺰاي ﻓﻮق ﺣﺎد ﭘﺮﻧﺪﮔﺎن )Highly Pathogenic Avian Influenza( و از ﺑﯿﻦ رﻓﺘﻦ آنﻫﺎ ﺧﺴﺎرات ﺳﻨﮕﯿﻨﯽ ﺑﻪ ﺻﻨﻌﺖ دام و ﻃﯿﻮر و ﺑﻬﺪاﺷﺖ ﻋﻤﻮﻣﯽ ﮐﺸﻮر ﺗﺤﻤﯿﻞ ﻣﯽﮐﻨﺪ. اﻣﺮوزه ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ ﺣﺠﻢ و ﺗﻨﻮع دادهﻫﺎ، ﺿﺮورت اﺳﺘﻔﺎده از ﻓﻨﺎوريﻫﺎي ﻣﮑﺎن ﻣﺤﻮر و ﻋﻠﻮم دادهﮐﺎوي ﺿﺮوري ﺑﻪ ﻧﻈﺮ ﻣﯽرﺳﺪ. ﻫﺪف اﯾﻦ ﺗﺤﻘﯿﻖ ﻣﺪلﺳﺎزي ﺷﯿﻮع ﺑﯿﻤﺎري آﻧﻔﻠﻮاﻧﺰا ﻓﻮق ﺣﺎد ﭘﺮﻧﺪﮔﺎن ﺑﻪ ﮐﻤﮏ ﻗﺎﺑﻠﯿﺖﻫﺎي ﺗﺤﻠﯿﻞ ﻣﮑﺎﻧﯽ ﻣﯽﺑﺎﺷﺪ.
ﻣﻮاد و روشﻫﺎ: در ﭘﮋوﻫﺶ ﺣﺎﺿﺮ ﮐﻪ ﺑﻪ ﺻﻮرت ﺗﺤﻠﯿﻠﯽ-اﮐﻮﻟﻮژﯾﮑﯽ اﺳﺖ، ﺳﺎل 1395 ﺑﺎ ﻣﯿﺰان ﺑﺎﻻي ﺷﯿﻮع اﯾﻦ ﺑﯿﻤﺎري ﺑﻪ ﻋﻨﻮان ﺳﺎل ﺗﻬﯿﻪ ﻣﺘﻐﯿﺮﻫﺎي 17ﮔﺎﻧﻪ )اﻗﻠﯿﻤﯽ، ﻣﺤﯿﻄﯽ و اﻧﺴﺎن ﺳﺎﺧﺖ( و اﯾﺠﺎد ﻻﯾﻪﻫﺎي ﻣﮑﺎﻧﯽ، در اﺳﺘﺎن ﮔﯿﻼن اﻧﺘﺨﺎب ﮔﺮدﯾﺪ. ﺑﺎ اﺳﺘﻔﺎده از ﺗﻠﻔﯿﻖ ﺗﺤﻠﯿﻞ رﮔﺮﺳﯿﻮﻧﯽ ﺗﻘﻮﯾﺖ ﺷﺪه )Boosted Regression Trees; BRT( و رﮔﺮﺳﯿﻮن وزندار ﺟﻐﺮاﻓﯿﺎﯾﯽ، وزنﻫﺎي اﯾﻦ ﻣﺘﻐﯿﺮﻫﺎ ﻣﺤﺎﺳﺒﻪ و ﻣﺪل ﺷﯿﻮع ﺑﯿﻤﺎري ﺗﻬﯿﻪ و ﺗﻮﺳﻂ ﻣﻨﺤﻨﯽ ﻋﻤﻠﯿﺎﺗﯽ درﯾﺎﻓﺖ ﮐﻨﻨﺪه ) ;Receiver Operating Characteristic ROC( اﻋﺘﺒﺎر آن ﻣﻮرد ارزﯾﺎﺑﯽ ﻗﺮار ﮔﺮﻓﺖ.
ﯾﺎﻓﺘﻪﻫﺎ: ﻣﺘﻐﯿﺮﻫﺎي ﺗﺎﻻب، ﺑﺎزار ﻓﺮوش ﻣﺮغ زﻧﺪه و اﺳﺘﺨﺮﻫﺎ ﺑﯿﺶﺗﺮﯾﻦ وزن را در ﺗﺤﻠﯿﻞ BRT ﺑﻪ ﺗﺮﺗﯿﺐ 12/8 ،15/59 ،18/91 ﺑﻪ ﺧﻮد اﺧﺘﺼﺎص دادﻧﺪ. ﻫﻢﭼﻨﯿﻦ، از ﻧﻈﺮ زﻣﺎﻧﯽ ﻣﺎه ﺑﻬﻤﻦ ﺑﯿﺶﺗﺮﯾﻦ ﻣﯿﺰان ﺷﯿﻮع را در ﺑﯿﻦ 3 ﻣﺎه ﺳﺮد ﺳﺎل داﺷﺘﻪ اﺳﺖ. ﻧﺘﯿﺠﻪﮔﯿﺮي: اﯾﻦ ﺑﯿﻤﺎري در ﻧﻮاﺣﯽ اﻃﺮاف ﺗﺎﻻبﻫﺎ و اﺳﺘﺨﺮﻫﺎ، ﻧﺰدﯾﮑﯽ ﺑﺎزارﻫﺎي ﻓﺮوش ﻣﺮغ زﻧﺪه ﻣﺸﺎﻫﺪه ﺷﺪه اﺳﺖ. ﺑﻨﺎﺑﺮاﯾﻦ اداره ﮐﻞ دامﭘﺰﺷﮑﯽ ﺑﻪﻋﻨﻮان ﻧﻬﺎد ﻧﻈﺎرﺗﯽ و ﺳﯿﺎﺳﺖﮔﺬار و ﺗﻮﻟﯿﺪﮐﻨﻨﺪﮔﺎن و ﻓﺮوﺷﻨﺪﮔﺎن ﻣﺮغ ﺑﻪﻋﻨﻮان ﻋﻮاﻣﻞ اﺟﺮاﯾﯽ ﻣﯽﺗﻮاﻧﻨﺪ ﻧﻘﺶ ﺑﺴﯿﺎر ﻣﻬﻤﯽ در ﭘﺎﯾﺶ، ﮐﻨﺘﺮل و ﺟﻠﻮﮔﯿﺮي از ﺷﯿﻮع اﯾﻦ ﺑﯿﻤﺎري اﯾﻔﺎء ﻧﻤﺎﯾﻨﺪ.
چكيده لاتين :
Infection of birds to Highly Pathogenic Avian Influenza (HPAI) and their extinction
impose heavily losses on the livestock and poultry industry along with public health. Nowadays, due to the volume
and variety of data, the need of using location-based technologies and data mining sciences has become inevitable.
This study aims to model the prevalence of avian influenza, using the capabilities of spatial analyses.
Materials and Methods: In this analytical-ecological study, the year 2016 is selected as the target year to prepare 17
variables (climate, environment, and man-made) and their spatial layers in Guilan province because of the high
prevalence of the disease in this year. The weights of the variables were computed through combination of Boosted
Regression Trees (BRT) analysis and Geographically Weighted Regression (GWR), and then prevalence of the disease
was prepared and evaluated by the Receiver Operating Characteristic (ROC) curve.
Results: The variables of wetlands, live poultry markets, and pools have the highest weights according to BRT analysis,
with 18.91, 15.59, and 12.8 percent, respectively. Also, in terms of time, the month of February has the highest
prevalence among the three cold months of the year.
Conclusion: The disease has been observed in the areas around wetlands, pools, and live poultry markets. Therefore,
the General Veterinary Administration, as a regulatory and policy-making body, and poultry producers and sellers as
executive agents can play a significant role in monitoring, controlling, and preventing the spread of the disease.
