Use of GIS and machine learning to predict disease in shrimp farmed on the east coast of the Mekong Delta, Vietnam
15/02/23 03:17PM
Khiem Nguyen Minh; Takahashi, Yuki; Yasuma Hiroki; Oanh Dang Thi Hoang; Hai Tran Ngoc; et al. Fisheries Science; Tokyo Vol. 88, Iss. 1, (Jan 2022): 1-13. DOI:10.1007/s12562-021-01577-8
Abstract:
Diseases in shrimp farms in the Mekong Delta of Vietnam cause significant crop losses and are therefore of great concern to producers. Once a pond becomes infected, it is difficult to prevent spread of the disease to nearby shrimp farming areas. Thus, predicting the occurrence of disease is an essential part of reducing the risk for shrimp farmers. In this study, we applied an integrated geographic information system and machine learning system to predict three serious diseases of shrimp, namely, acute hepatopancreatic necrosis, white spot syndrome disease, and Enterocytozoon hepatopenaei infection, based on data collected from shrimp farms in the Tra Vinh, Bac Lieu, Soc Trang, and Ca Mau provinces of Vietnam. We first constructed a map showing the distribution of these diseases using the locations of affected farms, and then we conducted spatial analysis to acquire the geographical features of the affected locations. This latter information was combined with environmental factors and clinical signs to form the set of independent variables affecting the outbreak of diseases. The neural network model outperformed the logistic regression, random forest, and gradient boosting methods in terms of predicting infection to estimate the probability of disease occurrence in farmed areas. Acute hepatopancreatic necrosis disease infected farms downstream of the Co Chien and Hau Rivers of Tra Vinh and west of Ca Mau. Enterocytozoon hepatopenaei infection is distributed in Soc Trang Province, while white spot syndrome virus has spread to the coastal districts of Soc Trang and Bac Lieu Provinces, where it is highly associated to water from a complex canal system.
Fulltext: https://doi.org/10.1007/s12562-021-01577-8
(Source:https://www.proquest.com/scholarly-journals/use-gis-machine-learning-predict-disease-shrimp/docview/2624600813/se-2?accountid=28030)