Predicting chronic wasting disease in white-tailed deer at the county scale using machine learning.

Autor: Ahmed MS; Wildlife Health Lab, Cornell University, Ithaca, NY, USA. sohelcu06@gmail.com.; Texas A & M Transportation Institute, Austin, TX, USA. sohelcu06@gmail.com., Hanley BJ; Wildlife Health Lab, Cornell University, Ithaca, NY, USA., Mitchell CI; Desert Centered Ecology, LLC, Tucson, AZ, USA.; U.S. Fish and Wildlife Service, Tucson, AZ, USA., Abbott RC; Wildlife Health Lab, Cornell University, Ithaca, NY, USA., Hollingshead NA; Wildlife Health Lab, Cornell University, Ithaca, NY, USA., Booth JG; Department of Statistics and Data Science, Cornell University, Ithaca, NY, USA., Guinness J; Department of Statistics and Data Science, Cornell University, Ithaca, NY, USA., Jennelle CS; Minnesota Department of Natural Resources, Nongame Wildlife Program, Saint Paul, MN, USA., Hodel FH; Department of Fisheries and Wildlife, Michigan State University, East Lansing, MI, USA., Gonzalez-Crespo C; Center for Animal Disease Modelling and Surveillance, University of California, Davis, CA, USA., Middaugh CR; Arkansas Game and Fish Commission, Little Rock, AR, USA., Ballard JR; Arkansas Game and Fish Commission, Little Rock, AR, USA., Clemons B; Florida Fish and Wildlife Conservation Commission, Gainesville, FL, USA., Killmaster CH; Georgia Department of Natural Resources, Social Circle, GA, USA., Harms TM; Iowa Department of Natural Resources, Ames, IA, USA., Caudell JN; Indiana Department of Natural Resources, Bloomington, IN, USA., Benavidez Westrich KM; Indiana Department of Natural Resources, Bloomington, IN, USA., McCallen E; Indiana Department of Natural Resources, Bloomington, IN, USA., Casey C; Kentucky Department of Fish and Wildlife Resources, Frankfort, KY, USA., O'Brien LM; Maryland Department of Natural Resources, Annapolis, MD, USA., Trudeau JK; Maryland Department of Natural Resources, Annapolis, MD, USA., Stewart C; Michigan Department of Natural Resources, Grand Rapids, MI, USA., Carstensen M; Minnesota Department of Natural Resources, Wildlife Health Program, Forest Lake, MN, USA., McKinley WT; Mississippi Department of Wildlife, Fisheries, and Parks, Jackson, MS, USA., Hynes KP; New York State Department of Environmental Conservation, Delmar, NY, USA., Stevens AE; New York State Department of Environmental Conservation, Delmar, NY, USA., Miller LA; New York State Department of Environmental Conservation, Delmar, NY, USA., Cook M; North Carolina Wildlife Resources Commission, Raleigh, NC, USA., Myers RT; North Carolina Wildlife Resources Commission, Raleigh, NC, USA., Shaw J; North Carolina Wildlife Resources Commission, Raleigh, NC, USA., Tonkovich MJ; Ohio Department of Natural Resources, Athens, OH, USA., Kelly JD; Florida Fish and Wildlife Conservation Commission, Gainesville, FL, USA., Grove DM; University of Tennessee, Nashville, TN, USA., Storm DJ; Wisconsin Department of Natural Resources, Madison, WI, USA., Schuler KL; Wildlife Health Lab, Cornell University, Ithaca, NY, USA.
Jazyk: angličtina
Zdroj: Scientific reports [Sci Rep] 2024 Jun 22; Vol. 14 (1), pp. 14373. Date of Electronic Publication: 2024 Jun 22.
DOI: 10.1038/s41598-024-65002-7
Abstrakt: Continued spread of chronic wasting disease (CWD) through wild cervid herds negatively impacts populations, erodes wildlife conservation, drains resource dollars, and challenges wildlife management agencies. Risk factors for CWD have been investigated at state scales, but a regional model to predict locations of new infections can guide increasingly efficient surveillance efforts. We predicted CWD incidence by county using CWD surveillance data depicting white-tailed deer (Odocoileus virginianus) in 16 eastern and midwestern US states. We predicted the binary outcome of CWD-status using four machine learning models, utilized five-fold cross-validation and grid search to pinpoint the best model, then compared model predictions against the subsequent year of surveillance data. Cross validation revealed that the Light Boosting Gradient model was the most reliable predictor given the regional data. The predictive model could be helpful for surveillance planning. Predictions of false positives emphasize areas that warrant targeted CWD surveillance because of similar conditions with counties known to harbor CWD. However, disagreements in positives and negatives between the CWD Prediction Web App predictions and the on-the-ground surveillance data one year later underscore the need for state wildlife agency professionals to use a layered modeling approach to ensure robust surveillance planning. The CWD Prediction Web App is at https://cwd-predict.streamlit.app/ .
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje