作者: W.B. Roush , R.F. Wideman
DOI: 10.1093/PS/79.2.180
关键词:
摘要: Abstract An evaluation was made of the relationship between individual daily growth patterns and susceptibility broiler chickens to pulmonary hypertension syndrome (PHS). In first experiment, 46 male broilers were weighed for each 50 d, during which time 13 developed PHS. Three temporal phases (0 15, 16 35, 36 d) velocity acceleration examined. Correlation dimensions Lyapunov exponents suggested evidence chaos in acceleration, but absence detectable differences normal PHS categories led us reject hypotheses that is more chaotic than susceptible Growth values mean SD statistically evaluated as response variables phase. Mean third phase different (velocity: 68.8 vs 48.9 g/d, P = 0.03, respectively) (acceleration: 0.3 -1.4 g/d2, 0.07, respectively). The (reflecting oscillation acceleration) greater birds 26.1 21.3 0.13, respectively; acceleration: 39.7 28.2 hypothesis accepted have oscillations A general regression neural network (GRNN) with genetic adaptive calibration trained predict based on their combinations. Data representing first, two, all three determined potential computerized diagnostic weighing. With GRNN, data sets successfully classified (100%) or without hypothesis, therefore, artificial networks could be used distinguish difference those second only one bird diagnosed Velocity from Phase 1 Phases 2 experiment applied Experiment 2. most promising they correctly identified 71.6 72.4% data, respectively. general, exceeded range training both reflected increased growth.