Use of pattern recognition techniques for early detection of morbidity in receiving feedlot cattle.

Two groups of cattle were used to develop (model data set: 384 heifers, 228 ± 22.7 kg BW, monitored over a 225-d feeding period) and to validate (naïve data set: 384 heifers, 322 ± 34.7 kg BW, monitored over a 142-d feeding period) the use of feeding behavior pattern recognition techniques to predict morbidity in newly arrived feedlot cattle. In the model data set, cattle were defined as morbid (MO) if they were removed from their pen to be treated due to visual observation of clinical signs of bovine respiratory disease and healthy (HL) if they remained within their pen and lacked lung lesions at slaughter. Individual feeding behavior parameters collected with a GrowSafe automated feeding behavior monitoring system were reduced via principal component analysis to 5 components that captured 99% of the variability in the data set. Combinations of clustering and cluster classification strategies applied to those components, along with pattern recognition techniques over different time windows, produced a total of 105 models from which precision, negative predictive value, sensitivity, specificity, and accuracy were calculated by comparing its predictions with the actual health status of individual cattle as determined by visual assessment. When the models with the best specificity (models 79 and 87), sensitivity (models 33 and 66), and accuracy (models 3 and 14) in the model data set were used in a naïve data set, models 79 and 87 were not able to predict any MO heifers (0%), with all animals (100%) being predicted as HL. Model 33 predicted 58.3% of the HL and 66.7% of the MO heifers, with MO heifers identified 3.1 ± 1.64 d earlier than by visual observation. Model 66 predicted 50.0% of the HL and 75.0% of the MO heifers, with MO heifers predicted 3.1 ± 1.76 d earlier than by visual observation. Model 3 predicted 100% of the HL and 50.0% of the MO cattle, with MO cattle predicted 1 d earlier than by visual observation. Model 14 predicted 83.3% of the HL and 58.3% of the MO cattle, with MO cattle detected 2.4 ± 1.99 d earlier than visual observation. The application of pattern recognition algorithms to feeding behavior has potential value in identifying MO cattle in advance of overt physical signs of morbidity. Work on an integrated system that would automatically process data collected from automated feed bunk monitoring systems is still required, however, for this method to have value to the commercial feedlot industry as a practical means of identifying MO cattle in real time.

Relationship between feeding behavior and performance of feedlot steers fed barley-based diets.

The relationship between feeding behavior and performance of 274 feedlot cattle was evaluated using Charolais cross steers from 2 consecutive years averaging 293 ± 41 kg for yr 1 (n = 115) and 349 ± 41 for yr 2 (n = 159). Steers were blocked by BW and assigned to 3 (yr 1) or 4 (yr 2) feedlot pens equipped with a radio frequency identification system (GrowSafe Systems). Each pen contained 5 feeding stalls that allowed individual animal access to a feed tub suspended on load cells. The system recorded animal identification, duration, and frequency of feedings as well as the amount of feed consumed during each visit. Daily variation in DMI (DVI), calculated as the absolute difference in DMI from one day to the next, as well as eating rate were determined for each steer. Barley-based diets were delivered to meet steer ad libitum intake over the 213- and 181-d feeding periods for yr 1 and 2 of the study, respectively. The backgrounding periods included the first 85 and 56 d of yr 1 and 2, respectively, in which steers were fed a 14 to 30% concentrate diet, whereas the finishing periods included the last 116 and 101 d of feeding in yr 1 and 2, respectively, with the diet consisting of 77.9% concentrate. Steers were weighed individually every 14 d. To relate feeding behavior to performance, steers were grouped by ADG and G:F and categorized as high, average, or low (based on 1 SD greater than and less than the mean). In the backgrounding and finishing periods of both years of the study, steers classified as having high ADG exhibited greater (P < 0.001) DVI than steers classified as having average or low ADG. Total daily DMI was also greater (P < 0.001) for steers in the high ADG group than those in the low ADG group. Overall, those steers with the greatest G:F also tended (P = 0.15) to have greater DVI than average or low G:F steers. Compared with average or low G:F steers, DMI by high G:F steers in both years of the study was less during backgrounding, finishing, and overall (P = 0.02). Bunk visits and bunk attendance duration were less frequent and shorter (P ≤ 0.01) overall for high compared with low G:F steers. In this study, steers with more variable eating patterns exhibited greater ADG and tended to have greater G:F, a finding that is contrary to industry perception.

Individual intake and antiparasitic efficacy of free choice mineral containing fenbendazole for grazing steers.

A 95-day study (June 25-September 27, 2001) was conducted using 120 steers (311.9+/-2.4 kg) randomly allocated to two treatments: (1) mineral containing 0.55% fenbendazole (FBZ) and (2) control, no FBZ in the mineral. Animals in the FBZ group were individually identified by an electronic tag that was read each time an animal attended the mineral feeder. The feeder was equipped with load cells that enabled individual mineral intakes to be estimated. The FBZ group was provided with non-medicated mineral during a 14-day adaptation period (July 23-August 5) and an 8-day post-medication period (September 17-24). The intake of FBZ was monitored for 14 days during each of the two treatment periods; August 6-19 and September 3-16, separated by a 14-day non-medicated period, August 20-September 2. Control animals had access to non-medicated mineral for the entire 95-day study period. All steers were grazed on alfalfa-grass pasture for the duration of the study and had free access to flocculated, filtered and chlorinated water via an automatic waterer. Fecal samples were collected from steers three times during the experiment July 23, August 27 and September 27, and analyzed for nematode eggs and Giardia sp. cysts. Seventy-five and 83% of the steers in the FBZ group visited the mineral feeder during the first and second treatment periods, respectively. Individual daily mineral and FBZ intake for the first and second treatment periods was 52.9+/-6.6g per day and 10.1+/-1.2mg/kg BW; 72.3+/-8.4 g per day and 11.8+/-1.4 mg/kg BW, respectively. FBZ animals were separated into three groups during each treatment period based on the recommended dose (RD) of FBZ (5 mg/kg/BW), those that received > the RD, those that received < RD but > 50% RD and those that received < 50% of RD. Nematode egg counts and the number of animals infected with nematodes was reduced (p < 0.05) in all cattle that consumed FBZ as compared to control animals. In contrast to nematode eggs, numbers of Giardia cysts was not reduced (p > 0.05) by FBZ as compared to controls in either treatment period. These results may be a reflection of Giardia re-infection occurring following treatment and highlight the need for variation in treatment regimes specifically targeted at the parasite of interest.

Comparison of image analysis, exertion force, and behavior measurements for use in the assessment of beef cattle responses to hot-iron and freeze branding.

Thirty-three steers (328 +/- 2 kg) from a total of 300 animals were randomly selected for a comparison of techniques designed to quantify the behavioral response to painful procedures. The steers were randomly assigned to freeze-branding, (F), hot-iron branding (H), and sham branding (S) treatments. The responses of all steers were videotaped to quantify the amount and intensity of head movements during branding. In addition, the force that steers exerted on the headgate and squeeze chute during branding was recorded using strain gauges and load cells. Behaviors believed to be indicative of pain (tail-flicking, kicking, falling, and vocalizing) were also recorded during branding. These techniques were compared for their effectiveness in measuring behavioral responses of steers during branding. Hot-iron-branded steers had greater maximum and average head movement distances and velocities than F or S steers (P < .05), and F steers only had greater maximum values than S animals (P < .05). The maximum exertion forces obtained from headgate load cells were also greater in H than in F or S steers (P < .05); however, no differences were observed between H and F treatments for squeeze load cell or headgate strain gauge data. Hot-iron-branded steers had the greatest incidence of tail-flicks, kicks, falls in the chute, and vocalizations, and S steers had the least. Results indicate that H steers experienced more discomfort at the time of branding than F and S steers, whereas F steers experienced more discomfort than shams. Image analysis was a superior technique for detecting treatment differences compared with exertion force measurements and frequency counts of tail-flicks, kicks, falls, and vocalization during branding.