A study of operator effects on ultrasonic measures of fat depth and longissimus muscle area in cattle, sheep and pigs

Abstract

Use of real-time ultrasonography in meat animal improvement programs has increased largely as a result of the perceived accuracy of such instruments in measuring longissimus muscle area (LMA) and fat depth (BF). The objective of this study was to examine operator effects for ultrasonic determination of BF and LMA in market-weight cattle, sheep and pigs. The day prior to slaughter, five persons scanned 45 pigs, 32 sheep and 18 cattle at the last rib using an A-mode Ithaco Model 731AM101 body composition meter (ABF) and a Johnson and Johnson Ultrasound 210DX B-mode scanner fitted with a 3.0 MHz probe. All operators measured their own B-mode recordings (OBF, OLMA); an independent, experienced technician also obtained measurements from all recordings (TBF, TLMA). Species x machine x operator interaction effects were detected (P less than .001) for ultrasonic BF. Within-species analyses also revealed that the machine x operator interaction was important (P less than .001) for all species. Analyses of ABF, OBF, TBF, OLMA and TLMA detected species x operator interactions (P less than .001) for all traits except TLMA (P less than .20). Assuming a random model with animal, operator and residual effects, variance components were estimated by maximum likelihood for each species. Repeatabilities, calculated as the intraclass correlation among animals for BF and LMA scanned and interpreted by different operators, were .13, .21 and .51 (OBF) and .28, .15 and .29 (OLMA) for cattle, sheep and pigs, respectively, indicating the importance of operator effects. Repeatabilities of TBF were .90, .42 and .68 and of TLMA were .19, .55 and .65 for cattle, sheep and pigs, respectively, indicating the importance of image interpretation in determining operator effects associated with use of the 210DX B-mode scanner. Technical training standards must, therefore, be established for operators of such equipment.