Estimating backfat depth, loin depth, and intramuscular fat percentage from ultrasound images in swine

Fast, accurate, and reliable estimates of backfat depth, loin depth, and intramuscular fat percentage in swine breeding stock are used to increase genetic improvement and farm profitability. The objective of this study was to develop an equation-based model for the estimation of swine backfat depth, loin depth, and intramuscular fat percentage estimates obtained from longitudinal ultrasound images. Images were collected from purebred Duroc (n = 230), purebred Large White (n = 154), and commercial (n = 190) pigs born in January 2021 at three farms located in North Carolina. An Exapad ultrasound machine captured longitudinal images across the 10th to 13th ribs at 182 (±12.8 SD) days of pig age. The total number of images processed for Duroc, Large White, and commercial pigs was 1 385, 928, and 1 168 images, respectively. To establish a standard measurement for model comparison, trained personnel following standard company procedures using the BioSoft Toolbox (v4.0.1.2; Biotronics Inc., Ames, IA) obtained backfat and loin depth measurements from the images. Longissimus muscle intramuscular fat percentage was predicted using near-infrared spectroscopy at approximately 22 h postmortem. Backfat and loin depth estimation were conducted only for commercial pigs (n = 190) while intramuscular fat estimation was conducted on all pigs (n = 574). Average backfat depth, loin depth, and intramuscular fat percentage were 14.6 (±2.6 SD) mm, 63.7 (±5.5 SD) mm, and 2.21 (±0.82 SD) %. Image analysis and estimation model development were conducted in MATLAB R2021a. Edge detection via the image gradient was applied to segment ultrasound images into backfat, loin, and rib regions. Segmented images were used to estimate backfat depth, loin depth, and loin intramuscular fat percentage. After image quality control and filtering, the image inclusion rate for each breed-trait combination ranged from 76 to 97%. All Duroc and commercial pigs and 97% of Large White pigs were represented by at least one image for trait estimation. Coefficient of determination of models for the estimation of backfat depth, loin depth, and intramuscular fat percentage were 0.58, 0.57, and 0.56, respectively. Root mean square error of backfat depth, loin depth, and intramuscular fat estimation were 1.65 mm, 3.58 mm, and 0.54%, respectively. Results demonstrate the feasibility of using ultrasound image gradient and an equation-based approach to estimate swine backfat and loin depth, and intramuscular fat percentage. This equation-based approach to estimate carcass traits in live swine can enhance genetic improvement.

Nondestructive Estimation of Lean Meat Yield of South Korean Pig Carcasses Using Machine Vision Technique

In this paper, we report the development of a nondestructive prediction model for lean meat percentage (LMP) in Korean pig carcasses and in the major cuts using a machine vision technique. A popular vision system in the meat industry, the VCS2000 was installed in a modern Korean slaughterhouse, and the images of half carcasses were captured using three cameras from 175 selected pork carcasses (86 castrated males and 89 females). The imaged carcasses were divided into calibration (n=135) and validation (n=39) sets and a multilinear regression (MLR) analysis was utilized to develop the prediction equation from the calibration set. The efficiency of the prediction equation was then evaluated by an independent validation set. We found that the prediction equation—developed to estimate LMP in whole carcasses based on six variables—was characterized by a coefficient of determination (R_v²) value of 0.77 (root-mean square error [RMSEV] of 2.12%). In addition, the predicted LMP values for the major cuts: ham, belly, and shoulder exhibited R_v² values≥0.8 (0.73 for loin parts) with low RMSEV values. However, lower accuracy (R_v⁽²⁾=0.67) was achieved for tenderloin cuts. These results indicate that the LMP in Korean pig carcasses and major cuts can be predicted successfully using the VCS2000-based prediction equation developed here. The ultimate advantages of this technique are compatibility and speed, as the VCS2000 imaging system can be installed in any slaughterhouse with minor modifications to facilitate the on-line and real-time prediction of LMP in pig carcasses.

The use of ultrasound measurement of perirenal fat thickness to estimate changes in body condition of young female rabbits

Forty-two New-Zealand x Californian young female rabbits of 4·5 months of age were used to evaluate the use o f a computerized ultrasound system in estimating changes in body condition. In order to get a wide range of fatness, animals were randomly subjected to four food restriction regimes for 2 weeks. Female rabbits were scanned at six different back sites to estimate perirenal fat thickness, using a real-time ultrasound unit equipped with a 5·0-MHz sector probe. After scanning, all animals were weighed and slaughtered. The weights of hot carcass, cold carcass, perirenal fat, scapular fat, liver and kidneys were recorded. There was a large variation in the weight of the main fat deposits, especially for the perirenal fat (CV = 0·632), indicating its suitability for estimating changes in body condition. Ultrasound measurements of perirenal fat thickness correlated strongly with all carcass fat weight values (r= 0·692 to 0·959;P< 0·001) and the estimated carcass energy content (r= 0·777 to 0·866;P< 0·001) and seem to be more precise predictors than live weight measurements. Multiple regression equations for estimating the main fat deposit weight and estimated carcass energy content using only the live weight as an independent variable hadR2values ranging from 0·47 to 0·59. The accuracy of estimates was improved when ultrasound measurements were used in the regression model (R2values ranging from 0·81 to 0·95). Ultrasound measurement of mean perirenal fat deposit thickness at 3 cm ahead of the 2ndto 3rdlumbar vertebrae was the best predictor for perirenal fat weight (R2= 0·95;P< 0·001), total fat weight (R2= 0·93;P< 0·001) and estimated carcass energy content (R2= 0·90;P< 0·001). A validation group of 11 rabbit does was used to validate the ultrasound regression equations, showing that their mean accuracy was approximately 0·895 and 0·967 for total fat weight and estimated carcass energy content, respectively. Results suggest that ultrasound measurements of perirenal fat thickness may be an accurate method for studying changes in body condition of young female rabbits at different times.

Studies on Intramuscular Fat Percentage in Live Swine Using Real-time Ultrasound to Determine Pork Quality

In the modern pork industry, selection of high intramuscular fat (IMF) in pigs is necessary to improve pork quality. Ultrasound has been used previously to predict subcutaneous fat thickness and IMF in the longissimus muscles of line pigs and Real-time ultrasound has also been reported as a reliable method for estimating IMF in live pigs. So we estimate the correlation between meat quality traits and IMF percentage to investigate the possibility of utilizing real-time ultrasound technology for predicting IMF percentage in line pigs to improve pork quality. The genetic and phenotypic correlations for chemical intramuscular fat (CIMF) and ultrasound intramuscular fat (UIMF) were estimated to be 0.75 and 0.76, respectively. These results suggest that genetic factors strongly influence meat quality. The genetic and phenotypic correlation between UIMF and CIMF were 0.75, 0.76, respectively. The heritability of UIMF and CIMF were 0.48 and 0.50, respectively. So we concluded that CIMF can be replaced with UIMF and Ultrasound machines can be used to test IMF in live swine. In future, UIMF can be utilized to improve pork quality as an alternative to CIMF.

A new simple method for estimating the pork carcass mass of primal cuts and lean meat content of the carcass

The aim of this study was to develop regression equations for estimating lean meat content and the mass of primal cuts (ham, loin, shoulder, belly) based on selected linear measurements. The experiment involved a classification of 141 pigs from the Polish commercial pig population, with hot carcass weight ranging between 60 and 120 kg. The study population was characterised by high variability in terms of analysed measurements. Eight measurements were made including: mass of half-carcass, backfat thickness at different points (over shoulder, over last rib, over the middle of M. gluteus medius), width and thickness of the M. longissimus dorsi measured over the last rib, thickness of the lumbar and the gluteal muscle layer located between the spinal cord and beginning of the M. gluteus medius and waist width – the width of the carcass measured at the narrowest point of the lumbar. A subjective five-point scale was used to score difficulties in obtaining linear measurements (workload rate). The lean meat percentage and mass of cuts were determined by dissection. The study enabled equations to be devised for estimating lean meat content with an accuracy greater than most devices used for carcass classification (estimation error 1.67). Regression coefficients for the mass of primal cuts were: 0.92 for ham, 0.87 for loin, 0.87 for shoulder, and 0.74 for belly. The error of equations used to estimate the mass of primal cuts were: 391 g for ham, 447 g for loin, 263 g for shoulder and 257 g for belly. The workload rate for all the developed regression equations ranged from 1.3 to 1.6 points. The outcome of this study was the development of equations to predict carcass value without the need to use expensive classification equipment.

Electromagnetic scanning to estimate carcass lean content of Taiwan native broilers

To estimate lean content of Taiwan native broiler carcasses accurately, objectively and rapidly, electrical conductivity measurements of broiler carcasses and other relative factors (carcass weight, length and temperature) were used in multiple linear regression analysis for lean prediction. Forty native broiler carcasses, with average market weight 2477.5±465.5 g, were scanned through a 10 MHz electromagnetic field created by an electromagnetic scanner (SA-3203) to measure the total body electrical conductivity (TOBEC) index. After scanning, each broiler carcass was separated into wing, breast, leg and back. Each primal cut was dissected into lean, fat and bone. The weight, length, temperature and TOBEC index of broiler carcass were significantly correlated with lean weight of broiler carcass (P<0.001). Regression analysis for lean estimation with carcass weight, length, temperature and TOBEC index showed higher coefficient of determination (R(2)=0.968) and lower coefficient of variation (CV=4.178) with an equation using beheaded carcass weight, temperature and TOBEC index as variables.

Prediction of lean and fat composition in swine carcasses from ham area measurements with image analysis

Video images of ham cross-sections were recorded from 71 pork carcasses (ranging in weight from 72 to 119kg). Three sets of prediction equations were developed to estimate pork carcass lean and fat composition from video image analysis (VIA) of ham cross-sectional area measurements, 10th rib back fat depth (TENFAT) and hot carcass weight (HCKg). Carcass data of dissected lean and fat in the four primal cuts (ham, loin, Boston button and picnic shoulder) were used as dependent variables in establishing regression equations. The first set of equations combined VIA ham measurements and total ham weight (HTKg). Regression models containing the single variable HTKg times ham percentage lean area (Vol. 1) or HTKg times ham percentage fat area (Vol. 2) accounted for 88% and 68% of the variation in total carcass lean weight (CLKg) and total carcass fat weight (CFKg) from the right side of each carcass, respectively. The second set of equations combined VIA ham measurements and TENFAT (cm). Multiple regression models involving TENFAT, Vol. 1, and Vol. 2 accounted for 91% and 90% of the variation in CLKg and CFKg. The third set of equations used VIA ham measurements, TENFAT and HCKg. Carcass lean weight was best predicted by HCKg, TENFAT, and ham lean area (HLA) (R(2)=.92). Carcass fat weight was best predicted by HCKg, TENFAT, and Vol. 2 (R(2)=.91). Overall correlations showed a high association between Vol. 1 and CLKg (r=.94, P<.0001) and Vol. 2 and CFKg (r=.83, P<.0001). Ham lean area was related to CLKg (r=.74, P<.0001) and ham fat area to CFKg (r=.81, P<.0001). The results of this study indicated video image analysis of ham cross-section slices combined with backfat depth at the 10th rib can be used for accurate estimation of total carcass lean or fat composition.

Evaluation of the prediction of alternative measures of pork carcass composition by three optical probes

The accuracy of 3 optical probes (HGP4 Hennessey Grading Probe, Destron-Feering PG-100 probe, and Giraldo OPTO-Electronic PG-200 probe) to predict the carcass percentage of 5 alternative measures of carcass composition (fat-tissue-free lean, lipid-free soft tissue, lipid-free lean, total fat tissue, and soft tissue lipid) was evaluated on 203 barrows and gilts of 7 genetic populations. The optical probe backfat depths were more closely correlated (P < 0.001, 0.963 to 0.983) than the LM depths (r = 0.695 to 0.734). The optical probe backfat depths were related to lean percentage (r = -0.82 to -0.88), total fat tissue percentage (r = 0.84 to 0.88), and soft tissue lipid percentage (r = 0.86 to 0.87). Optical probe LM depths were weakly related (P < 0.05; r = 0.23 to 0.34) to measures of carcass lean percentage and total fat tissue percentage (r = -0.16 to -0.26). Fat-free lean percentage was predicted with residual SD (RSD) of 3.7% for equations including last-rib midline backfat thickness, 2.4 to 2.7% for equations including optical probe backfat and LM depth, and 2.3% for ribbed carcass measurements. The RSD for the optical probe equations ranged from 2.1 to 2.4% for lipid-free soft tissue percentage and from 2.0 to 2.3% for lipid-free lean percentage. The RSD for the optical probe equations ranged from 2.9 to 3.3% for total fat tissue percentage and 2.5 to 2.8% for soft tissue lipid percentage. Quadratic and cross-product variables of optical probe fat depth, LM depth, and carcass weight were significant (P < 0.05) and reduced the RSD of the equations. Optical probe backfat and LM measurements can be used to predict alternative measures of carcass composition. The predicted relationships in fat-free lean percentage to backfat depth were nearly identical for each optical probe.

Evaluation of the impact of errors in the measurement of backfat depth on the prediction of fat-free lean mass

The development of regression equations to predict carcass composition typically assumes that the independent variables, such as backfat depth, are measured without error. However, technological and operator-specific types of measurement errors do exist. To evaluate the impact of measurement error for backfat depth, Monte Carlo simulation was used to model carcass fat-free lean mass (FFLM) in pigs. In the simulation, FFLM was a linear function of carcass weight and actual backfat depth (ABFD). Carcass weight was assumed to be measured without error, but measurement errors were generated such that the correlation (r(BF)) of the measured backfat depth (BFD) and ABFD ranged from 0.70 to 0.95. Two types of measurement errors were simulated: 1) constant variation that was additive to the variance of ABFD, and 2) variation proportional to the ABFD that was additive to the variance in ABFD. A total of 1,000 replications of 1,000 pigs were simulated. Within each type of measurement error, the absolute values of the regression coefficients and R2 values of the equations decreased as r(BF) decreased. The probability of the backfat depth squared (BFD2) being significant (P < 0.05) in the regression equation was increased when the measurement errors were proportional to ABFD. The occurrence of a significant BFD2 variable was 792 times out of 1,000 replications when r(BF) = 0.95 and increased to 996 times out of 1,000 when r(BF) = 0.85 for BFD with type 2 measurement errors. The inclusion of a CW x BFD variable in the regression equations (P < 0.05) increased (270 to 423 times out of 1,000) as r(BF) decreased from 0.85 to 0.70 for BFD with type 2 errors. Equations developed from BFD with measurement errors resulted in biased predictions of FFLM and changes in FFLM per unit change in BFD. The level and type of measurement errors that exist in the independent variables should be evaluated.

Does vagal nerve stimulation affect body composition and metabolism? Experimental study of a new potential technique in bariatric surgery

BACKGROUND

It has been shown that vagal nerve stimulation (VNS) can affect body mass. The aim of this study was to evaluate effect of VNS on body mass, body composition, metabolic rate, and plasma leptin and IGF-I levels.

METHODS

Eight female pigs were included in the study. Under general anesthesia, a bipolar electrode was implanted on the anterior vagal nerve by laparoscopy. Group A was treated by VNS, and group B was the control. After 4 weeks, stimulation was discontinued in group A and started in group B. The following parameters were evaluated: body mass, body composition, metabolic rate, plasma leptin and IGF-1 levels and intramuscular fat content (IMF).

RESULTS

VNS attenuated body weight gain (2.28 +/- 3.47 kg vs 14.04 +/- 6.75 kg; P = .0112, for stimulation and nonstimulation periods, respectively), backfat gain (0.04 +/- 0.26 mm vs 2.31 +/- 1.12 mm) and IMF gain (-3.76 +/- 6.06 mg/g MS vs 7.24 +/- 12.90 mg/g MS; P = .0281). VNS resulted in lower backfat depth/loin muscle area ratio (0.33 +/- 0.017 vs 0.38 +/- 0.35; P = .0476). Lower plasma IGF-I concentration was found after VNS (-3.67 +/- -11.55 ng/mL vs 9.86 +/- 10.74 ng/mL; P = .0312). No significant changes in other parameters were observed.

CONCLUSIONS

VNS affects body weight mainly at the expense of body fat resources; however, metabolic rate is not affected.

Comparing neutron and X-ray-based dual beam gauges for characterising industrial organic-based materials

Comparisons are made of the neutron gamma transmission (NEUGAT) and dual energy X-ray absorption (DEXA) methods of measuring the composition of organic-based industrial products. A simple model is developed to allow comparisons to be made particularly of the measurement precision and the industrial performance. These gauges have similar applications but the latter gauge is shown to be more suitable for high and variable product throughputs. X-ray tube source and detector combinations provide higher beam fluxes, superior imaging and require less bulky shielding.

Evaluation of procedures to predict fat-free lean in swine carcasses12

The objectives were to develop equations for predicting fat-free lean in swine carcasses and to estimate the prediction bias that was due to genetic group, sex, and dietary lysine level. Barrows and gilts (n = 1,024) from four projects conducted by the National Pork Board were evaluated by six procedures, and their carcass fat-free lean was determined. Pigs of 16 genetic groups were fed within weight groups one of four dietary regimens that differed by 0.45% in lysine content and slaughtered at weights between 89 and 163 kg. Variables in equations included carcass weight and measures of backfat depth and LM. Fat-free lean was predicted from measures of fat and muscle depth measured with the Fat-O-Meater (FOM), Automated Ultrasonic System (AUS), and Ultrafom (UFOM) instruments, carcass 10th-rib backfat and LM area (C10R), carcass last-rib backfat (CLR), and live animal scan of backfat depth and LM area with an Aloka 500 instrument (SCAN). Equations for C10R (residual standard deviation, RSD = 2.93 kg) and SCAN (RSD = 3.06 kg) were the most precise. The RSD for AUS, FOM, and UFOM equations were 3.46, 3.57, and 3.62 kg, respectively. The least precise equation was CLR, for which the RSD was 4.04 kg. All procedures produced biased predictions for some genetic groups (P < 0.01). Fat-free lean tended to be overestimated in fatter groups and underestimated in leaner ones. The CLR, FOM, and AUS procedures overestimated fat-free lean in barrows and underestimated it in gilts (P < 0.01), but other procedures were not biased by sex. Bias due to dietary lysine level was assessed for the C10R, CLR, FOM, and SCAN procedures, and fat-free lean in pigs fed the lowlysine dietary regimen was overestimated by CLR, FOM, and SCAN (P < 0.05). Positive regressions of residuals (measured fat-free lean minus predicted fat-free lean) on measured fat-free lean were found for each procedure, ranging from 0.204+/-0.013 kg/kg for C10R to 0.605+/-0.049 kg/kg for UFOM, indicating that all procedures overestimated fat-free lean in fat pigs and underestimated it in lean pigs. The pigs evaluated represent the range of variation in pigs delivered to packing plants, and thus the prediction equations should have broad application within the industry. Buying systems that base fat-free lean predictions on measures of carcass fat depth and muscle depth or area will overvalue fat pigs and undervalue lean pigs.

Performance of growing-finishing pigs fed diets containing Roundup Ready corn (event nk603), a nontransgenic genetically similar corn, or conventional corn lines

Two studies were conducted at two locations to evaluate growth performance and carcass characteristics of growing-finishing pigs fed diets containing either glyphosate-tolerant Roundup Ready (event nk603) corn, a nontransgenic genetically similar control corn (RX670), or two conventional sources of nontransgenic corn (RX740 and DK647). A randomized complete block design (three and four blocks in Studies 1 and 2, respectively) with a 2 x 4 factorial arrangement of treatments (two genders and four corn lines) was used. Study 1 used 72 barrows and 72 gilts (housed in single-gender groups of six; six pens per dietary treatment) with initial and final BW of approximately 22 and 116 kg, respectively. Study 2 used 80 barrows and 80 gilts (housed in single-gender groups of five; eight pens per dietary treatment) with initial and final BW of approximately 30 and 120 kg, respectively. Pigs were housed in a modified open-front building in Study 1 and in an environmentally controlled finishing building in Study 2. The test corns were included at a fixed proportion of the diet in both studies. Animals had ad libitum access to feed and water. Pigs were slaughtered using standard procedures and carcass measurements were taken. In Study 1, overall ADG, ADFI (as-fed basis), and gain:feed (G:F) were not affected (P > 0.05) by corn line. In Study 2, there was no effect of corn line on overall ADFI (as-fed basis) or G:F ratio. In addition, overall ADG of barrows fed the four corn lines did not differ (P > 0.05); however, overall ADG of gilts fed corn DK647 was greater (P < 0.05) than that of pigs fed the other corn lines. There was no effect (P > 0.05) of corn line on carcass yield or fatness measurements in either study. Differences between barrows and gilts for growth and carcass traits were generally similar for both studies and in line with previous research. Overall, these results indicate that Roundup Ready corn (nk603) gives equivalent animal performance to conventional corn for growing pigs.

Evaluation of the E+V video image analysis system as a predictor of pork carcass meat yield

This study was conducted to assess the ability of the VCS2001 (E+V, Oranienburg, Germany) video image analysis system to predict pork carcass composition. Pork carcasses (n = 278) were selected from a commercial packing plant to differ in weight, Fat-O-Meater (FOM) predicted percentage lean, and gender. Carcasses were imaged three times with the VCS2001, chilled overnight, and then sequentially fabricated into boneless subprimals. The VCS2001 accurately predicted the weight of total saleable product (R2 = 0.88, root mean square error [RMSE] = 1.84) and fat-corrected lean (R2 = 0.92, RMSE = 1.66), but autocorrelation existed between dependent and independent variables. The VCS2001 was acceptably accurate and precise in predicting weights of bone-in ham (R2 = 0.83, RMSE = 0.80), bone-in loin (R2 = 0.74, RMSE = 1.17), loin lean (R2 = 0.77, RMSE = 0.82), belly (R2 = 0.78, RMSE = 0.94), sparerib (R2 = 0.55, RMSE = 0.28), and boneless shoulder (R2 = 0.73, RMSE = 0.79). Weights were more accurately predicted than yields (as a percentage of hot carcass weight) of total saleable product (R2 = 0.47, RMSE = 1.97) or total fat-corrected lean (R2 = 0.44, RMSE = 1.89) using VCS2002, and it did not accurately predict percentages of bone-in ham (R2 = 0.45, RMSE = 1.13), ham lean (R2 = 0.32, RMSE = 1.46), bone-in loin (R2 = 0.29, RMSE = 1.36), loin lean (R2 = 0.56, RMSE = 0.90), belly (R2 = 0.43, RMSE = 1.08), sparerib (R2 = 0.08, RMSE = 0.32), or boneless shoulder (R2 = 0.30, RMSE = 0.88). New prediction models and equations were developed using VCS2001 output variables plus hot carcass weight to predict weight of total saleable product (R2 = 0.89, RMSE = 1.72) and fat-corrected lean (R2 = 0.93, RMSE = 1.55) with very minimal increases in accuracy and precision over that achieved using E+V-programmed models and equations. Use of new prediction models and equations marginally improved accuracy and precision of estimations of total saleable product yield (R2 = 0.56, RMSE = 1.81) and fat-corrected lean yield (R2 = 0.57, RMSE = 1.67) over that achieved using E+V-programmed models and equations. The VCS2001 was not able to predict pork carcass composition more accurately than existing technology; therefore, further development is needed to assure commercial viability of this instrument.

Ractopamine treatment biases in the prediction of pork carcass composition

Carcass and live measurements of 45 barrows were used to evaluate the magnitude of ractopamine (RAC) treatment prediction biases for measures of carcass composition. Barrows (body weight = 69.6 kg) were allotted by weight to three dietary treatments and fed to an average body weight of 114 kg. Treatments were: 1) 16% crude protein, 0.82% lysine control diet (CON); 2) control diet + 20 ppm RAC (RAC16); 3) a phase feeding sequence with 20 ppm RAC (RAC-P) consisting of 18% crude protein (1.08% lysine) during wk 1 and 4, 20% crude protein (1.22% lysine) during wk 2 and 3, 16% crude protein (0.94% lysine) during wk 6, and 16% crude protein (0.82% lysine) during wk 6. The four lean cuts from the right side of the carcasses (n = 15/treatment) were dissected into lean and fat tissue. The other cut soft tissue was collected from the jowl, ribs, and belly. Proximate analyses were completed on these three tissue pools and a sample of fat tissue from the other cut soft tissue. Prediction equations were developed for each of five measures of carcass composition: fat-free lean, lipid-free soft tissue, dissected lean in the four lean cuts, total carcass fat tissue, and soft-tissue lipid mass. Ractopamine treatment biases were found for equations in which midline backfat, ribbed carcass, and live ultrasonic measures were used as single technology sets of measurements. Prediction equations from live or carcass measurements underpredicted the lean mass of the RAC-P pigs and underpredicted the lean mass of the CON pigs. Only 20 to 50% of the true difference in fat-free lean mass or lipid-free soft-tissue mass between the control pigs and pigs fed RAC was predicted from equations including standard carcass measurements. The soft-tissue lipid and total carcass fat mass of RAC-P pigs was overpredicted from the carcass and live ultrasound measurements. Prediction equations including standard carcass measurements with dissected ham lean alone or with dissected loin lean reduced the residual standard deviation and magnitude of biases for the three measures of carcass leanmass. Prediction equations including the percentage of lipid of the other cut soft tissue improved residual standard deviation and reduced the magnitude of biases for total carcass fat mass and soft-tissue lipid. Prediction equations for easily obtained carcass or live ultrasound measures will only partially predict the true effect of RAC to increase carcass leanness. Accurate prediction of the carcass composition of RAC-fed pigs requires some partial dissection, chemical analysis, or alternative technologies.