P.PSH.1181 - Microwave systems for measuring lamb and beef carcase composition and intramuscular fat

Summary

This project involved the design and testing of a low cost microwave system using either an array of sensors to measure lamb and beef carcase composition in live animals and in commercial abattoirs, or using single site probes re-designed to detect meat quality attributes at specific sites. The project was divided into three parts. The first was structured around the design and validation of a microwave system with an array of sensors designed specifically for use in the lamb industry to measure carcase composition in both live animals and their carcases. The second was to repeat this development work in beef, with this likely to require completely reworked designs for transceivers, antenna and prediction algorithms, due mainly to the size/scale differences of cattle compared to sheep. The third part was to design and validate a single site microwave system for use in detecting intramuscular fat. Ultimately this project aimed to deliver a range of microwave devices for measuring fat depth in live sheep and cattle as well as their carcases.

Objectives

The key objective of this is to produce accurate and low cost microwave solutions that can be used by industry to provide enhanced feedback to producers, develop more transparent pricing systems, and to optimise the use of carcases to meet market specifications and volumes.

The project was divided into three parts. The objectives of each were as follows:

• Part 1: Sheep carcase composition prediction - Design and develop a microwave system with an array of sensors for measuring lamb carcase composition in both live animals and their carcases.
• Part 2: Cattle carcase composition prediction - Design and develop a microwave system with an array of sensors for measuring cattle fat depth in both live animals and their carcases.
• Part 3: Sheep and Cattle meat quality trait prediction - Design and develop a single-site microwave system for measuring intramuscular fat in beef and lamb carcases.

Key findings

The Sheep carcase array system was developed and scanned carases in a vertical position hanging on the rail, with the eight antenna measuring 160 points from the rump to the shoulder. The carcase array system could predict CT traits; CT Fat% (R2 0.49, RMSEP 2.17), CT Lean% (R2 0.34, RMSEP 2.68) and CT Bone% (R2 0.36, RMSEP 1.76). The second system developed was a farm-based live animal measurement device. The device was a handheld dual antenna microwave system and when positioned over the C-site could predict ultrasound measured C-site fat depth (R2 0.42, RMSEP 0.76) and eye muscle depth (R2 0.38, RMSEP 3.49). Further work is required to improve signal consistency and calibration across the antenna, and to change the structural design of the array. It is expected this will enhance the existing predictions reported.

Multiple experiments evaluated the ability of a portable microwave system (MiS) to predict IMF% in
lamb and beef carcases. MiS could predict IMF% in lamb carcases with an average RMSEP of 0.654 and R2 of 0.44 using the OCP and RMSEP of 0.671 and R2 of 0.42 using the VPA. Predictions of MiS coupled to VPA met AUS-MEAT national accreditation standards in the first three quarters of the data range. Accreditation standards were not met in quarter 4, the high IMF% range.

In Experiment 2 MiS could predict lamb IMF% with some precision in only 1 of 5 kill groups. Experiment 4 demonstrated the ability of MiS VPA to predict hot carcase IMF% in beef. There was good prediction in one of four kill groups, with an average R2 of 0.46 and RMSEP 2.19. This work showed some potential for MiS to predict IMF% in both beef and lamb. However, these experiments have demonstrated considerable variability in precision which cannot be immediately explained by the current data.

Benefits to industry

In the red meat industry, overfat carcases cause significant economic loss due to the labour required for trimming, and the waste that it represents. Subcutaneous fat is the most variable component both in its amount and distribution in the carcase or live animal, and on this basis the measurement of fat depth in the carcase is the cornerstone of most carcase classification schemes for beef, lamb and pork worldwide. The fat depth measurement is often taken manually, however, this has the disadvantage of being destructive, subjective and time-consuming. The ability to estimate fat depth accurately via a non-invasive and non-destructive technique is therefore highly sought after. It will be crucial for carcase grading, but also for performance testing of live animals to ensure compliance to market specifications prior to slaughter.

The portable handheld design developed in this project offers a relatively affordable option for processors wanting a whole carcase objective fatness estimation, without requiring extensive modifications to existing abattoir infrastructure, as is currently required by whole carcase measurement systems such as dual energy x-ray absorptiometry (DEXA), Frontmatec BCC-3TM (Frontmatec, Denmark) and E+V (Marel, Germany).

MLA action

MLA will disseminate these findings to industry and encourage adoption of the technologies successfully developed.

Future research

The multiple microwave devices developed proved accurate to be able to measure fat depth in cattle and sheep and their carcases and by extension showed promise in prediciting LMY%. However, the experiments undertaken to measure IMF% in carcases showed variable results. The reason for the variable prediction performance between datasets could not be determined, thus future work is required to understand the variables that impact this prediction and develop new antenna, data analysis and calibration methods to counter these effects.