Back to R&D main

Objective Carcase Measurement CT Project - Phase 1 - Ability for CT to measure meat parameters

Project start date: 01 February 2008
Project end date: 30 June 2009
Publication date: 01 March 2009
Project status: Completed
Livestock species: Sheep, Lamb, Grassfed cattle, Grainfed cattle
Relevant regions: National
Download Report (0.5 MB)

Summary

Background
Radiography is the use of penetrating electromagnetic radiation such as x-rays to view sub-surface structures and objects.  Various technologies can be grouped under this area.  They include:
Projectional radiography or plain film radiography is the practice of producing two-dimensional images using a single energy source x-ray radiation (SEXA). Dual-energy X-ray absorptiometry (DEXA) utilises two X-ray beams with different energy levels that are aimed at the carcase. DEXA is the most widely used and most thoroughly studied bone density measurement technology, and best suited for body composition measurement of meat/fat/bone.Multiple energy x-ray MEXA is also an emerging technology whereby photons of detected xray signal are segmented into 256 energy levels, once again providing a potentially rich source of data applicable to disease detection for instance;Industrial computed tomography (CT) scanning is any computer-aided tomographic process, usually x-ray computed tomography, that (like its medical imaging counterparts) uses multiple irradiation images (usually with x-rays) to produce three-dimensional representations of the scanned object both externally and internally.
MLA has funded a series of projects that are primarily focussed on the demonstration of technical and commerical feasibility in the measurement of key eating quality or objective carcase measurements (OCM) via the above technologies, with the aim of measuring these at line speeds within a processor.
These measurements could then be made available to other participants in the supply chain (such as producers), allowing feedback on eating quality and facilitating value based marketing options.
As a secondary benefit, the same technologies can also be used as front end visioning systems to drive cutting automation (e.g. x-ray front end to LEAP III primal cutters), allowing more precise cutting lines and increasing processing efficiency.
Research
Early R&D
In 1998, project M.796 assessed a technology that could determine the fat content of meat with the aid of nucleonic techniques.
Project A.SCT.0029
This summarised the A.SCT.00xx series of company managed activities into red meat OCM CT technologies.
Project A.SCC.0037
This covered a literature review of the critical impact of x-ray on RFID and shelf life of meat and lamb products.
Project A.SCT.0059
The main objective was to establish a suitable measurement methodology which would allow for accurate comparisons between hot and cold scans, and to assess if there was any difference in the CT measurement results, particularly of fat content, between hot and cold carcases.
The actual trial design consisted of two muscle primals, strip loin and cube roll. These primals were harvested hot from five different carcasses and divided into five replicate slices within each muscle and measured at both 2 hours post mortem, and again after a minimum of 24 hours chill at 5°C.
Establishment of the methodology included both the data capture process and the data analysis method. During the course of the trial work a suitable measurement technique was established for capturing the data, and several analysis methods were explored.
Perhaps most notably a difference was observed in the Hounsfield units of the peak for fat between hot and cold primals. Strip loin primals showed an increase of 51.5 HU from -111 (HU) at hot to -59.5 (HU). At a broad level further analysis was conducted using a thresholding analysis method to determine the total number of voxels within the fat range vs. total number of voxels outside the fat range. A significant correlation was observed between the predicted fat content of whole primals between hot and cold measurements (R2 0.99).
This observation is important as it suggests the ability to determine fat content in either hot or cold primals using CT scanning is the same.
A slight difference in density of muscle was observed between hot and cold scans, evidenced by a shift in the location of the muscle peak on the histogram. Again this trend was significant across all primals types showing an increase of 8 (HU) from hot to cold scans.
A.TEC.0096
The aim of this project was to establish whether CT scanning can be used to determine intramuscular fat in lamb meat. This project value added to an existing Sheep CRC funded project where lamb carcasses from the sheep CRC information nucleus flock (INF) were already being CT scanned for whole body composition analysis.
A.TEC.0100
Experimental studies conducted in Phase 1 (final report submitted to MLA) indicated:That it is feasible to measure fat under hot-boning conditions.Confirmed the large variability that exists within primals for Meat Standards Australia Australia (MSA) grade versus chemical intramuscular fat (%). The correlation between MSA grade and chemical intramuscular fat (%) ranged from 0 to 0.94 across 20 primals. Developed an equation (AdjR2 = 0.84) to estimate intramuscular fat in a primal but this equation still needs an independent evaluation.
Subsequently this cost benefit analysis (CBA) study was conduced (i.e., phase 2).  This study found that if a processor were to purchase a $500,000 CT machine (assumption) and if the current error rate on carcase yield measurement was approximately 1 mm at the P8 site then abattoirs processing > 500 head of cattle/day would benefit from installing a CT machine. Hot boning carcasses and the utilisation of a CT machine at a grading station on the slaughter floor has the potential to increase profitability.
A.MQA.0016 Helical scanning
This project was a 'proof of concept' that was to determine whether an industrial cone beam dual energy X-ray system generating a 3D carcase model by rotating the carcase at line speed, could provide sufficient data capture to directly measure a high number of carcase attributes. The results wee inconclusive owing to limitations in the software processing and camera, with further R&D recommended to address these limitations.
P.PSH.0753 Lamb DEXA producer feedback demonstration
This project was terminated by the processor due to space constraints and costs related to implementation in plant.  Another demonstration site is being pursued.
A.TEC.0123 In-situ CT Evaluation
An investigation into the potential of helical medical CT in the meat processing environment.
P.PIP.0530 Evaluation of 256EXA/MEXA
A preliminary evaluation of MEXA xray detectors.

More information

Project manager: Sean Starling
Primary researcher: IMTEC