Preliminary results with possible alternative methods 

Robert Romvari

CT scanning

Cooled half carcasses were scanned by means of a Siemens Somatom S40 spiral scanner, with an approximate measuring time of 55 minutes. Depending on the carcass size 120-150 images were taken with total overlapping, with a slice thickness of 10 mms. With the applied scanning procedure the physical picture resolution was 0,96 mm (the size of an elementary dot was 0.96x0.96 mms); consequently the voxel volume was approximately 0.01 cm3.

Data evaluation

Direct volumetric estimation was performed with a special post-processing software. From the total area of all images the muscle and the fat volume was measured. Based on the Hounsfield scale muscle was defined within the interval of 0-200 HU value and fat within 200 - -20 HU, respectively. The spreadsheet with the above mentioned data was complemented in Ireland with the lean meat data originating from the experimental dissection.

Result

Preliminary data evaluation was performed on raw data, which resulted in an r value of 0.96 between the muscle volume values and the EU lean meat percentage.

Further evaluation can be done on the basis of anatomically defined scans. Optional muscle and fat surface and also scale measurement can be performed according to the butchers.

MR scanning

The cooled half carcasses were packed with an isolating layer before the scanning and the core temperature was monitored and stored. (At the first experimental days of both weeks the time between slaughtering and scanning was too short to reach the necessary sample temperature. Therefore the first 3 samples from the 6 were examined at a something higher temperature.)

Scanning was performed by means of a Siemens Magnetom Vision Plus whole body 1.5 T MR tomograph. Depending on the size of the carcasses 5-7 sequences were acquisited with an approximate time of 50 minutes. Sequences were worked out in cooperation with the Cemagref. Images stored on CD-s were sent to French partner immediately after the trial.