Overview
DashboardIntelligenceFlock profitAsk GenemapTake a tour
Animals
SearchUpload flockPedigree & relationships
Records
PhenotypesUpload CSV dataWeighingsReproductionCarcaseFeed efficiencyNutritionMicrobiomeTranscriptsExpression profileG×E reaction normsSampling planUltrasound scansUltrasound scan dayUltrasound measurementsUltrasound calibrationUltrasound validation
Rankings
Live ram indexDynamic indexGenotypes & ssGBLUPCalibrationProperty rollupGenetic gainGenetic gain forecast
Decisions
Ram impactSire picksBid advisorMarket data
Sales
Prospective buyersListings & marketplaceSire carcase reportTell your ram breeder
Land & climate
Forage & drought plannerMethane & carbon
Settings
Farm profileEconomics profileBioeconomic profilePropertiesEnterprisesTeam & permissionsOrganisationActivity logExportStatusTechnical reference

Ultrasound calibration

Calibration kills · rib samples · lab chemical IMF · progress to the first gate

Genemap's ultrasound model is calibrated on what the processor grades: MSA carcase feedback — MSA marble score, AUS-MEAT marble score and, where the plant has one, camera IMF %. Laboratory chemical IMF is kept as an optional gold standard. A model is trusted only once it has been checked on animals it was never trained on, and it is always named after what it was trained on — a model trained on marble scores is a marbling grade, never an IMF %.

Research stage — no IMF numbers. Lab results here are the measuring stick for a future model. No predicted IMF is shown anywhere and nothing enters breeding values until a model passes its test.

1. Progress to the first gate

2. MSA carcase feedback (Livestock Data Link) — the main calibration source

Load the MSA feedback file for a kill (CSV or Excel, as downloaded from Livestock Data Link). Each carcase is matched to your animals by NLIS or EID, and the body number is checked against any calibration-kill record. Every row is checked first; if any row is wrong, nothing is imported and the problems are listed. Every column in the file is kept. Marble scores stay as grades — they are never turned into an IMF %. Scanned animals become calibration records (scan-to-kill over 14 days is flagged); unscanned animals are kept as your farm's kill history.

3. Calibration kills (chemical IMF sampling)

5. Lab results — chemical IMF (optional gold standard)

Columns: sample_code, chemical_imf_pct, optional sample_site, lab_ref, notes. Every row is checked first — unknown sample IDs, duplicates, values out of range, missing sites. If any row is wrong, nothing is imported and the problems are listed. Download the template.

6. Ultrasound and RNA samples on the same animals

7. Repeatability studies

From sessions planned as repeatability studies (each animal scanned twice the same day, re-positioned between scans).