National Hog Farmer is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Automating ultrasound measurements of backfat, loin depth

National Pork Board Morris Murphy North Carolina (2).jpg
Results demonstrate the feasibility of using an equation-based approach to automate swine backfat and loin depth measurements.

The objective of this study was to develop an equation-based model for the automation of swine backfat depth and loin depth estimates obtained from ultrasound images. Commercial pigs (n=190; Duroc × (Landrace × Large White)) were reared in a mechanically ventilated facility in north central North Carolina. An Exapad ultrasound machine was used to capture longitudinal images (n=1168) of the last three ribs at 194 (±5) days of age.

To establish standard measurements for model validation, trained personnel obtained backfat and loin depth measurements manually from the images following standard company procedures. Average backfat and loin depth were 0.57 (±0.1) and 2.5 (±0.2) inches, respectively.

NC State Fig 1 081822.JPG

During image processing, the directional change in color values was used to segment ultrasound images into backfat (a.), loin (b.), and rib (c.) regions. Segmented images were then used to estimate backfat and loin depth.

Each trait's respective interquartile range was used to identify outlier predictions and remove poor quality images. After image quality control and filtering, 1018 images (190 pigs) were analyzed for trait prediction and linear models were developed.

On average the algorithm processed 2.07 images per second. Root mean square error of backfat and loin depth prediction were 0.06 and 0.14 inches, respectively.

Results demonstrate the feasibility of using an equation-based approach to automate swine backfat and loin depth measurements. We hope to continue on this path of automating the interpretation of images to aid the selection of profitable, robust swine.

Source: Zack Peppmeier, Suzanne Leonard, Mark Knauer and Jeremy Howard, who are solely responsible for the information provided, and wholly own the information. Informa Business Media and all its subsidiaries are not responsible for any of the content contained in this information asset. 

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.