The Effect of Rider:Horse Bodyweight Ratio on the Superficial Body Temperature of Horse's Thoracolumbar Region Evaluated by Advanced Thermal Image Processing.
Authors: Domino Małgorzata, Borowska Marta, Trojakowska Anna, Kozłowska Natalia, Zdrojkowski Łukasz, Jasiński Tomasz, Smyth Graham, Maśko Małgorzata
Journal: Animals : an open access journal from MDPI
Summary
Rider-to-horse bodyweight ratio significantly influences thermal patterns across the thoracolumbar region, yet conventional infrared thermography has proven inconsistent at detecting these differences across certain weight categories. Researchers applied advanced texture analysis to thermal images from 12 horses worked by six riders in light (<12%), moderate (>12–15%), and heavy (>15–18%) bodyweight ratio groups, capturing thermographic data before and after standardised exercise and processing images through histogram statistics, gray-level run-length matrix, and gray-level co-occurrence matrix approaches. Whilst conventional thermal measurements failed to discriminate consistently between groups, texture heterogeneity measures—particularly inverse difference moment, sum entropy, entropy, difference variance, and difference entropy derived from the red colour component—demonstrated reliable, measurable changes correlated with increasing rider weight. This finding suggests that advanced texture analysis offers superior sensitivity for detecting biomechanical stress responses in the horse's back compared to standard thermal imaging techniques, potentially providing equine professionals with a more objective tool for evaluating saddle fit and appropriate rider-horse matching. Given that inappropriate bodyweight ratios are increasingly prevalent as the human population becomes heavier, access to such diagnostic refinements could better protect ridden horses from soft tissue strain and long-term spinal compromise.
Read the full abstract on PubMed
Practical Takeaways
- •Advanced thermal image texture analysis may provide a more sensitive objective assessment tool than conventional thermal imaging for detecting physiological stress from excessive rider weight burden
- •Rider:horse bodyweight ratios in the 10-18% range produce measurable differences in thoracolumbar surface thermal patterns that could inform better rider-horse matching decisions
- •Texture analysis methodology using RGB component decomposition requires further validation but shows promise as a non-invasive welfare monitoring tool for riding establishments
Key Findings
- •Texture analysis features (InvDefMom, SumEntrp, Entropy, DifVarnc, DifEntrp) detected significant differences in thermal image heterogeneity across three rider:horse bodyweight ratio groups (10-12%, >12≤15%, >15<18%), whereas conventional thermal imaging did not
- •Among 372 texture features analyzed, 95 HS, 48 GLRLM, and 96 GLCM features differed with exercise; 29 HS, 16 GLRLM, and 30 GLCM features differed with bodyweight ratio
- •Red component analysis of thermal images provided the most consistent measurable differences in texture heterogeneity measures across bodyweight ratios