FREE SHIPPING over $40
OVER 10000 FIVE-STAR REVIEWS
5% OFF ALL SUBSCRIPTIONS
100% HAPPINESS GUARANTEE
Analyze Diet
0
Translational animal science2025; 9; txaf121; doi: 10.1093/tas/txaf121

Assessment of thermal imaging to objectively body condition score mature horses and multiparous gestating beef cows.

Abstract: This study explored whether thermal imaging could provide an objective means of body condition scoring (BCS) horses and multiparous, gestating beef cows. This study consisted of two parts: one part assessed BCS in horses of the Quarter Horse or Thoroughbred breed types while the other evaluated BCS of gestating beef cows. Ground truth BCS were assigned by five to eight trained scorers for each animal. Thermal images were also collected from one or both sides of the body and analyzed for surface temperature. Surface temperature and BCS were evaluated with the whole body, and for five (cows) or seven (horses) specific body regions. Covariates were monitored, including breed, ambient temperature, cloud coverage (%), housing conditions, blanketing status, individual scorer, and scorer's location. Considerable between-scorer variation in BCS assigned was apparent in both species, with 58% to 65% of observations falling within 0.5 BCS of the median. Much of the variability was attributable to scoring the degree of BCS deviation from normal, and 72% to 82% of scores agreed in directional accuracy. Scaled, centered scores within 0.5 units of the median BCS were defined as high-agreement data. Random forest regressions were derived to explore how well thermal-camera-obtained body surface temperature data could be used to estimate BCS of horses and cows using either all available data or the subset of high-agreement data. In both datasets, 15% of observations were used for hyperparameter tuning, 55% for model-training, and 30% for independent evaluation. Unique models were created for each body area assigned a BCS for horses (the neck, shoulder, withers, ribs, loin, tailhead, and overall) or beef cows (the brisket, shoulder, ribs and spine, hooks and pins, and overall). Models estimating horse BCS had lower RMSE (6.71% to 10.4%) and higher accuracy (62.7% to 90.3%) than those estimating cow BCS (RMSE: 6.39% to 15.9%; accuracy: 46.8% to 82.5%). Models derived from all data had higher RMSE (7.76% to 15.9%) and lower accuracy (53.8% to 82.3%) compared with those derived from only high-agreement data (RMSE: 5.80% to 8.18%; accuracy: 66.7% to 90.3%). Overall, thermal imaging shows promise as an objective tool for BCS assignment in both horses and beef cows. However, future refinement of the method for assigning BCS ground truth is important.
© The Author(s) (2025). Published by Oxford University Press on behalf of the American Society of Animal Science.
Get Full Text
Publication Date: 2025-09-15 PubMed ID: 41245604PubMed Central: PMC12614166DOI: 10.1093/tas/txaf121Google Scholar: Lookup
The Equine Research Bank provides access to a large database of publicly available scientific literature. Inclusion in the Research Bank does not imply endorsement of study methods or findings by Mad Barn.
LinkedInCopy
  • Journal Article

Summary

This research summary has been generated with artificial intelligence and may contain errors and omissions. Refer to the original study to confirm details provided. Submit correction.

Overview

  • This research evaluated whether thermal imaging could objectively measure body condition scores (BCS) in mature horses and multiparous gestating beef cows.
  • The study compared thermal surface temperature data to traditional, subjective BCS assigned by trained scorers, aiming to develop accurate, automated scoring models.

Study Design and Species

  • The research consisted of two parts: one focusing on horses of Quarter Horse or Thoroughbred breeds, and the other on multiparous, gestating beef cows.
  • For each animal, multiple trained scorers (five to eight) assigned traditional BCS to serve as a “ground truth” reference.
  • Thermal images were captured from one or both sides of the animal bodies to assess surface temperature patterns.
  • Body regions analyzed included seven areas for horses (neck, shoulder, withers, ribs, loin, tailhead, overall) and five areas for beef cows (brisket, shoulder, ribs and spine, hooks and pins, overall).

Data Collection and Covariates

  • Surface temperature measurements from thermal images were linked with BCS data for both full-body and specific body regions.
  • Several covariates were monitored to account for environmental and individual differences, including:
    • Breed differences
    • Ambient temperature
    • Percentage of cloud coverage
    • Housing conditions
    • Blanketing status
    • Individual scorer identity
    • Scorer’s physical location

Scorer Variability and Data Agreement

  • There was notable variability between scorers in assigning BCS for both horses and cows, reflecting subjectivity in traditional methods.
  • Between 58% and 65% of all scores were within 0.5 BCS units of the median score assigned across scorers, showing moderate agreement.
  • Most disagreement stemmed from how far the BCS deviated from normal, though 72% to 82% of scores agreed on whether conditions were above or below normal (directional accuracy).
  • The researchers defined “high-agreement” data as scores within 0.5 units of the median, isolating more reliable BCS assignments for analysis.

Model Development and Analysis

  • Random forest regression models were developed to predict BCS from thermal imaging data.
  • Two different model sets were trained for each species:
    • One using all available data
    • Another restricted to the high-agreement subset of BCS data.
  • Data were split into subsets for model tuning (15%), training (55%), and independent evaluation (30%).
  • Separate models were built for each body region to explore localized thermal patterns and their relationship with BCS.

Results and Performance Metrics

  • Models predicting horse BCS performed better than those for beef cows:
    • Horse BCS models showed lower root mean squared error (RMSE) ranging from 6.71% to 10.4% and higher accuracy between 62.7% and 90.3%.
    • Cow models had RMSEs from 6.39% to 15.9% and lower accuracies from 46.8% to 82.5%.
  • Using high-agreement data improved model accuracy and decreased RMSE:
    • High-agreement models had RMSEs between 5.80% and 8.18% and accuracy from 66.7% to 90.3%.
    • Models using all data showed higher errors (7.76% to 15.9%) and lower accuracy (53.8% to 82.3%).

Implications and Future Directions

  • The study indicates that thermal imaging has strong potential as an objective alternative to subjective body condition scoring for horses and beef cows.
  • The technology could improve consistency and reduce observer bias in animal health and management assessments.
  • Refining the method for establishing reliable BCS ground truth is critical, as scorer variability impacts model training and accuracy.
  • Additional research may focus on improving imaging protocols, integrating environmental factors, and expanding sample sizes or species.
  • Ultimately, thermal imaging could support more efficient animal welfare monitoring and nutritional management in livestock production.

Cite This Article

APA
Webster AP, Wright RK, Hammond JB, Kotey NA, Gleason CB, White RR. (2025). Assessment of thermal imaging to objectively body condition score mature horses and multiparous gestating beef cows. Transl Anim Sci, 9, txaf121. https://doi.org/10.1093/tas/txaf121

Publication

Translational animal science
ISSN: 2573-2102
NlmUniqueID: 101738705
Country: England
Language: English
Volume: 9
Pages: txaf121
PII: txaf121

Researcher Affiliations

Webster, Alexandra P
  • School of Animal Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, United States.
Wright, Ryan K
  • School of Animal Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, United States.
Hammond, Jillian B
  • School of Animal Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, United States.
Kotey, Naa A
  • School of Animal Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, United States.
Gleason, Claire B
  • School of Animal Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, United States.
White, Robin R
  • School of Animal Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, United States.

References

This article includes 41 references
  1. nAccuWeather. 2024. AccuWeather: Blacksburg, VA. https://www.accuweather.com/en/us/blacksburg/24060/weather-forecast/331254? city=blacksburg (Accessed January 19, 2024).
  2. Anagnostopoulos A. A study on the use of thermal imaging as a diagnostic tool for the detection of digital dermatitis in dairy cattle. J. Dairy Sci. 104:10194–10202.
    pubmed: 34099304
  3. Boosman R, Nemeth F, Gruys E. Bovine laminitis: clinical aspects, pathology and pathogenesis with reference to acute equine laminitis. Vet. Q. 13:163–171.
    pubmed: 1949543
  4. Brooks S A. Morphological variation in the horse: defining complex traits of body size and shape. Anim. Genet. 41:159–165.
    doi: 10.1111/j.1365-2052.2010.02127.xpubmed: 21070291google scholar: lookup
  5. Catalano D N. Estimation of actual and ideal bodyweight using morphometric measurements and owner guessed bodyweight of adult draft and warmblood horses. J. Equine Vet. Sci. 39:38–43.
    doi: 10.1016/j.jevs.2015.09.002pubmed: 31203974google scholar: lookup
  6. Catalano D N. Estimation of actual and ideal bodyweight using morphometric measurements of miniature, Saddle-Type, and thoroughbred horses. J. Equine Vet. Sci. 78:117–122.
    doi: 10.1016/j.jevs.2019.04.008pubmed: 31203974google scholar: lookup
  7. Chen C P J, White R. Common Pitfalls in Evaluating Model Performance and Strategies for Avoidance. .
    doi: 10.2139/ssrn.4829509google scholar: lookup
  8. Chudecka M, Lubkowska A, Kempińska-Podhorodecka A. Body surface temperature distribution in relation to body composition in obese women. J. Therm. Biol. 43:1–6.
    pubmed: 24956951
  9. Daradics Z. Obesity-related metabolic dysfunction in dairy cows and horses: comparison to human metabolic syndrome. Life. 11:1406.
    pmc: PMC8705694pubmed: 34947937
  10. Dugdale A H, Curtis G C, Harris P A, Argo C M. Assessment of body fat in the pony: part I. Relationships between the anatomical distribution of adipose tissue, body composition and body condition. Equine Vet. J. 43:552–561.
    doi: 10.1111/j.2042-3306.2010.00330.xpubmed: 21496091google scholar: lookup
  11. Evans D G. The interpretation and analysis of subjective body condition scores. Anim. Sci. 26:119–125.
    doi: 10.1017/S0003356100039520google scholar: lookup
  12. Ferguson J D, Azzaro G, Licitra G. Body condition assessment using digital images. J. Dairy Sci. 89:3833–3841.
    doi: 10.3168/jds.S0022-0302pubmed: 16960058google scholar: lookup
  13. Gamer M. irr: Various coefficients of interrater reliability and agreement. 2010.
  14. Geor R J, Harris P. Dietary management of obesity and insulin resistance: countering risk for laminitis. Vet. Clin. North Am. Equine Pract. 25:51–65, vi.
    pubmed: 19303550
  15. Henneke D R, Potter G D, Kreider J L. Body condition during pregnancy and lactation and reproductive efficiency of mares. Theriogenology 21:897–909.
    doi: 10.1016/0093-691X(84)90383-2google scholar: lookup
  16. Hogan J. P., Phillips C. J. C.. Starvation of Ruminant Livestock. Nutrition and the welfare of farm animals 2016 p. 29–57.
  17. Jarvis N., McKenzie H. C.. Nutritional considerations when dealing with an underweight adult or senior horse. Vet. Clin. North Am. Equine Pract. 37:89–110.
    doi: 10.1016/j.cveq.2020.12.003pubmed: 33820611google scholar: lookup
  18. Kojima T.. Estimation of beef cow body condition score: a machine learning approach using three-dimensional image data and a simple approach with heart girth measurements. Livest. Sci. 256:104816.
  19. Kristensen E.. Within- and across-person uniformity of body condition scoring in danish holstein cattle. J. Dairy Sci. 89:3721–3728.
    doi: 10.3168/jds.S0022-0302(06)72413-4pubmed: 16899709google scholar: lookup
  20. Kuznetsova A., Brockhoff P. B., Christensen R. H. B.. lmerTest package: tests in linear mixed effects models. J. Stat. Soft. 82:82.
    doi: 10.18637/jss.v082.i13google scholar: lookup
  21. Kuznetsova A., Christensen R. H. B., Brockhoff P. B.. Different tests on lmer objects (of the lme4 package): Introducing the lmerTest package. The R User Conference, useR 2013 p 66.
  22. Liaw A., Wiener M.. Classification and regression by randomForest. R News 2:18–22.
  23. MacCormack J., Bruce J.. The horse in winter—shelter and feeding. Farm Building Progress No. 105 :10–13.
  24. Martin-Gimenez T., Aguirre-Pascasio C. N., de Blas I.. Ultrasonographic assessment of regional fat distribution and its relationship with body condition in an easy keeper horse breed. J. Equine Vet. Sci. 39:69–75.
    doi: 10.1016/j.jevs.2016.01.010google scholar: lookup
  25. Martinson K. L., Coleman R. C., Rendahl A. K., Fang Z., McCue M. E.. Estimation of body weight and development of a body weight score for adult equids using morphometric measurements1. J. Anim. Sci. 92:2230–2238.
    doi: 10.2527/jas.2013-6689pubmed: 24663191google scholar: lookup
  26. Maśko M., Witkowska‐Piłaszewicz O., Jasiński T., Domino M.. Thermal features, ambient temperature and hair coat lengths: Limitations of infrared imaging in pregnant primitive breed mares within a year. Reprod. Domest. Anim. 56:1315–1328.
    pmc: PMC9292174pubmed: 34310786
  27. Mottet R., Onan G., Hiney K.. Revisiting the henneke body condition scoring system: 25 years later. J. Equine Vet. Sci. 29:417–418.
    doi: 10.1016/j.jevs.2009.04.116google scholar: lookup
  28. Okur S.. The effectiveness of thermography in determining localization of orthopedic diseases in horses. Van Veterinary Journal 34:51–54.
  29. Pfeifer L. F., Castro N. A., Neves P. M., Cestaro J. P., Siqueira L. G.. Development and validation of an objective method for the assessment of body condition scores and selection of beef cows for timed artificial insemination. Livest. Sci. 197:82–87.
  30. Pyrek P., Siwinska N., Zak-Bochenek A.. Reproducibility of the body condition score assessment in silesian horses, using the 9-point BCS scale. Vet. Res. Commun. 47:273–278.
    pubmed: 35316481
  31. R core team. R: A Language and Environment for Statistical Computing. 2024.
  32. Rashmi R., Snekhalatha U.. Thermal imaging method in the evaluation of obesity in various body regions–a preliminary study. IOP Conf. Ser: Mater. Sci. Eng. 912:062022.
    doi: 10.1088/1757-899X/912/6/062022google scholar: lookup
  33. Roche J. R.. Invited review: Body condition score and its association with dairy cow productivity, health, and welfare. J. Dairy Sci. 92:5769–5801.
    doi: 10.3168/jds.2009-2431pubmed: 19923585google scholar: lookup
  34. Salamunes A. C. C., Stadnik A. M. W., Neves E. B.. The effect of body fat percentage and body fat distribution on skin surface temperature with infrared thermography. J. Therm. Biol. 66:1–9.
    pubmed: 28477901
  35. Silva L. F. P., Coimbra L. G. S., Eyre K.. Malnutrition of pregnant beef cows and the impact on passive immunity transfer to calves. Anim. Prod. Sci. 63:1970–1982.
    doi: 10.1071/AN23076google scholar: lookup
  36. Suagee J. K.. Effects of diet and weight gain on body condition scoring in thoroughbred geldings. J. Equine Vet. Sci. 28:156–166.
  37. Thatcher C., Pleasant R., Geor R., Elvinger F.. Prevalence of overconditioning in mature horses in southwest Virginia during the summer. J. Vet. Intern. Med. 26:1413–1418.
    pubmed: 22946995
  38. Wagner J. J.. Carcass composition in mature hereford cows: estimation and effect on daily metabolizable energy requirement during winter. J. Anim. Sci. 66:603–612.
    doi: 10.2527/jas1988.663603xpubmed: 3378920google scholar: lookup
  39. Wickham H., Bryan J.. Readxl: Read Excel Files. 2022 R package version 1(2).
  40. Wyse C., McNie K., Tannahil V., Murray J., Love S.. Prevalence of obesity in riding horses in Scotland. Vet. Rec. 162:590–591.
    pubmed: 18453379
  41. Xiong Y., Condotta I. C., Musgrave J. A., Brown-Brandl T. M., Mulliniks J. T.. Estimating body weight and body condition score of mature beef cows using depth images. Transl. Anim. Sci. 7:txad085.
    pmc: PMC10424719pubmed: 37583486

Citations

This article has been cited 0 times.
Subscribe to our newsletter
Join 100,113 horse owners like you who receive our email updates on horse health and nutrition.