FREE SHIPPING over $40
OVER 9000 FIVE-STAR REVIEWS
5% OFF ALL SUBSCRIPTIONS
100% HAPPINESS GUARANTEE
Analyze Diet
0
Home / Equine Research Bank / Animals (Basel) / Development of an Explainable Machine Learning Computational...
Animals : an open access journal from MDPI2026; 16(3); 377; doi: 10.3390/ani16030377

Development of an Explainable Machine Learning Computational Model for the Prediction of Severe Complications After Orchiectomy in Stallions.

Abstract: The objective of the present study was to apply supervised Machine Learning to predict severe complications after equine orchiectomy. A dataset of 612 cases of orchiectomies in stallions was used for the development of a computational model, among which in 8.5% of cases severe complications (colic, continued stallion-like behaviour, evisceration, funiculitis, haemorrhage, and scrotal infection) were diagnosed post-orchiectomy. Three supervised Machine Learning tools were employed: Logistic Regression (12 different models evaluated), Random Forest (64 models), and Gradient Boosting (8 models). For the prediction of the development of severe complications post-orchiectomy, Logistic Regression was the tool that produced the best discrimination measures, where accuracy, precision and recall were 0.9134, 0.8391, and 0.9133, respectively. The results of the analysis for SHapley Additive exPlanations values for the impact of the independent variables in the prediction of the development of complications indicated that (a) the age of the horse and (b) the surgical technique employed were the two variables that mostly influenced the prediction outcome, findings that were unambiguous in the models developed by any Machine Learning tool. The findings of this study indicate that computational models could be used as adjunct tools to support clinical decisions in the peri-operative management of horses.
Get Full Text
Publication Date: 2026-01-25 PubMed ID: 41681358PubMed Central: PMC12897072DOI: 10.3390/ani16030377Google 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 study developed and evaluated machine learning models to predict the risk of severe complications following orchiectomy (castration surgery) in stallions.
  • The goal was to support veterinarians in making informed peri-operative decisions to improve stallion outcomes.

Background and Objective

  • Orchiectomy in stallions can sometimes lead to serious post-surgical complications such as colic, abnormal behavior, evisceration, funiculitis, hemorrhage, and scrotal infection.
  • Timely prediction of these complications before or immediately after surgery could help vets manage care proactively and potentially reduce adverse outcomes.
  • The researchers aimed to leverage supervised machine learning methods to create a predictive computational model using clinical data from past stallion orchiectomies.

Data and Methods

  • The dataset consisted of 612 cases of orchiectomies performed on stallions.
  • Severe complications occurred in 8.5% of these surgeries.
  • Three supervised machine learning algorithms were tested to predict severe complications:
    • Logistic Regression (with 12 different model versions evaluated)
    • Random Forest (64 models tested)
    • Gradient Boosting (8 models tested)

Model Performance

  • Logistic Regression outperformed both Random Forest and Gradient Boosting in predicting severe complications.
  • Performance metrics for the best Logistic Regression model included:
    • Accuracy: 0.9134 (indicating overall correct prediction rate)
    • Precision: 0.8391 (how many predicted complications were true complications)
    • Recall: 0.9133 (the model’s ability to identify all actual complications)

Interpreting Model Predictions: Explainability

  • The researchers used SHapley Additive exPlanations (SHAP) values to interpret how each independent variable influenced model predictions.
  • Two key factors consistently emerged as the strongest predictors of severe complications across all models:
    • Age of the horse — older or younger age possibly linked to complication risk.
    • Surgical technique used — differences in how the orchiectomy was conducted impacted outcomes.
  • This explainability provides clinicians with insights that align with clinical understanding and helps justify reliance on the predictive model.

Conclusions and Implications

  • The study demonstrates that machine learning models, particularly logistic regression, can effectively predict severe post-orchiectomy complications in stallions.
  • Using explainable AI methods like SHAP enhances trust and facilitates clinical adoption by clarifying how input factors relate to risk.
  • Such computational tools could serve as valuable adjuncts to support veterinary clinical decision-making, potentially improving peri-operative care plans and outcomes for horses undergoing orchiectomy.
  • Further validation and integration into clinical workflows would be important next steps to realize practical benefits.

Cite This Article

APA
Tyrnenopoulou P, Kalatzis D, Kiouvrekis Y, Flouraki E, Folias L, Loukopoulos E, Starras A, Chalvatzis P, Tsioli V, Mavrogianni VS, Fthenakis GC. (2026). Development of an Explainable Machine Learning Computational Model for the Prediction of Severe Complications After Orchiectomy in Stallions. Animals (Basel), 16(3), 377. https://doi.org/10.3390/ani16030377

Publication

Animals : an open access journal from MDPI
ISSN: 2076-2615
NlmUniqueID: 101635614
Country: Switzerland
Language: English
Volume: 16
Issue: 3
PII: 377

Researcher Affiliations

Tyrnenopoulou, Panagiota
  • Veterinary Faculty, University of Thessaly, 43100 Karditsa, Greece.
Kalatzis, Dimitris
  • Faculty of Public and One Health, University of Thessaly, 43100 Karditsa, Greece.
Kiouvrekis, Yiannis
  • Faculty of Public and One Health, University of Thessaly, 43100 Karditsa, Greece.
  • Business School, University of Nicosia, Nicosia 1700, Cyprus.
  • Department of Information Technologies, University of Limassol, Limassol 3020, Cyprus.
Flouraki, Eugenia
  • Veterinary Faculty, University of Thessaly, 43100 Karditsa, Greece.
Folias, Leonidas
  • Private Veterinary Practice, 41110 Larissa, Greece.
Loukopoulos, Epameinondas
  • Veterinary Faculty, University of Thessaly, 43100 Karditsa, Greece.
Starras, Alexandros
  • Private Veterinary Practice, 73100 Chania, Greece.
Chalvatzis, Panagiotis
  • Private Veterinary Practice, 50100 Kozani, Greece.
Tsioli, Vassiliki
  • Veterinary Faculty, University of Thessaly, 43100 Karditsa, Greece.
Mavrogianni, Vasia S
  • Veterinary Faculty, University of Thessaly, 43100 Karditsa, Greece.
Fthenakis, George C
  • Veterinary Faculty, University of Thessaly, 43100 Karditsa, Greece.

Conflict of Interest Statement

The authors declare no conflicts of interest.

References

This article includes 31 references
  1. Schumacher J. Testis. 2006. pp. 775–807.
  2. Kilcoyne I, Spier S.J. Castration complications: A review of castration techniques and how to manage complications. Vet. Clin. N. Am. Equine Pract. 2021;37:259–273.
    doi: 10.1016/j.cveq.2021.04.002pubmed: 34116922google scholar: lookup
  3. Robert M.P., Chapuis R.J.J., de Fourmestraux C., Geffroy O.J. Complications and risk factors of castration with primary wound closure: Retrospective study in 159 horses. Can. Vet. J. 2017;58:466–471.
    pmc: PMC5394602pubmed: 28487590
  4. Rodden E.B.K., Suthers J.M., Busschers E., Burford J.H., Freeman S.L. A scoping review on intraoperative and postoperative surgical castration complications in domesticated equids. Equine Vet. J. 2024;56:1115–1128.
    doi: 10.1111/evj.14122pubmed: 38993145google scholar: lookup
  5. Mason B.J., Newton J.R., Payne R.J., Pilsworth R.C. Costs and complications of equine castration: A UK practice-based study comparing “standing nonsutured” and “recumbent sutured” techniques. Equine Vet. J. 2005;37:468–472.
    doi: 10.2746/042516405774479988pubmed: 16163951google scholar: lookup
  6. Baldwin C.M. A review of prevention and management of castration complications. Equine Vet. Educ. 2024;36:97–106.
    doi: 10.1111/eve.13880google scholar: lookup
  7. Basran P.S., MacLean M., Kaba S., Villiers E., Woodward A.P., Sánchez J., O’Sullivan T.L. The unmet potential of artificial intelligence in veterinary epidemiology. Am. J. Vet. Res. 2022;83:392–395.
    doi: 10.2460/ajvr.22.03.0038pubmed: 35353711google scholar: lookup
  8. Sarker I.H. Machine learning: Algorithms, real-world applications and research directions. SN Comp. Sci. 2021;2:160.
    doi: 10.1007/s42979-021-00592-xpmc: PMC7983091pubmed: 33778771google scholar: lookup
  9. Kumar S. How Machine Learning is Revolutionizing Animal Welfare with Enhanced Livestock Monitoring and Communication Technologies. 2025.
  10. Kiouvrekis Y., Vasileiou N.G.C., Katsarou E.I., Lianou D.T., Michael C.K., Zikas S., Katsafadou A.I., Bourganou M.V., Liagka D.V., Chatzopoulos D.C.. The use of machine learning to predict prevalence of subclinical mastitis in dairy sheep farms. Animals 2024;14:2295.
    doi: 10.3390/ani14162295pmc: PMC11350869pubmed: 39199829google scholar: lookup
  11. Hooper S.E., Hecker K.G., Artemiou E. Using machine learning in veterinary medical education: An introduction for veterinary medicine educators. Vet. Sci. 2023;10:537.
    doi: 10.3390/vetsci10090537pmc: PMC10536867pubmed: 37756059google scholar: lookup
  12. Fraiwan M.A., Abutarbush S.M. Using artificial intelligence to predict survivability likelihood and need for surgery in horses presented with acute abdomen (colic). J. Equine Vet. Sci. 2020;90:102973.
    doi: 10.1016/j.jevs.2020.102973pubmed: 32534764google scholar: lookup
  13. Low D., Stables S., Kondrotaite L., Garland B., Rutherford S. Machine-learning-based prediction of functional recovery in deep-pain-negative dogs after decompressive thoracolumbar hemilaminectomy for acute intervertebral disc extrusion. Vet. Surg. 2025;54:665–674.
    doi: 10.1111/vsu.14250pmc: PMC12063718pubmed: 40130766google scholar: lookup
  14. Bourganou M.V., Kiouvrekis Y., Chatzopoulos D.C., Zikas S., Katsafadou A.I., Liagka D.V., Vasileiou N.G.C., Fthenakis G.C., Lianou D.T. Assessment of published papers on the use of machine learning in diagnosis and treatment of mastitis. Information 2024;15:428.
    doi: 10.3390/info15080428google scholar: lookup
  15. Moll H.D., Pelzer K.D., Pleasant R.S., Modransky P.D., May K.A. A survey of equine castration complications. J. Equine Vet. Sci. 1995;15:522–526.
    doi: 10.1016/S0737-0806(07)80421-7google scholar: lookup
  16. Carmalt J.L., Shoemaker R.W., Wilson D.G. Evaluation of common vaginal tunic ligation during field castration in draught colts. Equine Vet. J. 2008;40:597–598. doi: 10.2746/042516408X281162.
    doi: 10.2746/042516408X281162pubmed: 18290261google scholar: lookup
  17. Schumacher J. Testis. In: Auer J.A., Stick J.A., editors. Equine Surgery. 4th ed. Saunders; St. Louis, MO, USA: 2012. pp. 804–840.
  18. Kilcoyne I., Watson J.L., Kass P.H., Spier S.J. Incidence, management, and outcome of complications of castration in equids: 324 cases (1998–2008) J. Am. Vet. Med. Assoc. 2013;242:820–825. doi: 10.2460/javma.242.6.820.
    doi: 10.2460/javma.242.6.820pubmed: 23445295google scholar: lookup
  19. Comino F., Giusto G., Caramello V., Gandini M. Do different characteristics of two emasculators make a difference in equine castration? Equine Vet. J. 2018;50:141–144. doi: 10.1111/evj.12713.
    doi: 10.1111/evj.12713pubmed: 28636747google scholar: lookup
  20. Racine J., Vidondo B., Ramseyer A., Koch C. Complications associated with closed castration using the Henderson equine castration instrument in 300 standing equids. Vet. Surg. 2019;48:21–28. doi: 10.1111/vsu.12960.
    doi: 10.1111/vsu.12960pubmed: 30367692google scholar: lookup
  21. Shaikhanova A., Kuznetsov O., Iklassova K., Tokkuliyeva A., Sugurova L. Interpretable predictive modeling for educational equity: A workload-aware decision support system for early identification of at-risk students. Big Data Cogn. Comput. 2025;9:297. doi: 10.3390/bdcc9110297.
    doi: 10.3390/bdcc9110297google scholar: lookup
  22. Aria M., Cuccurullo C., Gnasso A. A comparison among interpretative proposals for Random Forests. Mach. Learn. Appl. 2021;6:100094. doi: 10.1016/j.mlwa.2021.100094.
    doi: 10.1016/j.mlwa.2021.100094google scholar: lookup
  23. Friedman J.H. Greedy function approximation: A gradient boosting machine. Ann. Statist. 2001;29:1189–1232. doi: 10.1214/aos/1013203451.
    doi: 10.1214/aos/1013203451google scholar: lookup
  24. Hastie T., Tibshirani R., Friedman J.H. Boosting and additive trees. In: Hastie T., Tibshirani R., Friedman J.H., editors. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd ed. Springer; New York, NY, USA: 2009. pp. 337–384.
  25. Zhang C., Zhang Y., Shi X., Almpanidis G., Fan G., Shen X. On incremental learning for Gradient Boosting Decision Trees. Neural Process. Lett. 2019;50:957–987. doi: 10.1007/s11063-019-09999-3.
    doi: 10.1007/s11063-019-09999-3google scholar: lookup
  26. Fukuyo R., Tokunaga M., Yamamoto H., Ueno H., Kinugasa Y. Which method best predicts postoperative complications: Deep learning, machine learning, or conventional logistic regression? Ann. Gastroenterol. Surg. 2026. in press .
    doi: 10.1002/ags3.70145google scholar: lookup
  27. James G., Witten D., Hastie T., Tibshirani R. An Introduction to Statistical Learning with Applications in R. Springer; New York, NY, USA: 2013.
  28. Nakas C., Bantis L., Gatsonis C. ROC Analysis for Classification and Prediction in Practice. CRC Press; Boca Raton, FL, USA: 2023.
  29. Beardshall M. Understanding Explainable AI (XAI): Enhancing Transparency with SHAP and LIME. [(accessed on 3 December 2024)]. Available online: https://www.linkedin.com/pulse/understanding-explainable-ai-xai-enhancing-shap-lime-mike-beardshall-tmjyf/
  30. Lundberg S.M., Lee S.I. A unified approach to interpreting model predictions; Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS’17); Long Beach, CA, USA. 4–9 December 2017; pp. 4768–4777.
  31. Hinton S., Schroeder O., Aceto H.W., Berkowitz S., Levine D. Prevalence of complications associated with use of the henderson equine castrating instrument. Equine Vet. J. 2019;51:163–166. doi: 10.1111/evj.12982.
    doi: 10.1111/evj.12982pubmed: 29926934google scholar: lookup

Citations

This article has been cited 0 times.
Subscribe to our newsletter
Join 99,002 horse owners like you who receive our email updates on horse health and nutrition.