EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Sheep pulmonary image recognition (SPIRO): An explainable deep learning framework for macroscopic assessment of ovine pulmonary lesions

Muhammad Furqan Arshad, Tesfalem Mehari Berhe, Giovanni Lai, Ludovica Mura, Antonio Varcasia, Maria Luisa Pinna Parpaglia, Elisabetta Antuofermo, Marta Polinas, Marina Antonella Sanna, Maria Giovanna Cancedda, Md Sabbir Ahmed, Delaram Ghadir und 5 weitere

The Veterinary Journal · 2026-12 · 320 · 106936

DOI: 10.1016/j.tvjl.2026.106936 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Veterinary Journal, 2026-12. Verfasst von Muhammad Furqan Arshad und 16 weiteren Autor:innen. Beteiligte Einrichtung: Department of Veterinary Medicine, University of Sassari, Sassari, Italy. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate detection and classification of pulmonary diseases in sheep are critical for ensuring animal welfare, food safety, and effective disease surveillance in abattoir settings. In this study, we propose a two-stage, explainable, histopathology-informed deep learning framework for the automated classification of ovine lung lesions. The dataset comprises 1052 sheep lung images, annotated through a dual-level diagnostic process involving macroscopic inspection and histopathological confirmation, and classified as healthy, inflammatory, or neoplastic. Four convolutional neural network architectures-VGG16, VGG19, ResNet50, and EfficientNetB0-were benchmarked under identical conditions. EfficientNetB0 achieved the highest observed standalone performance and was integrated into a two-stage pipeline, in which YOLOv11 first localized lung regions and EfficientNetB0 subsequently classified the cropped images. Model interpretability was evaluated using Gradient-weighted Class Activation Mapping (Grad-CAM). The YOLOv11-EfficientNetB0 pipeline achieved an overall test accuracy of 90% and a weighted F1-score of 0.88. Neoplastic lesions were classified with the highest performance, reaching precision, recall, and F1-score values of 0.98, 0.99, and 0.99, respectively, while healthy and inflammatory lungs achieved F1-scores of 0.83 and 0.82, respectively. Grad-CAM visualizations indicated that predictions were mainly driven by diagnostically relevant pulmonary regions. These findings demonstrate the feasibility of explainable artificial intelligence for objective, scalable, and biologically grounded post-mortem assessment of sheep lungs, with potential applications in abattoir surveillance, veterinary pathology, and flock health monitoring.

Schlagwörter

Stichwörter der Autor:innen

Deep learning (DL)EfficientNetB0Sheep lung diseaseVeterinary pathologyWeb applicationYOLOv11

Autor:innen und Institutionen

  • Department of Veterinary Medicine, University of Sassari, Sassari, Italy.
  • Consorzio Nazionale Interuniversitario per le Telecomunicazioni, Dipartimento di Ingegneria dell'Informazione, Università di Pisa, Italy.
  • Department of Veterinary Medicine, University of Sassari, Sassari, Italy; Mediterranean Center for Disease Control, Sassari, Italy.
  • Department of Animal Health, Istituto Zooprofilattico Sperimentale della Sardegna, Sassari, Italy.
  • Department of Veterinary Medicine, University of Sassari, Sassari, Italy; Mediterranean Center for Disease Control, Sassari, Italy. Electronic address: gpburrai@uniss.it.

Zitieren

APA

Arshad, M. F., Berhe, T. M., Lai, G., Mura, L., Varcasia, A., Parpaglia, M. L. P., Antuofermo, E., Polinas, M., Sanna, M. A., Cancedda, M. G., Ahmed, M. S., Ghadir, D., Ghadir, S., Giordano, S., Ligios, C., Adami, D., & Burrai, G. P. (2026). Sheep pulmonary image recognition (SPIRO): An explainable deep learning framework for macroscopic assessment of ovine pulmonary lesions. The Veterinary Journal, 320, 106936. https://doi.org/10.1016/j.tvjl.2026.106936

Vancouver

Arshad MF, Berhe TM, Lai G, Mura L, Varcasia A, Parpaglia MLP, et al. Sheep pulmonary image recognition (SPIRO): An explainable deep learning framework for macroscopic assessment of ovine pulmonary lesions. The Veterinary Journal. 2026;320:106936. doi:10.1016/j.tvjl.2026.106936

BibTeX
@article{Arshad2026,
  author = {Arshad, Muhammad Furqan and Berhe, Tesfalem Mehari and Lai, Giovanni and Mura, Ludovica and Varcasia, Antonio and Parpaglia, Maria Luisa Pinna and Antuofermo, Elisabetta and Polinas, Marta and Sanna, Marina Antonella and Cancedda, Maria Giovanna and Ahmed, Md Sabbir and Ghadir, Delaram and Ghadir, Setayesh and Giordano, Stefano and Ligios, Ciriaco and Adami, Davide and Burrai, Giovanni Pietro},
  title = {Sheep pulmonary image recognition (SPIRO): An explainable deep learning framework for macroscopic assessment of ovine pulmonary lesions},
  journal = {The Veterinary Journal},
  year = {2026},
  volume = {320},
  pages = {106936},
  doi = {10.1016/j.tvjl.2026.106936}
}
Kostenlos starten →