• ISSN [ Online ] : 3139-1141

Volume 1 - Issue 5, November - December 2026

  • Home
  • November - December 2026

Cattle Care AI A Deep Learning-Based Framework for Preliminary Cattle Skin Disease Screening


📑 Paper Information
📑 Paper Title Cattle Care AI A Deep Learning-Based Framework for Preliminary Cattle Skin Disease Screening
👤 Authors Shruti Indrajeet Patil, Kalashree Parshuram Pakhare, Prof. Vishwas Kenchi
📘 Published Issue Volume 1 Issue 5
📅 Year of Publication 2026
🆔 Unique Identification Number IJARSMET-V1I5P3
📑 Search on Google Click Here
📝 Abstract
Healthy cattle are central to dairy output, animal welfare, and the income of farming families. Lumpy Skin Disease (LSD) is a viral infection that produces skin nodules, fever, and other clinical signs, and noticing the symptoms early can prompt farmers to seek veterinary help sooner. This paper describes Cattle Care AI, a prototype that labels cattle skin photographs as lumpy skin or normal skin and offers preliminary decision support. Its classifier uses transfer learning: a frozen EfficientNetB0 backbone pretrained on ImageNet feeds a small trainable classification head. Input images are resized to 224 by 224 pixels, and augmentation is applied during training only. On an independent test set of 373 images, in which no original photograph appears in more than one subset, the model reached 93.57 percent accuracy and a macro-averaged F1-score of 93.56 percent. The model is embedded in a Streamlit application that offers image upload, prediction display, symptom checking, risk-oriented recommendations, prediction history, image comparison, and report generation. The system is meant as an educational and preliminary screening aid. It cannot confirm a disease, and it does not replace a veterinary examination or laboratory testing.
📝 How to Cite
Shruti Indrajeet Patil, Kalashree Parshuram Pakhare, Prof. Vishwas Kenchi, "Cattle Care AI A Deep Learning-Based Framework for Preliminary Cattle Skin Disease Screening" International Journal of Advance Research in Science, Management, Engineering and Technology, V1(5): Page(17-21) November - December 2026. ISSN: 3139-1141. www.ijarsmet.com. Published by Scientific and Academic Research Publishing.