Transparent Peer Review By Scholar9
AI-Driven Classification of External Diseases in Cattle
Abstract
Timely detection and accuracy of diseases in dairy and poultry industries are extremely important features on which cattle health depends and elimination minimize economic losses. External diseases like Lumpy Skin disease (LSD), Foot & Mouth disease (FMD) and Infectious Bovine Keratoconjunctivitis (IBK) are serious threats to livestock productivity and livelihood of rural community. Traditional diagnostic methods frequently used manual observation for diagnosis sometimes, and this is a time-consuming and error-prone approach. The present initiative thus revolves around employing a customized CNN model for automated classification of these diseases in cattle based on visual symptoms observed in images by an AI-oft driven approach. Artificially intelligent this system drives on sensor inputs such as physiological monitoring through IoT sensors, particularly the MLX90614 sensor for detecting the surface temperature, and a MAX30102 sensor for heart rate and oxygen saturation measurement. Thus, the barrage of inputs now enables the multi-dimensional evaluation of cattle health. The experiment results show promising accuracy and efficiency in classifying diseases. This indicates that the AI-based framework has a great chance in improving livestock health management, reduction in disease spread, and thereby ensuring sustainable and economical viable farming practices.
Niravkumar K Patel Reviewer
Hello Researcher,
I hope you are doing well. I can see the process of handling the disease for the body, but I can't see that it's handled properly. I have feedback on this research to improve the diagnosis process of the disease. Which is not there in the research paper. Please try to improve because this is one important key factor to identify the disease
If you can do some process like this, then it will be easy to handle the healthcare as per your research.
This is an example of the pipeline of the algorithms.
Data Collection
plaintext CopyEdit - Collect patient data: symptoms, test results, imaging, EHRs - Ensure anonymization and HIPAA/GDPR compliance
Data Preprocessing
plaintext CopyEdit - Clean missing data - Normalize test values - Encode categorical variables - Extract features from images using CNNs
: Model Training
plaintext CopyEdit - Split data into train/test sets - Choose algorithm (e.g., XGBoost for lab tests, CNN for imaging) - Train model on labeled diagnosis data - Validate with cross-validation
Model Evaluation
plaintext CopyEdit - Use metrics: Accuracy, Precision, Recall, AUC-ROC - Avoid overfitting, check generalizability
Diagnosis Prediction
plaintext CopyEdit - Deploy model in a hospital system or app - New patient data → real-time disease prediction
Decision Support
plaintext CopyEdit - Output diagnosis probability or recommendation - Suggest additional tests or flag urgent cases
Please let me know. If you have any questions. this paper needs lots of improvements.
Niravkumar K Patel Reviewer
23 Jun 2025 10:34 AM
Approved
Hello Researcher,
I hope you are doing well. I have given few comments to improve this research. If you can correct it then it will be good for your research.
IJ Publication Publisher
Dear Sir,
Thank you for your timely response. We appreciate your valuable comments and have forwarded them to the respective author for their consideration.
Niravkumar K Patel Reviewer