AI-Driven Disease Diagnosis in Falcons Using Machine Learning

Abstract

Falcons are highly prized avian species, particularly in regions where falconry is an important cultural and economic activity. Ensuring their health and detecting diseases at an early stage is crucial, yet traditional veterinary diagnosis can often be delayed due to limited accessibility to expert care. To address this gap, we present Falcon Doctor, an AI-powered, web-based intelligent system designed to predict falcon diseases based on observable symptoms using machine learning algorithms. The system enables users to select symptoms through an intuitive user interface and leverages a trained Random Forest model to predict the most probable disease from a set of ten common falcon diseases. The prediction module was developed and evaluated using a dataset of symptom-disease combinations sourced from a reputed falcon care centre. To ensure accuracy and reliability, the system compares the performance of two classification models—Random Forest and Naïve Bayes—by analysing their evaluation metrics such as accuracy, precision, recall, and F1-score. Based on the comparative results, Random Forest provided superior prediction capability and was selected as the final model for integration. In addition to disease detection, Falcon Doctor includes a chatbot module powered by Gemini AI, enabling users to ask natural language queries regarding symptoms, treatments, preventive measures, and care routines. Moreover, the system provides a hospital locator feature, allowing users to find nearby veterinary clinics specialized in avian medicine. An admin panel is integrated into the platform for managing users, hospitals, and feedback, ensuring streamlined backend operations. The overall goal of Falcon Doctor is to make falcon disease diagnosis more accessible, faster, and efficient, supporting bird conservation efforts and reducing dependency on immediate physical consultations. Future enhancements aim to include real-time vet consultations, broader disease coverage, and intelligent chatbot improvements to elevate the quality of digital avian healthcare.

Introduction

Falcons, especially species like the Peregrine Falcon, hold significant cultural and economic value, particularly in regions where falconry is traditional. Their health is crucial but often hard to monitor due to limited veterinary access and expertise. To address this, the Falcon Doctor system is proposed—a web-based AI tool that predicts falcon diseases from user-selected symptoms using a Random Forest machine learning model trained on a dataset of common falcon diseases.

The system features:

• Symptom-based disease prediction with high accuracy, outperforming alternatives like Naïve Bayes.

• A Gemini-powered chatbot for interactive support on falcon health, care, and treatment.

• A hospital locator to find nearby specialized veterinary clinics.

• Separate user and admin panels for managing users, hospitals, feedback, and data.

Common falcon diseases include Avian Influenza, Aspergillosis, and Newcastle Disease, with symptoms that often overlap, making manual diagnosis challenging. Early detection is critical to prevent severe health issues and zoonotic risks.

The Falcon Doctor combines machine learning with user-friendly interfaces and geolocation tools, making falcon disease management accessible even in remote areas. The system is grounded in extensive literature review, adapting best practices in veterinary AI, and filling a niche where few digital tools for avian health currently exist.

Conclusion

The Falcon Doctor project aims to revolutionize the way falcon breeders manage the health of their birds by leveraging artificial intelligence for disease detection and prediction. By integrating a trained Random Forest model, the system provides accurate disease predictions based on user-selected symptoms, helping falcon breeders and caretakers make informed decisions quickly. Unlike traditional veterinary methods that may require expert availability and manual diagnosis, Falcon Doctor offers a faster, more accessible solution through its interactive web-based interface. The prediction results are not only limited to identifying the most probable disease but also include vital information such as disease overview, transmission methods, preventive measures, control strategies, and health alerts, thereby offering a complete understanding of the condition. To further support users, the system incorporates a geolocation feature that enables easy identification of nearby falcon clinics and hospitals for timely medical intervention. Additionally, the integration of a chatbot powered by Gemini allows for real time conversational support, enabling users to ask questions, clarify doubts, and receive instant responses related to falcon health and care. The user-friendly interface and features like feedback submission and password management make the system accessible to users with varying levels of technical knowledge. On the administrative side, the system provides functionalities to manage users, hospitals, and feedback, ensuring the platform remains organized and updated. By comparing machine learning models during development, the project demonstrated the effectiveness of Random Forest over Naïve Bayes using evaluation metrics such as accuracy, precision, recall, and F1-score. This model selection ensures reliable and consistent performance in disease prediction. The Random Forest model achieved 98.70% accuracy. Overall, Falcon Doctor is a significant step toward modernizing avian healthcare by combining technology, data, and expert insights into one platform.

References

[1] Smith, J., Thomas, L., and Patel, R., “Symptom-based disease prediction in cattle using machine learning classifiers,” International Journal of Veterinary Informatics, vol. 5, no. 2, pp. 45–52, 2018. [2] Kumar, V., and Reddy, K., “Mobile technology for poultry disease diagnosis using symptom selection and image input,” International Journal of Mobile Computing and Multimedia Communications, vol. 10, no. 3, pp. 34–41, 2019. [3] Al-Mahmood, H., Zhang, Y., and Liu, S., “A comparative study of machine learning models for pet disease prediction,” Computers in Biology and Medicine, vol. 123, pp. 103–110, 2020. [4] Rehman, S., Iqbal, A., and Zafar, S., “Development of a rule-based veterinary chatbot for pet care guidance,” Journal of Artificial Intelligence and Soft Computing Research, vol. 11, no. 1, pp. 77–85, 2021. [5] Al-Mutairi, F., Al-Busaidi, M., and Al-Najjar, A., “Digital health systems for falcon care: A contextual analysis of veterinary gaps,” Avian Research Review, vol. 8, no. 2, pp. 21–29, 2022. [6] Zubair Medammal, E. A. A. Shukkur, and P. A. Azeez, \"Health and healthcare of falcons in United Arab Emirates,\" Journal of Ornithology, vol. 147, no. 5, pp. 280, Dec. 2021. [7] A. Panthakkan, Z. Medammal, S. M. Anzar, F. Taher, and H. Al-Ahmad, “AI-Driven Solutions for Falcon Disease Classification: Concatenated ConvNeXt cum EfficientNet AI Model Approach,” Advances in Science and Engineering Technology International Conferences, 2024 [8] A. F. Ahmed, A. I. Elbermawy and A. El-Aziz, “Comparing K-Nearest Neighbors, Random Forest and Naïve Bayes Models to Classify Fetal Health Using Resampling Methods,” International Conference on Advances in Computing, Communication and Information Science, 2023 [9] A. Khan and F. Asghar, “Design of Animal Health Monitoring System Using IoT and ML Techniques,” International Journal of Advanced Computer Science and Applications (IJACSA), vol. 11, no. 12, pp. 541–548, 2020. [10] G. Zaid, Y. Hameed, and M. Salah, “Diagnosis of Avian Diseases Using Deep Learning Approaches,” Procedia Computer Science, vol. 199, pp. 876–883, 2022. [11] M. A. Al Hemeiri, K. Mohteshamuddin, A. Arias de la Torre, B. A. Degafa, and G. Ameni, “A retrospective study on the health problems of falcons in Al Ain, United Arab Emirates,” Emirates Journal of Food and Agriculture, 2022 [12] M. Farooq, K. Nisar, and A. Rasheed, “Design and Evaluation of a Chatbot for Livestock Healthcare Guidance,” IEEE Access, vol. 9, pp. 125456–125468, 2021. [13] M. Saeed, A. Hussain, and M. Waqas, “ML-Based Early Detection of Zoonotic Diseases in Birds,” Biomedical Signal Processing and Control, vol. 73, p. 103424, 2022. [14] M. V. K. Reddy, G. V. P. S. Abhijit, K. S. Nath, and M. Sthyanarayana, “Disease Predictor Based on Symptoms Using Machine Learning,” International Journal for Research in Applied Science & Engineering Technology , 2023 [15] R. S. Rajalakshmi, D. R. Deshmukh, and P. C. Joshi, “Animal Disease Identification Using Ensemble Classifiers,” International Journal of Scientific Research in Science and Technology (IJSRST), vol. 9, no. 2, pp. 223–230, 2022. [16] S. S. Kale, P. H. Wani, and A. G. Thombare, “Disease Detection in Animals Using Machine Learning Techniques,” Journal of Emerging Technologies and Innovative Research (JETIR), vol. 9, no. 4, pp. 543–549, 2022. [17] Shafiqul Abidin,Mutkule Prasad Raghunath,P. Rajashekar,Ashok Kumar,Debarati Ghoshal,Mohammed Ishrat “Identification of disease based on symptoms by employing ML,” International Conference on Inventive Computation Technologies , 2022 [18] T. S. Malviya and D. Jadhav, “Implementation of Smart Disease Prediction System for Pets Using ML,” International Journal of Innovative Research in Science, Engineering and Technology, vol. 10, no. 2, pp. 890–895, 2021. [19] Y. Liu, J. Lin, and Z. Song, “AI-Based Veterinary Diagnostic System Using Ensemble Learning for Livestock Monitoring,” Computers and Electronics in Agriculture, vol. 194, p. 106788, 2022. [20] M. Gadekar, S. Jamadar, P. Pachpute, S. Shinde, and S. Bhosale, “Symptoms Based Disease Prediction,” International Research Journal of Modernization in Engineering Technology and Science,2022 [21] F. Azam, N. Shah, and U. Khan, “Veterinary Disease Diagnosis Using Deep Neural Networks: A Case Study of Poultry Farms,” Sensors, vol. 21, no. 5, p. 1801, 2021.

Copyright

Copyright © 2025 Aaesha V V, Athika Rabeea M, Fathima Manna A K, Sajila A, Swaradh P, Zubair M. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.