A hybrid deep learning and machine learning framework for automated freshwater fish disease diagnosis towards sustainable aquaculture
Department of Computer Science, Government College (Autonomous), Rajahmundry, AP, India.
Research Article
Open Access Research Journal of Science and Technology, 2026, 16(02), 049-060.
Article DOI: 10.53022/oarjst.2026.16.2.0030
Publication history:
Received on 04 February 2026; revised on 10 March 2026; accepted on 12 March 2026
Abstract:
Aquaculture plays a crucial role in global food security, particularly in South Asia, where freshwater fish farming contributes significantly to nutrition and rural livelihoods. However, frequent disease outbreaks severely impact productivity, leading to substantial economic losses. Early and accurate disease detection remains challenging due to the dependence on expert knowledge, manual inspection, and laboratory testing, which are often time-consuming and inaccessible to small-scale farmers. In this study, a hybrid artificial intelligence framework is proposed for automated freshwater fish disease classification using deep feature embeddings and machine learning models. The framework employs a modified Inception-v3 convolutional neural network to extract high-level discriminative features from fish skin images, followed by supervised classification using multiple machine learning algorithms. The dataset consists of seven disease categories, including bacterial, fungal, parasitic, viral infections, and healthy fish samples, collected from freshwater aquaculture farms in South Asia. A rigorous evaluation protocol based on 10-fold cross-validation and independent testing is adopted to ensure robustness and generalization. Experimental results demonstrate high classification accuracy, precision, recall, and F1-score across all disease categories, with ensemble models achieving superior performance. The proposed system offers a reliable, scalable, and cost-effective solution for early fish disease diagnosis, enabling timely intervention and improved aquaculture management. The findings highlight the potential of integrating deep learning and machine learning techniques for building intelligent decision-support systems in sustainable aquaculture.
Keywords:
Fish Disease Detection; Aquaculture; Deep Learning; Inception-v3; Image Embedding; Machine Learning; Computer Vision; Smart Farming; Artificial Intelligence; Decision Support Systems
Full text article in PDF:
Copyright information:
Copyright © 2026 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0