Sürdürülebilir Kalkınma Amaçları
Yazarlar
Yayınlar
- MakaleYalnız künye2026Scopus SJRQ1Innovative solutions for aquaculture: detecting fish diseases with hybrid deep learning model and explainable artificial intelligence
Utku, Anıl; Barut, Abdulkadir; Alofaysan, Hind
Early identification of fish diseases is critical to maintain aquaculture and minimize its financial burden. In the diagnosis of diseases, time is critical and scientists’ experience is imperative, which makes the artificial intelligence-based models necessary. It is intended by this study to integrate deep learning with explainable artificial intelligence methods to establish an artificial intelligence-powered framework for early detection of disease in fish. In this study, we developed a hybrid artificial intelligence system designed to detect diseases in freshwater fish by analyzing skin images. The primary objectives of the system are to minimize economic losses, prevent further disease spread, and ultimately increase aquaculture productivity. For this purpose, we used a dataset consisting of images from seven different disease categories, with 250 images in each class. A new hybrid model, Residual Network-50–Vision Transformer, was developed by integrating the strengths of Residual Network-50 and Vision Transformer architectures to improve classification accuracy. The proposed model was subjected to rigorous comparison with established architectures, including Visual Geometry Group 16, Mobile Convolutional Neural Network Version 2, Efficient Neural Network–Baseline Model B0, standalone Vision Transformer, and the original Residual Network-50. The results demonstrated that Residual Network-50–Vision Transformer outperformed the model, achieving an impressive 99.14% accuracy rate. Furthermore, the study employed local interpretable model-independent explanations to explain which image features most significantly influence the model’s classification decisions, thereby enhancing the interpretability and confidence in the model’s output. The application of such advanced image processing models in combination with explainable AI techniques demonstrates significant potential to support sustainable aquaculture by reducing disease outbreaks and associated financial losses.
AtıfScopus3OpenAlex2
İncele - MakaleYalnız künye2026Scopus SJRQ1Modeling stack voltage and remaining useful life of proton exchange membrane fuel cells using deep neural networks
Dogan, Gurkan; Imak, Andaç
Currently, various energy conversion systems are being developed to reduce environmental impacts. Proton exchange membrane fuel cells (PEMFCs), which stand out with their high efficiency characteristics, have emerged as an important alternative. However, these advantages, along with high costs and uncertainties regarding remaining useful life (RUL) estimation, restrict their further commercialization, especially in the automotive industry. This study presents a forecasting framework using a long short-term memory (LSTM) architecture enhanced with the Bahdanau attention mechanism for predicting RUL and long-term degradation trends. The model thus exhibits a performance that can more effectively capture dependencies in time series and adaptively weight significant time steps. Furthermore, the tree-structured parzen estimator (TPE) was chosen for the appropriate selection of hyperparameters in the developed forecasting framework. The validity, stability, and performance of the proposed model under different operating conditions were examined on the FC1 and FC2 datasets. Experimental studies have shown that the proposed method is the most successful among the best methods, achieving RE metric values of 1.628 × 10−3, 0.003 × 10−1, 99.16 × 10−2, and 0.044 × 10−1 respectively in estimating the best RMSE, MAPE, R2, and RUL.