Accurate classification of water quality and precise prediction of the Water Quality Index (WQI) are essential for safeguarding aquatic ecosystems and mitigating disease risks in aquaculture. This study systematically evaluates multiple machine learning models including LightGBM, XGBoost, Random Forest, and Support Vector Machines using grid search optimization to enhance performance. Results show that ensemble models, particularly LightGBM and XGBoost, achieved superior accuracy in water quality classification (up to 99.65 %). For WQI prediction, XGBoost regression delivered the highest performance with an R² of 0.9685. SHAP (SHapley Additive exPlanations) analysis was employed to interpret model predictions and quantify feature contributions. Dissolved oxygen and BOD emerged as the most influential predictors, followed by turbidity, nitrate, and electrical conductivity, aligning with known risk factors for aquatic disease outbreaks. These findings underscore the potential of combining advanced machine learning with explainable AI techniques and real-time water quality data to enable proactive monitoring and early warning systems for sustainable aquatic health management.

There are about 33,000 different species of fish and they are visually identified using variety of traits, i.e., size and shape of body, head's size and shape, skin pattern, fin pattern, mouth pattern, scale pattern, and eye pattern etc. In traditional manner, identifying these fish species is always difficult with necked eye. Identification and detection of fish species from images using deep learning and computer vision based techniques is challenging topic among researchers worldwide as an interesting problem. Automatic fish species classification and detection has practical importance for both smart aquaculture and fish industry. AI powered deep learning and computer vision based automatic fish species recognition and sorting system becoming significant factor for making aquaculture industry more productive and sustainable. However, the performance of machine learning classifier greatly depends on the size of image dataset and the quality of the images in the dataset. This article demonstrate BD-Freshwater-Fish, an image dataset contain 4389 images of 12 different species captured in natural environment using HD mobile camera from local fish market of Sylhet and Jessore district of Bangladesh. Twelve (12) different data classes are: Rohu (Labeo rohita), Catla (Catla catla), Mrigal (Cirrhinus cirrhosus), Grass Carp (Ctenopharyngodon idella), Common Carp (Cyprinus carpio), Mirror Carp (Cyprinus carpio var. specularis), Black Rohu (Labeo calbasu), Silver Carp (Hypophthalmichthys molitrix), Striped Catfish (Pangasius pangasius), Nile Tilapia (Oreochromis niloticus), Long-whiskered Catfish (Sperata aor), Freshwater Shark (Wallago attu) has been included in the dataset with a different number of images of different species. The BD-Freshwater-Fish dataset is hosted by Department of Computer Science and Engineering mutually with the help of the Department of Aquaculture, Sylhet Agricultural University, Sylhet, Bangladesh.

As aquaculture increasingly seeks sustainable alternatives to traditional feed ingredients such as fishmeal and fish oil, Moringa oleifera , an eco‐friendly and widely available plant, has emerged as a promising plant‐based feed ingredient. This review synthesizes research from 1988 to 2024, sourced from PubMed, Google Scholar, and Research Gate, to evaluate the nutritional and medicinal potentials of M. oleifera in aquaculture feeds. Rich in protein, essential amino acids, vitamins (A, C, and E), minerals (calcium, iron, and potassium), and polyunsaturated fatty acids, M. oleifera offers significant benefits. Its leaves, flowers, and rhizomes utilized as crude extracts or bioactive compounds promote growth, enhance immunity, and provide antimicrobial defense against parasites, fungi, bacteria, and viruses. Studies demonstrate its effectiveness in improving growth, nutrition, and hematology in aquaculture species while lowering production costs. Furthermore, its antioxidant properties attributed to phenolic and bioactive compounds, bolster fish health, and resilience. This review underscores the potential of M. oleifera to advance sustainable aquaculture practices through its dual nutritional and medicinal benefits.

Surma River, which flows through Sylhet, is an essential source of aquatic resources and water resources in the region. But expansion of urban region has also increased the pollutants in the river’s water which has immersed as a huge problem for Surma’s physiochemical properties. This study has been focused on evaluating the impact of water pollution on these physiochemical properties of Surma River’s water over six months in 2023 by concentrating on its advantages and cons on fish habitat. Using YSI Pro DSS Multi parameter, water samples were collected from four different stations along the river, including Lamakazi (Station 1), Kanishail Ghat (S-2), Sheikhpara (S-3), and Kazir Bazar (S-4) where every station contains 4 Collection points. Key parameters, such as temperature, pH, dissolved oxygen (DO), total dissolved solids (TDS), and nitrate levels, were analyzed. A statistical model was then applied to assess the importance of seasonal variations and their influence on these water quality parameters at the different stations. The results show seasonal variations, with higher temperatures and lower DO levels recorded during the monsoon season, particularly at Station 4, suggesting potential stress on fish populations. The study emphasizes the importance of simultaneous monitoring and pollution management to safeguard the aquatic ecosystem and ensure the sustainability of the river's resources. Future research may focus on the long-term effects of pollution on fish populations and the implementation of effective conservation strategies.