Accurate forecasting of renewable energy generation is essential for efficient grid management and sustainable power planning. However, traditional supervised models often require access to labeled data from the target site, which may be unavailable due to privacy, cost, or logistical constraints. In this work, we propose FreeGNN, a Continual Source-Free Graph Domain Adaptation framework that enables adaptive forecasting on unseen renewable energy sites without requiring source data or target labels. Our approach integrates a spatio-temporal Graph Neural Network (GNN) backbone with a teacher-student strategy, a memory replay mechanism to mitigate catastrophic forgetting, graph-based regularization to preserve spatial correlations, and a drift-aware weighting scheme to dynamically adjust adaptation strength during streaming updates. This combination allows the model to continuously adapt to non-stationary environmental conditions while maintaining robustness and stability. We conduct extensive experiments on three datasets: GEFCom2012, Solar PV, and Wind SCADA, encompassing multiple sites, temporal resolutions, and meteorological features. The ablation study confirms that each component-memory, graph regularization, drift-aware adaptation, and teacher-student strategy-contributes significantly to overall performance. The experiments show that FreeGNN achieves an MAE of 5.237 and an RMSE of 7.123 on the GEFCom dataset, an MAE of 1.107 and an RMSE of 1.512 on the Solar PV dataset, and an MAE of 0.382 and an RMSE of 0.523 on the Wind SCADA dataset. These results demonstrate its ability to achieve accurate and robust forecasts in a source-free, continual learning setting, highlighting its potential for real-world deployment in adaptive renewable energy systems. Project details are available at https://github.com/AraoufBh/FreeGNN.
Herein, we report the green synthesis of Copper oxide nanoparticles (CuO NPs) made from the aqueous extract of Morus alba leaves; an agricultural waste, rarely used; in order to mitigate persistent pollutants and microbial threats. The biomolecules found in the plant extract acted as the reducing and stabilizing agents. The identification of the generated CuO NPs was performed using UV–Vis spectroscopy, FTIR, XRD, SEM, and zeta potential analysis, revealing distinct structural, morphological, and surface features. The size of CuO NPs ranges between 10 and 35 nm, they displayed an absorption peak at 269 nm, a band gap energy of 2.2 eV, and a zeta potential of − 29.4 mV, indicating high stability and photocatalytic capacity. They demonstrated exceptional photocatalytic efficiency by degrading 97
The aim of the current study is assessing the in vitro anti-inflammatory and antioxidant properties of Laurus nobilis methanolic extract (LNME) and to investigate their impacts on experimental oxidative stress in ulcerative colitis caused by acetic acid (AA). The in vitro antioxidant ability of LNME was evaluated using four tests (DPPH, ABTS, FRAP, and CUPRAC). The anti-inflammatory capacity was assessed using the protein denaturation technique, on the basis of total polyphenol measurement. In an in vivo study, 28 rats were equitably divided into four groups: (1) control group, (2) Laurel group: Rats receiving 250 mg/kg B.W of LNME, (3) AA group: Rats receiving 2 mL/kg B.W of AA (3 The current study suggests that LNME displays anti-inflammatory, and antioxidant potential and a cytoprotective impact supporting its uses to alleviate ulcerative and colonic oxidative stress.
Bee venom (BV) is a well-studied nephroprotective agent, however, its efficacy against ethylene glycol (EG)-induced nephrotoxicity has not been previously investigated. Therefore, this study aimed to assess the protective effect of BV in mitigating EG-mediated alterations in renal function markers, oxidative stress, and histopathological injury in male mice. Renal toxicity was induced in adult male mice by daily oral gavage of 20
The widespread use of medical imaging in telemedicine and EHRs demands robust watermarking that preserves diagnostic quality. Conventional spread spectrum methods, despite their robustness, are limited by shared secret keys, geometric vulnerabilities, and weak resilience to generative AI attacks. This paper proposes a spread spectrum discrete wavelet transform (DWT) watermarking framework for medical images, which uses a deep perceptual masking network (JNDnet) to place the watermark where it remains invisible, a lightweight CNN to adaptively select resilient sub‑bands, and a neural detector that replaces fixed‑threshold correlation for improved extraction accuracy while remaining blind. Experiments on three medical datasets show imperceptibility (PSNR > 44 dB, SSIM > 0.98) and robust performance against common and generative AI attacks, with bit error rates below 6
