FedCarbonNet - An AI-Blockchain-IoT Framework for Carbon Footprint Management and Optimization Grefith Gohel, Dev Jani, Vishal Barot Lecture Notes in Networks and Systems, 2026 Addressing the global climate crisis necessitates innovative frameworks for effective carbon footprint management. FedCarbonNet integrates artificial intelligence, blockchain, and the Internet of Things to overcome limitations in traditional carbon management systems, including delayed emissions tracking, insecure trading mechanisms, and inadequate predictive analytics. The framework employs IoT sensors for real-time emissions monitoring, federated learning with Recurrent Neural Networks, Long Short-Term Memory networks, and TabTransformer models for accurate forecasting and optimization, and blockchain with smart contracts for secure and transparent carbon credit transactions. Evaluated on a global dataset spanning power generation and aviation, FedCarbonNet achieves a Mean Absolute Error of 0.082 million metric tons (Mt) with a 95% confidence interval of [0.078, 0.086] Mt and an average emissions reduction of 9.96%. Operating across 50 nodes, the system processes 120 transactions per second with an energy consumption of 0.01 kWh per transaction, demonstrating scalability despite challenges with sparse data and computational demands. Future work includes a pilot project targeting a 12% emissions reduction and global deployment. FedCarbonNet provides a robust, scalable solution for real-time carbon management, enhancing accountability and efficiency in alignment with net-zero objectives.
A Hybrid ResNet-ViT Architecture for Skin Cancer Classification Shubham Gajjar, Om Rathod, Deep Joshi, Harshal Joshi, Vishal Barot 2025 IEEE 4th World Conference on Applied Intelligence and Computing Aic 2025, 2025 Correct identification of different skin lesions, such as melanoma, from dermatoscopic images is a major challenge for automated systems, thus causing delay in early diagnosis. Most of the current models are unable to achieve an optimal balance between local feature extraction and global context understanding. Our work overcomes this challenge by proposing a hybrid ResNet50-ViT model. This architecture is significant because it allows the categorization of seven lesion classes from the HAM10000 dataset by fusing ViT's global context awareness with ResNet's strong local feature extraction. For improving its robustness, data augmentation methods were used. The designed model demonstrated an testing accuracy of 96.3% and showed better discriminative ability, as indicated through ROC values of nearly 1.00 for all test set classes. This hybrid model classifies different types of skin lesions with high accuracy, which is a major breakthrough in computer-aided dermatological diagnosis.
