<p>The enhancement of security for sensitive healthcare data stored in cloud environments has become crucial due to growing concerns about data privacy and security. Traditional centralized systems face limitations in terms of data integrity and access control. In this paper, we propose a novel blockchain-based infrastructure integrated with decentralized network nodes to safeguard healthcare data and improve transmission performance. Smart contracts are employed to ensure secure data access and facilitate reliable data transmission to the destination. Medical records, including images and treatment details, are stored securely using off-chain storage solutions such as the InterPlanetary File System and validated through a consensus mechanism. To address malicious threats and enhance data privacy, we introduce the Tri-level Attention Boosted Sliced Recurrent Neural Network, which effectively detects and mitigates security breaches. Furthermore, the Federated Learning model is incorporated to preserve privacy while training models on distributed data, avoiding the need to share sensitive information. The security framework is strengthened with Differential Privacy and Homomorphic Encryption techniques to ensure higher scalability and privacy. The experimental outcome of the Tri-level Attention Boosted Sliced Recurrent Neural Network-based blockchain improves the throughput and diminishes the generation time to 1000 Transaction per second as well as 5&#xa0;s. Further, the efficiency is maximized, and the accuracy achieves 95%. Thus, the proposed method outperformed all other traditional approaches that improve the security and enhance the transmission process with better privacy in real-time applications.</p>