Scholarlink

<p>The increasing reliance on electronic health records (EHRs) in smart healthcare systems highlights critical challenges in data security, privacy, and timely diagnosis, particularly for neurological disorders such as epilepsy. Although blockchain technology and deep learning models have shown individual promise, existing solutions often lack an integrated approach to ensure both secure data management and accurate early detection. This paper proposes a novel hybrid framework that integrates cryptographic optimization, blockchain-based secure data sharing, and deep learning for effective epilepsy prediction. The framework employs a Simulated Annealing-Gradient-Based Optimizer (SA-GBO) for optimal cryptographic key generation, Elliptic Curve Cryptography (ECC) for efficient encryption, and SHA3-512 for data integrity. Ethereum blockchain combined with the InterPlanetary File System (IPFS) ensures tamper-proof and decentralized storage with confidential data exchange. For intelligent diagnostics, the Optimized Deep Belief Neural Network (ODBNN) utilizes benchmark datasets such as BONN and BIDS-CHB-MIT to detect epileptic seizures by capturing complex EEG patterns. The Experimental results demonstrate the superior key optimization performance of SA-GBO and enhanced system security via blockchain-IPFS integration. The ODBNN model achieves high accuracy rates of 99.25% and 99.39% on BONN and BIDS-CHB-MIT datasets, respectively, with precision, recall, and F1-scores consistently above 98%. These improvements represent up to a 1.5% increase in accuracy and 2.5-4% gains across other metrics compared to existing methods. These findings underscore the framework’s effectiveness for reliable, precise automated seizure detection, highlighting its strong potential for real-world clinical deployment in EEG-based epilepsy diagnosis.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Blockchain-based framework for secure medical data sharing and disease diagnosis using optimized deep belief networks

  • 期刊论文

  • 发表日期：

    2025年10月17日

查看全文
  • Nikhil Sharma,
  • Prashant Giridhar Shambharkar

摘要

The increasing reliance on electronic health records (EHRs) in smart healthcare systems highlights critical challenges in data security, privacy, and timely diagnosis, particularly for neurological disorders such as epilepsy. Although blockchain technology and deep learning models have shown individual promise, existing solutions often lack an integrated approach to ensure both secure data management and accurate early detection. This paper proposes a novel hybrid framework that integrates cryptographic optimization, blockchain-based secure data sharing, and deep learning for effective epilepsy prediction. The framework employs a Simulated Annealing-Gradient-Based Optimizer (SA-GBO) for optimal cryptographic key generation, Elliptic Curve Cryptography (ECC) for efficient encryption, and SHA3-512 for data integrity. Ethereum blockchain combined with the InterPlanetary File System (IPFS) ensures tamper-proof and decentralized storage with confidential data exchange. For intelligent diagnostics, the Optimized Deep Belief Neural Network (ODBNN) utilizes benchmark datasets such as BONN and BIDS-CHB-MIT to detect epileptic seizures by capturing complex EEG patterns. The Experimental results demonstrate the superior key optimization performance of SA-GBO and enhanced system security via blockchain-IPFS integration. The ODBNN model achieves high accuracy rates of 99.25% and 99.39% on BONN and BIDS-CHB-MIT datasets, respectively, with precision, recall, and F1-scores consistently above 98%. These improvements represent up to a 1.5% increase in accuracy and 2.5-4% gains across other metrics compared to existing methods. These findings underscore the framework’s effectiveness for reliable, precise automated seizure detection, highlighting its strong potential for real-world clinical deployment in EEG-based epilepsy diagnosis.