Scholarlink

<p>Data management system security and efficiency are crucial in the fast-changing healthcare technology market. Cyberattackers target vital healthcare data. Advanced solutions are needed to handle current healthcare data’s complexity and size. This study uses EffiIncepNet, an ensemble deep learning network, for health data categorization and blockchain security. The goal is to develop a robust, efficient, scalable system that minimizes security threats and performs well. EffiIncepNet improves classification accuracy and execution efficiency by combining EfficientNet and Inception-ResNet-v2 architectures. EffiIncepNet data categorization, internal blockchain implementation to stop suspicious transactions, and continuous network monitoring with algorithms that look for strange behavior are all parts of a three-step security structure. The model is tested on Human Vital Signs, IEEE-CIS Fraud Detection, and Elliptic++. The Human Vital Signs dataset had a 98% success rate and a 96% AUC. The IEEE-CIS Fraud Detection dataset had a 97% success rate and a 98% AUC. And the Elliptic++ dataset had a 94% success rate and a 96% AUC. The model’s shortened training times made it scalable for massive datasets. The new BPRWS (Balanced Accuracy Recall Weighted Score) metric confirmed the model’s accuracy and recall. EffiIncepNet redefines the security and dependability of blockchain-based healthcare systems. Its high classification accuracy, AUC, execution efficiency, and security characteristics make it excellent for the cybersecurity of sensitive health data. This study demonstrates how useful EffiIncepNet is. It opens up opportunities to use machine learning and blockchain to improve healthcare security and efficiency.</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

Innovative AI ensemble model for robust and optimized blockchain-based healthcare systems

  • 期刊论文

  • 发表日期：

    2025年1月8日

查看全文
  • Abdulwahab Ali Almazroi

摘要

Data management system security and efficiency are crucial in the fast-changing healthcare technology market. Cyberattackers target vital healthcare data. Advanced solutions are needed to handle current healthcare data’s complexity and size. This study uses EffiIncepNet, an ensemble deep learning network, for health data categorization and blockchain security. The goal is to develop a robust, efficient, scalable system that minimizes security threats and performs well. EffiIncepNet improves classification accuracy and execution efficiency by combining EfficientNet and Inception-ResNet-v2 architectures. EffiIncepNet data categorization, internal blockchain implementation to stop suspicious transactions, and continuous network monitoring with algorithms that look for strange behavior are all parts of a three-step security structure. The model is tested on Human Vital Signs, IEEE-CIS Fraud Detection, and Elliptic++. The Human Vital Signs dataset had a 98% success rate and a 96% AUC. The IEEE-CIS Fraud Detection dataset had a 97% success rate and a 98% AUC. And the Elliptic++ dataset had a 94% success rate and a 96% AUC. The model’s shortened training times made it scalable for massive datasets. The new BPRWS (Balanced Accuracy Recall Weighted Score) metric confirmed the model’s accuracy and recall. EffiIncepNet redefines the security and dependability of blockchain-based healthcare systems. Its high classification accuracy, AUC, execution efficiency, and security characteristics make it excellent for the cybersecurity of sensitive health data. This study demonstrates how useful EffiIncepNet is. It opens up opportunities to use machine learning and blockchain to improve healthcare security and efficiency.