Scholarlink

<p>Water quality is a crucial factor influencing the health and stability of aquatic ecosystems, where even minor variations in chemical or physical parameters can lead to significant biological stress. Real-time monitoring of such fluctuations remains challenging due to the dynamic and spatially distributed nature of aquatic environments. To address this, the study proposes <i>HydroTwin</i>, a Digital Twin–inspired virtual ecosystem designed for real-time assessment and prediction of aquatic organism health under pollution-induced stress. By integrating high-resolution environmental parameters with biological indicators through a hybrid Convolutional Neural Network enhanced with Bayesian inference, <i>HydroTwin</i> captures the bidirectional interactions between organisms and their environment. A Reputation Aware Fault Tolerant Consensus (RAFTC) protocol within a consortium blockchain ensures secure, tamper-proof, and transparent data provenance. Validated on real-world data of environmental and biological records, <i>HydroTwin</i> achieved a detection latency of 8.37 seconds, Classification Performance (Precision (85.22%), Sensitivity (87.10%), Specificity (87.16%), and F-Measure (86.12%)), Prediction Performance (Error Rate 0.31), demonstrating strong diagnostic and predictive performance. Overall, <i>HydroTwin</i> establishes a scalable and intelligent foundation for early warning systems and proactive aquatic ecosystem governance.</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

A digital twin-inspired intelligent healthcare framework for aquatic animals

  • 期刊论文

  • 发表日期：

    2026年7月23日

查看全文
  • Abdullah Alqahtani,
  • Munish Bhatia,
  • Veerawali Behal

摘要

Water quality is a crucial factor influencing the health and stability of aquatic ecosystems, where even minor variations in chemical or physical parameters can lead to significant biological stress. Real-time monitoring of such fluctuations remains challenging due to the dynamic and spatially distributed nature of aquatic environments. To address this, the study proposes HydroTwin, a Digital Twin–inspired virtual ecosystem designed for real-time assessment and prediction of aquatic organism health under pollution-induced stress. By integrating high-resolution environmental parameters with biological indicators through a hybrid Convolutional Neural Network enhanced with Bayesian inference, HydroTwin captures the bidirectional interactions between organisms and their environment. A Reputation Aware Fault Tolerant Consensus (RAFTC) protocol within a consortium blockchain ensures secure, tamper-proof, and transparent data provenance. Validated on real-world data of environmental and biological records, HydroTwin achieved a detection latency of 8.37 seconds, Classification Performance (Precision (85.22%), Sensitivity (87.10%), Specificity (87.16%), and F-Measure (86.12%)), Prediction Performance (Error Rate 0.31), demonstrating strong diagnostic and predictive performance. Overall, HydroTwin establishes a scalable and intelligent foundation for early warning systems and proactive aquatic ecosystem governance.