Towards a Connected Heterogeneous All-Medium Integrated Network (CHAIN) for Converged Connectivity Across Land, Sea, Air, and Space
论文提出CHAIN框架，通过全光网络骨干和AI驱动的跨域控制层整合海陆空天各领域通信，解决下一代连接中的跨介质问题。
0 citationsRead paper
Selected work
Representative Papers
Multi-planar 2D-U-Net Segmentation of 3D-CT Abdominal Organs augmented by Spatial Occurrence Maps
This work addresses the challenge of accurate segmentation of five abdominal organs in large-field-of-view 3D CT scans by proposing a two-stage lightweight framework. In the first stage, an axial 2D U-Net coarsely localizes organ regions; in the second stage, multi-planar (axial, sagittal, and coronal) 2D U-Net predictions are fused within the localized region, augmented with a fuzzy 3D spatial occurrence map that encodes anatomical location priors for refined segmentation. The integration of the spatial occurrence map, multi-planar context, and region-of-interest cropping substantially enhances segmentation accuracy. Evaluated on 80 multi-source public CT cases, the proposed method achieves up to a 4% improvement in Dice coefficient over a baseline model without the spatial occurrence map.
0 citationsRead paper
Revising Bloom's Taxonomy for Dual-Mode Cognition in Human-AI Systems: The Augmented Cognition Framework
This study addresses a critical gap in Bloom’s taxonomy within the AI era: its failure to distinguish between individual cognition and human-AI collaborative (distributed) cognition, resulting in a lack of corresponding learning objectives and assessment criteria. To resolve this, the paper proposes the Augmented Cognition Framework (ACF), which reconceptualizes Bloom’s taxonomy by explicitly bifurcating cognitive processes into individual and distributed modes. ACF introduces mode-specific action verbs, asymmetric dependency relationships, and a novel seventh tier—“coordination”—to govern transitions between cognitive modes and optimize human-AI collaboration. As the first framework capable of generating assessable learning outcomes for individual cognition, distributed cognition, and mode governance alike, ACF effectively mitigates core pedagogical risks of the AI age, such as “fluent incompetence.”
0 citationsRead paper
Bridging Theory and Practice: A Stochastic Learning-Optimization Model for Resilient Automotive Supply Chains
UK automotive manufacturing relies heavily on just-in-time (JIT) production but faces dual challenges from demand volatility and supply disruptions. To address this, we propose the first adaptive inventory management framework integrating Bayesian inference with two-stage stochastic optimization, yielding a falsifiable stochastic learning–optimization model that establishes mathematical foundations and theoretical boundaries for AI-enhanced supply chain resilience. Our method unifies Bayesian dynamic updating, Monte Carlo simulation, and real-time decision optimization to enable online modeling of supply–demand uncertainty. In a 365-period simulation study, the framework reduces operational costs by 7.4% under stable conditions and improves performance by 5.7% during supply disruptions. Furthermore, it characterizes the applicability boundary of Bayesian conservatism under abrupt shocks. This work bridges a critical gap by providing the first quantitative validation of synergistic effects between AI and operations research methodologies.
0 citationsRead paper
Machine Learning Risk Intelligence for Green Hydrogen Investment: Insights for Duqm R3 Auction
In the Duqm region of Oman, the absence of historical operational data for green hydrogen facilities impedes quantitative assessment of maintenance and production risks under harsh desert conditions. Method: This study proposes a machine learning–based decision support framework leveraging publicly available meteorological data. It introduces a novel Maintenance Pressure Index (MPI), constructed from environmental proxy variables—including sandstorm frequency, extreme temperature excursions, and humidity volatility—to model equipment degradation trends without requiring prior maintenance records. Contribution/Results: The resulting model enables time-series forecasting of future maintenance requirements, thereby enhancing regulatory foresight and improving the scientific rigor and verifiability of Resource-Response-Reserve (R3) auction decisions. By bridging the data gap in early-stage green hydrogen infrastructure deployment, the approach supports risk-informed operations planning in arid environments.
0 citationsRead paper
Recent publications
Latest Papers
Towards a Connected Heterogeneous All-Medium Integrated Network (CHAIN) for Converged Connectivity Across Land, Sea, Air, and Space
论文提出CHAIN框架，通过全光网络骨干和AI驱动的跨域控制层整合海陆空天各领域通信，解决下一代连接中的跨介质问题。
0 citationsRead paper
Multi-planar 2D-U-Net Segmentation of 3D-CT Abdominal Organs augmented by Spatial Occurrence Maps
This work addresses the challenge of accurate segmentation of five abdominal organs in large-field-of-view 3D CT scans by proposing a two-stage lightweight framework. In the first stage, an axial 2D U-Net coarsely localizes organ regions; in the second stage, multi-planar (axial, sagittal, and coronal) 2D U-Net predictions are fused within the localized region, augmented with a fuzzy 3D spatial occurrence map that encodes anatomical location priors for refined segmentation. The integration of the spatial occurrence map, multi-planar context, and region-of-interest cropping substantially enhances segmentation accuracy. Evaluated on 80 multi-source public CT cases, the proposed method achieves up to a 4% improvement in Dice coefficient over a baseline model without the spatial occurrence map.
0 citationsRead paper
Revising Bloom's Taxonomy for Dual-Mode Cognition in Human-AI Systems: The Augmented Cognition Framework
This study addresses a critical gap in Bloom’s taxonomy within the AI era: its failure to distinguish between individual cognition and human-AI collaborative (distributed) cognition, resulting in a lack of corresponding learning objectives and assessment criteria. To resolve this, the paper proposes the Augmented Cognition Framework (ACF), which reconceptualizes Bloom’s taxonomy by explicitly bifurcating cognitive processes into individual and distributed modes. ACF introduces mode-specific action verbs, asymmetric dependency relationships, and a novel seventh tier—“coordination”—to govern transitions between cognitive modes and optimize human-AI collaboration. As the first framework capable of generating assessable learning outcomes for individual cognition, distributed cognition, and mode governance alike, ACF effectively mitigates core pedagogical risks of the AI age, such as “fluent incompetence.”
0 citationsRead paper
Bridging Theory and Practice: A Stochastic Learning-Optimization Model for Resilient Automotive Supply Chains
UK automotive manufacturing relies heavily on just-in-time (JIT) production but faces dual challenges from demand volatility and supply disruptions. To address this, we propose the first adaptive inventory management framework integrating Bayesian inference with two-stage stochastic optimization, yielding a falsifiable stochastic learning–optimization model that establishes mathematical foundations and theoretical boundaries for AI-enhanced supply chain resilience. Our method unifies Bayesian dynamic updating, Monte Carlo simulation, and real-time decision optimization to enable online modeling of supply–demand uncertainty. In a 365-period simulation study, the framework reduces operational costs by 7.4% under stable conditions and improves performance by 5.7% during supply disruptions. Furthermore, it characterizes the applicability boundary of Bayesian conservatism under abrupt shocks. This work bridges a critical gap by providing the first quantitative validation of synergistic effects between AI and operations research methodologies.
0 citationsRead paper
Machine Learning Risk Intelligence for Green Hydrogen Investment: Insights for Duqm R3 Auction
In the Duqm region of Oman, the absence of historical operational data for green hydrogen facilities impedes quantitative assessment of maintenance and production risks under harsh desert conditions. Method: This study proposes a machine learning–based decision support framework leveraging publicly available meteorological data. It introduces a novel Maintenance Pressure Index (MPI), constructed from environmental proxy variables—including sandstorm frequency, extreme temperature excursions, and humidity volatility—to model equipment degradation trends without requiring prior maintenance records. Contribution/Results: The resulting model enables time-series forecasting of future maintenance requirements, thereby enhancing regulatory foresight and improving the scientific rigor and verifiability of Resource-Response-Reserve (R3) auction decisions. By bridging the data gap in early-stage green hydrogen infrastructure deployment, the approach supports risk-informed operations planning in arid environments.
0 citationsRead paper