Abstract
Remote Patient Monitoring Systems (RPMS) require efficient resource management to prioritize life-critical data in latency-sensitive healthcare environments. This research introduces an Integrated Queuing and Certainty Factor Theory (IQCT) model aimed at optimizing bandwidth allocation and task scheduling within fog-edge-cloud architectures. IQCT prioritizes patient requests in real time by classifying them into emergency, warning, and normal categories using certainty factor (CF)-based urgency assessment. Simulated on Raspberry Pi fog nodes with the UCI Heart Disease dataset, its performance was benchmarked against FCFS, PQ, and WFQ using metrics such as latency, energy consumption, and response time. IQCT reduced latency for emergency requests by 54.5% and improved network efficiency by 30.08% compared to FCFS. It also lowered response and execution times by 49.5% and 36%, and decreased fog-layer energy consumption by 30.8%.
Introduction
In Remote Patient Monitoring Systems (RPMS), timely data processing is critical. Traditional cloud computing architectures often struggle with high latency due to the distance between data sources and central servers. While fog computing addresses this by moving processing power to the network edge, it introduces challenges related to limited local resources. Existing scheduling models like First-Come-First-Served (FCFS) or Weighted Fair Queuing (WFQ) often fail to distinguish between routine data and life-critical alerts, leading to potential delays in emergency responses. This paper proposes the Integrated Queuing and Certainty Factor Theory (IQCT) model to intelligently prioritize medical data based on clinical urgency.
Methodology
The IQCT model operates within a hierarchical three-layer architecture designed to balance local responsiveness with global analysis. The Sensing Layer consists of various medical sensors and IoT devices (e.g., ECG, blood pressure monitors, pulse oximeters) that collect raw physiological data from patients. This data is transmitted to the Fog Layer, which acts as the primary processing hub.
The core innovation of the IQCT model is the integration of Certainty Factor (CF) Theory with an M/M/1 Priority Queuing system. CF Theory provides a mathematical framework for managing uncertainty in rule-based systems, which is particularly useful in medical diagnosis where sensor readings may be noisy or ambiguous. The model utilizes a set of clinical rules derived from Braunwald's heart disease guidelines and validated by a panel of expert cardiologists. By correlating multiple attributes—such as heart rate, blood pressure, and chest pain type—the system calculates a certainty factor that represents the urgency of the patient's condition.
Based on these calculated weights, incoming requests are dynamically categorized into three distinct queues:
- Emergency Queue: For life-critical situations requiring immediate intervention.
- Warning Queue: For patients showing abnormal signs that require monitoring but are not immediately life-threatening.
- Normal Queue: For routine data collection and non-urgent status updates.
The system employs a pre-emptive priority scheduling mechanism. If an emergency packet arrives while the fog node is processing a lower-priority task, the current task is suspended, and the emergency request is moved to the front of the line. This ensures that critical medical alerts are never "starved" by routine traffic.
The model was rigorously tested using the UCI Heart Disease dataset, which contains 297 patient records with 13 clinical features. The simulation environment utilized Raspberry Pi nodes to mimic real-world edge computing constraints, with data arriving at a rate of 1.3 requests per millisecond to simulate a high-load hospital environment.
Results and Discussion
The IQCT model was benchmarked against standard scheduling algorithms, including First-Come-First-Served (FCFS), Priority Queuing (PQ), and Weighted Fair Queuing (WFQ). The results demonstrated that IQCT significantly outperforms these traditional methods in every key performance indicator:
- Latency Reduction: Emergency request latency was reduced by 54.5% compared to FCFS, ensuring that critical alerts reach medical personnel with minimal delay.
- Improved Response Times: The model achieved a 49.5% improvement in response time and a 36% reduction in execution time for emergency tasks.
- Energy Efficiency: By performing intelligent triage at the fog layer and reducing the volume of data transmitted to the cloud, the model lowered fog-layer energy consumption by 30.8%.
- Network Performance: Overall network usage time and efficiency improved by 30.08%.
Statistical validation using the Wilcoxon signed-rank test confirmed that IQCT's performance gains were consistent and statistically valid at both 95% and 99% confidence levels. These findings suggest that the integration of expert-driven certainty factors allows for a much more nuanced and effective prioritization of medical data than simple rule-based or health-blind scheduling.
Conclusion and Future Work
The IQCT model offers a robust framework for prioritizing critical healthcare data at the network edge. While the simulation results are promising, future work will focus on real-world deployment with physical medical sensors. Additionally, the researchers plan to address security concerns by integrating blockchain-based identity management and AI-driven encryption. Further refinement of Quality of Service (QoS) thresholds and testing across diverse edge hardware will also be pursued to enhance the model's generalizability and reliability in clinical settings.