IoT-Enabled Intelligent System for the Radiation Monitoring and Warning Approach

Table 4

Comparison of existing approaches.

Reference	Dataset used	Problem identified	Methodology	Evaluation measure
[11]	Real-time data	Radiation detection	Neural networks and field-programmable gate arrays (FPGAs)	7.31% accuracy improvement
[12]	Data from typical integrated research reactors	Radiation detection	Monitoring units based on air absorption	Not reported
[13]	12 dosimeter systems	Radiation detection	LiF: Mg, Ti/TLD-100 and LiF: Mg, Cu, P/(MCP-N, RADCARD)	0.8%–4.5% average angular response
[10]	Real-time data and external data	Radiation detection	Machine learning and radiation detection sensors	Not reported
[37]	Real-time data	Wound detection	Temperature and humidity sensors and decision trees	94%
[38]	Real-time data	Health monitoring	Sensors, microcontrollers, fuzzy neural networks	97%
[39]	Real-time data	Fire monitoring and warning systems for smart buildings	Sensors, microcontrollers, fuzzy control algorithms	Not reported
Proposed system	Real-time data	Radiation monitoring and warning approach	Radiation detection sensors, microcontrollers, adaptive boosting classifiers	81.7%