Expanding the Applicability of Air Dose Rate Conversion Artificial Neural Networks Constructed from Fukushima Experience to Other Detectors

Background: This study validated the applicability of employing artificial neural networks (ANNs) to convert aerial radiation survey data, constructed from experiences in Fukushima, into ambient dose rates at 1 m above ground level.Materials and Methods: The ANN utilized in this investigation was constructed based on the data collected after the Fukushima Daiichi Nuclear Power Plant accident. To validate the applicability of ANN, we applied it to the data used as input to ANN obtained from a detector that was different from the original detector.Results and Discussion: The measurement data from the different detectors were able to output the conversion results by ANN with equivalent accuracy as the original detectors using coefficients to convert the counting levels of the detectors.Conclusion: The results of this study suggest the potential for applying ANN generated from the accumulation of valuable monitoring data in Fukushima to other contexts.