Radiation dose estimation with multiple artificial neural networks in dicentric chromosome assay

Abstract

Purpose

The dicentric chromosome assay (DCA), often referred to as the ‘gold standard’ in radiation dose estimation, exhibits significant challenges as a consequence of its labor-intensive nature and dependency on expert knowledge. Existing automated technologies face limitations in accurately identifying dicentric chromosomes (DCs), resulting in decreased precision for radiation dose estimation. Furthermore, in the process of identifying DCs through automatic or semi-automatic methods, the resulting distribution could demonstrate under-dispersion or over-dispersion, which results in significant deviations from the Poisson distribution. In response to these issues, we developed an algorithm that employs deep learning to automatically identify chromosomes and perform fully automatic and accurate estimation of diverse radiation doses, adhering to a Poisson distribution.

Materials and methods

The dataset utilized for the dose estimation algorithm was generated from 30 healthy donors, with samples created across seven doses, ranging from 0 to 4 Gy. The procedure encompasses several steps: extracting images for dose estimation, counting chromosomes, and detecting DC and fragments. To accomplish these tasks, we utilize a diverse array of artificial neural networks (ANNs). The identification of DCs was accomplished using a detection mechanism that integrates both deep learning-based object detection and classification methods. Based on these detection results, dose-response curves were constructed. A dose estimation was carried out by combining a regression-based ANN with the Monte-Carlo method.

Results

In the process of extracting images for dose analysis and identifying DCs, an under-dispersion tendency was observed. To rectify the discrepancy, classification ANN was employed to identify the results of DC detection. This approach led to satisfaction of Poisson distribution criteria by 32 out of the initial pool of 35 data points. In the subsequent stage, dose-response curves were constructed using data from 25 donors. Data provided by the remaining five donors served in performing dose estimations, which were subsequently calibrated by incorporating a regression-based ANN. Of the 23 points, 22 fell within their respective confidence intervals at p < .05 (95%), except for those associated with doses at levels below 0.5 Gy, where accurate calculation was obstructed by numerical issues. The accuracy of dose estimation has been improved for all radiation levels, with the exception of 1 Gy.

Conclusions

This study successfully demonstrates a high-precision dose estimation method across a general range up to 4 Gy through fully automated detection of DCs, adhering strictly to Poisson distribution. Incorporating multiple ANNs confirms the ability to perform fully automated radiation dose estimation. This approach is particularly advantageous in scenarios such as large-scale radiological incidents, improving operational efficiency and speeding up procedures while maintaining consistency in assessments. Moreover, it reduces potential human error and enhances the reliability of results.

Keywords:

• Radiation
• dose estimation
• dicentric chromosome
• Poisson distribution
• deep learning
• Monte-Carlo method

Disclosure statement

The authors declare that they have no conflict of interest.

Additional information

Funding

This work was supported by a grant from the Korean Government (MSIT) to the Dongnam Institute of Radiological and Medical Sciences (DIRAMS), with grant numbers 50491-2023 and Ministry of Science and ICT, South Korea.

Notes on contributors

Seungsoo Jang

Seungsoo Jang is at the Department of Advanced Nuclear Engineering, Graduate Student, Pohang University of Science and Technology (POSTECH), Pohang, Korea.

Janghee Lee

Janghee Lee is at the Department of Advanced Nuclear Engineering, Graduate Student, Pohang University of Science and Technology (POSTECH), Pohang, Korea.

Song-Hyun Kim

Song-Hyun Kim is at the SierraBaSe Co. Ltd, CEO, Sejong, Korea.

Sangsoo Han

Sangsoo Han is at the SierraBaSe Co. Ltd, COO, Sejong, Korea.

Sung-Gyun Shin

Sung-Gyun Shin is at the SierraBaSe Co. Ltd, CTO, Sejong, Korea.

Sunghee Lee

Sunghee Lee is at the SierraBaSe Co. Ltd, Director of Research, Sejong, Korea.

Inhyuk Kang

Inhyuk Kang is at the SierraBaSe Co. Ltd, a Senior Researcher, Sejong, Korea.

Wol Soon Jo

Dr. Wol Soon Jo is a corresponding author of this paper and has been establishing a biodosimetry system in DIRAMS since 2019. She has various experiences such as cancer immunology, genetic toxicology, microalgae and radiation biology. She published more than 30 papers as the first author or corresponding author related to Radiation biology.

Sookyung Jeong

Sookyung Jeong, with a PhD and 10 years in biodosimetry, is developing a dicentric chromosome assay using deep learning. She has authored and co-authored two papers in this field.

Su Jung Oh

Su Jung Oh is a biodosimetry expert and has trained for 10 years. She is the executive secretary of biodosimetry network involving four institutions in South Korea and participates in cross-analysis tests using dicentric assay and translocation assay organized by the network.

Chang Geun Lee

Dr. Chang Geun Lee is working at the Research Center of Dongnam Institute of Radiological & Medical Sciences (DIRAMS). He received his Ph.D. degree from POSTECH (ROK) majoring in development of immunotherapy. He is working on the effect of low-dose radiation and the roles of radiation dose rate.