A machine learning method for improving the accuracy of radiation biodosimetry by combining data from the dicentric chromosomes and micronucleus assays
Abstract A large-scale malicious or accidental radiological event can expose vast numbers of people to ionizing radiation. The dicentric chromosome (DCA) and cytokinesis-block micronucleus (CBMN) assays are well-established biodosimetry methods for estimating individual absorbed doses after radiatio...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-12-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-022-25453-2 |
| _version_ | 1811188987525595136 |
|---|---|
| author | Igor Shuryak Ekaterina Royba Mikhail Repin Helen C. Turner Guy Garty Naresh Deoli David J. Brenner |
| author_facet | Igor Shuryak Ekaterina Royba Mikhail Repin Helen C. Turner Guy Garty Naresh Deoli David J. Brenner |
| author_sort | Igor Shuryak |
| collection | DOAJ |
| description | Abstract A large-scale malicious or accidental radiological event can expose vast numbers of people to ionizing radiation. The dicentric chromosome (DCA) and cytokinesis-block micronucleus (CBMN) assays are well-established biodosimetry methods for estimating individual absorbed doses after radiation exposure. Here we used machine learning (ML) to test the hypothesis that combining automated DCA and CBMN assays will improve dose reconstruction accuracy, compared with using either cytogenetic assay alone. We analyzed 1349 blood sample aliquots from 155 donors of different ages (3–69 years) and sexes (49.1% males), ex vivo irradiated with 0–8 Gy at dose rates from 0.08 Gy/day to ≥ 600 Gy/s. We compared the performances of several state-of-the-art ensemble ML methods and found that random forest generated the best results, with R2 for actual vs. reconstructed doses on a testing data subset = 0.845, and mean absolute error = 0.628 Gy. The most important predictor variables were CBMN and DCA frequencies, and age. Removing CBMN or DCA data from the model significantly increased squared errors on testing data (p-values 3.4 × 10–8 and 1.1 × 10–6, respectively). These findings demonstrate the promising potential of combining CBMN and DCA assay data to reconstruct radiation doses in realistic scenarios of heterogeneous populations exposed to a mass-casualty radiological event. |
| first_indexed | 2024-04-11T14:28:27Z |
| format | Article |
| id | doaj.art-1a15af263c514e6d8d0c2477e6a4ca66 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-04-11T14:28:27Z |
| publishDate | 2022-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-1a15af263c514e6d8d0c2477e6a4ca662022-12-22T04:18:47ZengNature PortfolioScientific Reports2045-23222022-12-0112111310.1038/s41598-022-25453-2A machine learning method for improving the accuracy of radiation biodosimetry by combining data from the dicentric chromosomes and micronucleus assaysIgor Shuryak0Ekaterina Royba1Mikhail Repin2Helen C. Turner3Guy Garty4Naresh Deoli5David J. Brenner6Center for Radiological Research, Columbia University Irving Medical CenterCenter for Radiological Research, Columbia University Irving Medical CenterCenter for Radiological Research, Columbia University Irving Medical CenterCenter for Radiological Research, Columbia University Irving Medical CenterRadiological Research Accelerator Facility, Columbia University Irving Medical CenterRadiological Research Accelerator Facility, Columbia University Irving Medical CenterCenter for Radiological Research, Columbia University Irving Medical CenterAbstract A large-scale malicious or accidental radiological event can expose vast numbers of people to ionizing radiation. The dicentric chromosome (DCA) and cytokinesis-block micronucleus (CBMN) assays are well-established biodosimetry methods for estimating individual absorbed doses after radiation exposure. Here we used machine learning (ML) to test the hypothesis that combining automated DCA and CBMN assays will improve dose reconstruction accuracy, compared with using either cytogenetic assay alone. We analyzed 1349 blood sample aliquots from 155 donors of different ages (3–69 years) and sexes (49.1% males), ex vivo irradiated with 0–8 Gy at dose rates from 0.08 Gy/day to ≥ 600 Gy/s. We compared the performances of several state-of-the-art ensemble ML methods and found that random forest generated the best results, with R2 for actual vs. reconstructed doses on a testing data subset = 0.845, and mean absolute error = 0.628 Gy. The most important predictor variables were CBMN and DCA frequencies, and age. Removing CBMN or DCA data from the model significantly increased squared errors on testing data (p-values 3.4 × 10–8 and 1.1 × 10–6, respectively). These findings demonstrate the promising potential of combining CBMN and DCA assay data to reconstruct radiation doses in realistic scenarios of heterogeneous populations exposed to a mass-casualty radiological event.https://doi.org/10.1038/s41598-022-25453-2 |
| spellingShingle | Igor Shuryak Ekaterina Royba Mikhail Repin Helen C. Turner Guy Garty Naresh Deoli David J. Brenner A machine learning method for improving the accuracy of radiation biodosimetry by combining data from the dicentric chromosomes and micronucleus assays Scientific Reports |
| title | A machine learning method for improving the accuracy of radiation biodosimetry by combining data from the dicentric chromosomes and micronucleus assays |
| title_full | A machine learning method for improving the accuracy of radiation biodosimetry by combining data from the dicentric chromosomes and micronucleus assays |
| title_fullStr | A machine learning method for improving the accuracy of radiation biodosimetry by combining data from the dicentric chromosomes and micronucleus assays |
| title_full_unstemmed | A machine learning method for improving the accuracy of radiation biodosimetry by combining data from the dicentric chromosomes and micronucleus assays |
| title_short | A machine learning method for improving the accuracy of radiation biodosimetry by combining data from the dicentric chromosomes and micronucleus assays |
| title_sort | machine learning method for improving the accuracy of radiation biodosimetry by combining data from the dicentric chromosomes and micronucleus assays |
| url | https://doi.org/10.1038/s41598-022-25453-2 |
| work_keys_str_mv | AT igorshuryak amachinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT ekaterinaroyba amachinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT mikhailrepin amachinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT helencturner amachinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT guygarty amachinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT nareshdeoli amachinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT davidjbrenner amachinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT igorshuryak machinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT ekaterinaroyba machinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT mikhailrepin machinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT helencturner machinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT guygarty machinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT nareshdeoli machinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays AT davidjbrenner machinelearningmethodforimprovingtheaccuracyofradiationbiodosimetrybycombiningdatafromthedicentricchromosomesandmicronucleusassays |