THE EFFECT OF BACKGROUND SIGNAL AND ITS REPRESENTATION IN DECONVOLUTION OF EPR SPECTRA ON ACCURACY OF EPR DOSIMETRY IN BONE
Autor: | Agnieszka Marciniak, Mateusz Cieszyński, Piotr Boguś, Bartlomiej Ciesielski, Agnieszka Zientek, Karolina Krefft, Anita Prawdzik-Dampc |
---|---|
Rok vydání: | 2016 |
Předmět: |
Materials science
Radiation Dosage Sensitivity and Specificity Signal Spectral line 030218 nuclear medicine & medical imaging law.invention 03 medical and health sciences 0302 clinical medicine Nuclear magnetic resonance law Calibration Background Radiation Humans Radiology Nuclear Medicine and imaging Irradiation Radiometry Electron paramagnetic resonance Radiation Radiological and Ultrasound Technology Electron Spin Resonance Spectroscopy Public Health Environmental and Occupational Health Reproducibility of Results General Medicine Radiation Exposure Trabecular bone 030220 oncology & carcinogenesis Epr dosimetry Biological Assay Deconvolution Artifacts Tooth Algorithms |
Zdroj: | Radiation Protection Dosimetry. 172:275-282 |
ISSN: | 1742-3406 0144-8420 |
Popis: | This study is about the accuracy of EPR dosimetry in bones based on deconvolution of the experimental spectra into the background (BG) and the radiation-induced signal (RIS) components. The model RIS's were represented by EPR spectra from irradiated enamel or bone powder; the model BG signals by EPR spectra of unirradiated bone samples or by simulated spectra. Samples of compact and trabecular bones were irradiated in the 30-270 Gy range and the intensities of their RIS's were calculated using various combinations of those benchmark spectra. The relationships between the dose and the RIS were linear (R2 > 0.995), with practically no difference between results obtained when using signals from irradiated enamel or bone as the model RIS. Use of different experimental spectra for the model BG resulted in variations in intercepts of the dose-RIS calibration lines, leading to systematic errors in reconstructed doses, in particular for high- BG samples of trabecular bone. These errors were reduced when simulated spectra instead of the experimental ones were used as the benchmark BG signal in the applied deconvolution procedures. |
Databáze: | OpenAIRE |
Externí odkaz: |