Related Experiment Videos
Evaluation of attenuation corrections using Monte Carlo simulated lung SPECT
A Gustafsson1, B Bake, L Jacobsson
1Department of Radiation Physics, University of Göteborg, Sahlgrenska University Hospital, Sweden. agnetha.gustafsson@radfys.gu.se
Physics in Medicine and Biology
|September 2, 1998
Summary
Attenuation correction significantly improves single photon emission computed tomography (SPECT) lung imaging homogeneity. A post-processing method using true thoracic densities yielded the best results, highlighting the importance of accurate lung contour definition.
Area of Science:
- Medical Imaging
- Nuclear Medicine
- Radiological Physics
Background:
- Photon attenuation in SPECT imaging causes image distortions, particularly in the thoracic region due to varying tissue densities.
- Accurate attenuation correction is crucial for quantitative analysis in lung SPECT studies.
Purpose of the Study:
- To compare the efficacy of pre-processing and post-processing attenuation correction methods in lung SPECT.
- To investigate the influence of specific attenuation correction parameters on image homogeneity.
Main Methods:
- Simulated SPECT studies using a digital thorax phantom with homogeneous lung activity distribution.
- Monte Carlo technique for simulating photon transport and attenuation.
- Image homogeneity quantified using the coefficient of variation (CV); isolated correction effects assessed via pixel value normalization.
Main Results:
- The post-processing attenuation correction method, utilizing true thoracic densities, reduced the CV from 12.8% to 4.4%.
- Variations in body contour definition had a minimal impact on image homogeneity.
- Lung contour definition significantly influenced the effectiveness of attenuation correction.
Conclusions:
- Post-processing attenuation correction with accurate thoracic densities substantially enhances lung SPECT image homogeneity.
- Precise definition of the lung contour is critical for optimal attenuation correction outcomes in SPECT imaging.