Medizinische Universität Graz Austria/Österreich - Forschungsportal - Medical University of Graz
Gewählte Publikation:
Rodriguez Dominguez, K.
Clinical Acceptable Image Noise in Pediatric Computed Tomography Examinations—A Retrospective Analysis.
Studium für die Gleichwertigkeit; Humanmedizin; [ Diplomarbeit ] Graz Medical University; 2020. pp. 67
[OPEN ACCESS]
FullText
- Autor*innen der Med Uni Graz:
- Betreuer*innen:
- Nagy Eszter
- Sorantin Erich
- Altmetrics:
- Abstract:
- INTRODUCTION The use of plain radiographies and CTs comes with an unavoidable radiation exposure. CT examinations in pediatric patients represent one of the highest challenges to radiologists because of their higher sensitivity to radiation’s harmful side effects. This study aims to obtain noise measurements in diagnostic CT scans to define clinically acceptable noise. MATERIALS AND METHODS A database of 620 CT examinations was retrieved at the Radiology Department, Division of Pediatric Radiology at University Hospital Graz. Data was anonymized. Image noise, as well as other key technical parameters, was retrieved from the images. Image noise was measured in head, chest and abdomen CT scans. Results were analyzed descriptively as well as with statistical tests. Subjective image analysis was performed according to two grading scales and analyzed descriptively. RESULTS In native head CTs, image noise in centrum semiovale was significantly lower at 100 kV than at 80 kV and 120 kV tube voltages (p<0,05). Significantly higher image noise was seen in the aortic arch with 100 kV than with 120 kV. Enhanced head and chest examinations with 100 kV presented significantly higher image noise than the native ones. Measurements in native abdominal examinations with 100 kV showed significantly higher image noise (p=0,012) when compared to those with 120 kV. No significant difference was seen between native and CM examinations. Subjective evaluation of native head examinations revealed minimal image noise in 61% of the cases as well as optimal image quality in 92% of the cases. 64% of native chest examinations was scored with “moderate” image noise, the subjective image quality was “acceptable” in 88% of the cases was. 77% of the enhanced chest examinations were “acceptable”, 56% had “moderate” image noise. 67% of the enhanced abdominal examinations had “moderate” image noise, and 84% were “acceptable”. CONCLUSION Establishing clinical acceptable image noise ranges within different anatomical regions may contribute in the development of CT protocols to provide the best possible diagnostic information at the lowest possible dose.