Details

Autor(en) / Beteiligte

• Gulliksrud, Kristine
• Stokke, Caroline
• Trægde Martinsen, Anne Catrine

Titel

How to measure CT image quality: Variations in CT-numbers, uniformity and low contrast resolution for a CT quality assurance phantom

Ist Teil von

• Physica medica, 2014-06, Vol.30 (4), p.521-526

Ort / Verlag

Italy: Elsevier Ltd

Erscheinungsjahr

2014

Quelle

MEDLINE

Beschreibungen/Notizen

• Abstract Purpose Quality assurance (QA) phantoms for testing different image quality parameters in computed tomography (CT) are commercially available. Such phantoms are also used as reference for acceptance in the specifications of CT-scanners. The aim of this study was to analyze the characteristics of the most commonly used QA phantom in CT: Catphan 500/504/600. Methods Nine different phantoms were scanned on the same day, on one CT-scanner with the same parameter settings. Interphantom variations in CT-number values, image uniformity and low contrast resolution were evaluated for the phantoms. Comparisons between manual image analysis and results obtained from the automatic evaluation software QAlite were performed. Results Some interphantom variations were observed in the low contrast resolution and the CT-number modules of the phantoms. Depending on the chosen regulatory framework, the variations in CT-numbers can be interpreted as substantial. The homogenous modules were found more invariable. However, the automatic image analysis software QAlite measures image uniformity differently than recommended in international standards, and will not necessarily give results in agreement with these standards. Conclusions It is important to consider the interphantom variations in relation to ones framework, and to be aware of which phantom is used to study CT-numbers and low contrast resolution for a specific scanner. Comparisons with predicted values from manual and acceptance values should be performed with care and consideration. If automatic software-based evaluations are to be used, users should be aware that large differences can exist for the image uniformity testing.