Early experience with automated B-mode quality assurance tests

Area of Science:

• Medical Imaging
• Quality Assurance
• Diagnostic Ultrasound

Background:

• Quality assurance (QA) programs are crucial for medical imaging but lack robust evidence of efficacy.
• Automated image analysis offers a potential solution for objective and consistent QA.

Purpose of the Study:

• To assess the efficacy of an automated image quality analysis method in a QA program.
• To determine if automated analysis can detect changes in scanner performance.

Main Methods:

• Analysis of test object images measuring lateral resolution, low contrast penetration, slice thickness, and grey-scale contrast/visibility.
• Investigation of known/suspected scanner faults and review of routine QA results.

Main Results:

• Automated analysis detected changes in performance variables corresponding to known/suspected scanner faults.
• Image shadowing affected resolution and grey-scale visibility; user-reported quality deterioration correlated with changes in slice thickness, lateral resolution, and grey-scale contrast.
• A probe failure was identified due to an unrecorded change in default settings.

Conclusions:

• The automated image quality analysis method provides evidence of efficacy for scanner performance testing.
• Further research is needed to optimize the method for prospective fault detection and reduce confounding factors.