Computed radiography dose data mining and surveillance as an ongoing quality assurance improvement process

Area of Science:

• Medical Imaging
• Radiologic Technology
• Health Informatics

Background:

• Computed radiography (CR) utilizes sensitivity numbers (S-numbers) to indicate exposure levels.
• Quality assurance (QA) in medical imaging is crucial for optimizing diagnostic quality and patient safety.
• Picture Archiving and Communication Systems (PACS) store and manage medical images and associated data.

Purpose of the Study:

• To implement a data-mining program for extracting CR S-number information from PACS.
• To establish an ongoing QA improvement project for monitoring and optimizing radiographic techniques.
• To utilize S-number trends for adjusting technique charts and achieving departmental S-number goals.

Main Methods:

• A data-mining program was developed to extract CR S-number data from PACS monthly.
• Extracted S-number data were compared against previous month's data and established departmental goals.
• Findings were presented at monthly QA meetings to inform technologists.
• Technologists used S-number trends to modify radiographic technique charts.

Main Results:

• The data-mining program successfully extracted and analyzed CR S-number information.
• S-number trends provided actionable insights for radiographic technique adjustments.
• The process facilitated a cyclic QA improvement loop.

Conclusions:

• Mining PACS data is an effective strategy for ongoing QA in medical imaging.
• This approach can lead to significant reductions in patient radiation dose.
• The QA process helps minimize interexamination dose variance, improving consistency.