Repeat Analysis Program As A Quality Assurance System For Radiology Management: Causal Repeat and Challenges

Abstract

Rejected or repeated images analysis remains a significant challenge, particularly in digital imaging. Despite the expectation that the transition from conventional to digital systems would reduce repetition rates, the reality is that repetition rates still exceed established standards. This literature review aims to shed light on the identification of causes and barriers in the reject/repeat program. We conducted a systematic review of this program in radiography units over several decades, examining the causes of repetition, types of examinations, and data sources used. We also described the methods employed to analyze reject/repeat instances in both conventional and digital systems. The study found that computed or digital radiography was the primary data source for image analysis. Despite the use of digital systems, repetition rates persisted, with chest radiography being the most significant contributor, accounting for over 30% of cases. Technical factors, particularly positioning errors, contributed to more than 30% of repetitions. Notably, determining the causes of rejection proved subjective. However, one study highlighted that artificial intelligence (AI) could accurately predict image rejection with a sensitivity of 93%. Thus, the incorporation of AI can greatly assist in classifying rejection causes, resulting in more efficient and streamlined radiology management

Keywords

• reject repeat analysis program
• radiology
• management
• ai