Suboptimal Chest Radiography and Artificial Intelligence: The Problem and the Solution

Chest radiographs (CXR) are the most performed imaging tests and rank high among the radiographic exams with suboptimal quality and high rejection rates. Suboptimal CXRs can cause delays in patient care and pitfalls in radiographic interpretation, given their ubiquitous use in the diagnosis and management of acute and chronic ailments. Suboptimal CXRs can also compound and lead to high inter-radiologist variations in CXR interpretation. While advances in radiography with transitions to computerized and digital radiography have reduced the prevalence of suboptimal exams, the problem persists. Advances in machine learning and artificial intelligence (AI), particularly in the radiographic acquisition, triage, and interpretation of CXRs, could offer a plausible solution for suboptimal CXRs. We review the literature on suboptimal CXRs and the potential use of AI to help reduce the prevalence of suboptimal CXRs.

Digital radiography reject analysis of examinations with multiple rejects: an Australian emergency imaging department clinical audit

INTRODUCTION

The largest source of manmade ionising radiation exposure to the public stems from diagnostic medical imaging examinations. Reject analysis, a form of quality assurance, was introduced to minimise repeat exposures. The purpose of this study was to analyse projection-specific reject rates and radiographic examinations with multiple rejects.

METHODS

A retrospective audit of rejected radiographs was undertaken in a busy Australian metropolitan emergency digital X-ray room from March to June 2018. The data were collected by reject analysis software embedded within the X-ray unit. Reject rates, and reasons for rejection for each X-ray projection were analysed.

RESULTS

Data from 11, 596 images showed overall reject rate was 10.3% and the overall multiple reject rate was 1.3%. The projections with both a high number and high percentage of rejects were antero-posterior (AP) chest (175, 18.1%), AP pelvis (78, 22.5%), horizontal beam hip (61, 33.5%) and horizontal beam knee (116, 30.5%). The projections with both a high frequency and multiple reject rate were horizontal beam knee (32, 8.4%) and horizontal beam hip (17, 9.3%). The top reasons for multiple rejects were positioning (67.1%) and anatomy cut-off (8.4%).

CONCLUSIONS

The findings of this study demonstrated that projection-specific reject and multiple reject analysis in digital radiography is necessary in identifying areas for quality improvement which will reduce radiation exposure to patients. Projections that were frequently repeated in this study were horizontal beam knee and horizontal beam hip. Future research could involve re-auditing the department following the implementation of improvement strategies to reduce unnecessary radiation exposure.