Purpose or learning objective
Chest radiography is the most commonly performed diagnostic radiographic examination [1]. Image quality assessment plays a vital role in improving image quality and diagnostic accuracy[2]. Quality assessment (QA) for chest radiographs is still a manual process, which leads to poor interobserver agreement and limits the number of images included in the evaluation[3], further hindering the analysis of the overall image quality in the health care institution.
Among the factors affecting the quality of chest radiography, layout and position errors are common image acquisition defects that...
Methods or background
Outline
This study consisted of three stages: a preassessment stage, where ten quantitative indices were created to capture the quality of the layout and position of chest radiographs (Table 1).[Fig 7] And a “study dataset” with the various image qualities of chest radiographs collected based on these indices. A training stage of developing the automatic assessment system consisted of a multiple linear regression model and an automatic measurement DL model. There was also a test stage to evaluate the performance of the system by using...
Results or findings
The stepwise regression showed a statistically significant relationship between 10 quantitative indice parameters and subjective scores (P<0.05) (Fig 4). [Fig 4] On the testing set, the PCK at the 4 mm distance threshold was 94%, 96%, 98% for three regions: lung, clavicle, and spinous process of thoracic vertebrae. The Dice values for left-right lung segmentation was 0.95 and 0.94. whereas the left-right scapula segmentation was 0.89 and 0.88, respectively (Fig 5).[Fig 5] The deep learning model showed high accuracy in predicting the quantitative indices (ICC=0.82-0.99,...
Conclusion
Ten quantitative indices correlated well with the subjective perceptions of radiologists on image layout and the positions of chest radiographs. The developed automatic system showed a high performance in measuring quantitative indices and assessing image quality.
References
Mettler FA, Jr., Mahesh M, Bhargavan-Chatfield M et al (2020) Patient Exposure from Radiologic and Nuclear Medicine Procedures in the United States: Procedure Volume and Effective Dose for the Period 2006-2016. Radiology 295:418-427
Tesselaar E, Dahlström N, Sandborg M (2016) Clinical Audit of Image Quality in Radiology Using Visual Grading Characteristics Analysis. Radiat Prot Dosimetry 169:340-346
Whaley JS, Pressman BD, Wilson JR, Bravo L, Sehnert WJ, Foos DH (2013) Investigation of the variability in the assessment of digital chest X-ray image quality. J Digit Imaging...
Personal information and conflict of interest
Y. Meng:
Nothing to disclose
J. Ruan:
Nothing to disclose
X. Y. Gong:
Nothing to disclose