Potential of a machine-learning model for dose optimization in CT quality assurance

• Published in: European radiology 2019-07, Vol.29 (7), p.3705-3713
• Main Authors: Meineke, Axel, Rubbert, Christian, Sawicki, Lino M., Thomas, Christoph, Klosterkemper, Yan, Appel, Elisabeth, Caspers, Julian, Bethge, Oliver T., Kröpil, Patric, Antoch, Gerald, Boos, Johannes
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Artificial intelligence; Chest; Computed tomography; Cross-Sectional Studies; Cut-off; Diagnostic Radiology; Error analysis; Female; Humans; Imaging; Internal Medicine; Interventional Radiology; Learning algorithms; Machine Learning; Male; Mathematical models; Medicine; Medicine & Public Health; Middle Aged; Multidetector Computed Tomography - standards; Neural networks; Neuroradiology; Optimization; Performance measurement; Quality assurance; Quality Assurance, Health Care; Quality control; Radiation; Radiation Dosage; Radiation Injuries - prevention & control; Radiography, Thoracic - standards; Radiology; Retrospective Studies; Root-mean-square errors; Scanners; Thoracic Diseases - diagnosis; Training; Ultrasound; Young Adult
• ISSN: 0938-7994; 1432-1084
• DOI: 10.1007/s00330-019-6013-6

Objectives To evaluate machine learning (ML) to detect chest CT examinations with dose optimization potential for quality assurance in a retrospective, cross-sectional study. Methods Three thousand one hundred ninety-nine CT chest examinations were used for training and testing of the feed-forward, single hidden layer neural network (January 2016–December 2017, 60% male, 62 ± 15 years, 80/20 split). The model was optimized and trained to predict the volumetric computed tomography dose index (CTDI vol ) based on scan patient metrics (scanner, study description, protocol, patient age, sex, and water-equivalent diameter ( D W )). The root mean-squared error (RMSE) was calculated as performance measurement. One hundred separate, consecutive chest CTs were used for validation (January 2018, 60% male, 63 ± 16 years), independently reviewed by two blinded radiologists with regard to dose optimization, and used to define an optimal cutoff for the model. Results RMSE was 1.71, 1.45, and 1.52 for the training, test, and validation dataset, respectively. The scanner and D W were the most important features. The radiologists found dose optimization potential in 7/100 of the validation cases. A percentage deviation of 18.3% between predicted and actual CTDI vol was found to be the optimal cutoff: 8/100 cases were flagged as suboptimal by the model (range 18.3–53.2%). All of the cases found by the radiologists were identified. One examination was flagged only by the model. Conclusions ML can comprehensively detect CT examinations with dose optimization potential. It may be a helpful tool to simplify CT quality assurance. CT scanner and D W were most important. Final human review remains necessary. A threshold of 18.3% between the predicted and actual CTDI vol seems adequate for CT quality assurance. Key Points • Machine learning can be integrated into CT quality assurance to improve retrospective analysis of CT dose data. • Machine learning may help to comprehensively detect dose optimization potential in chest CT, but an individual review of the results by an experienced radiologist or radiation physicist is required to exclude false-positive findings.