Abstract
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 (CTDIvol) based on scan patient metrics (scanner, study description, protocol, patient age, sex, and water-equivalent diameter (DW)). 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 DW 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 CTDIvol 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 DW were most important. Final human review remains necessary. A threshold of 18.3% between the predicted and actual CTDIvol 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.
Similar content being viewed by others
Abbreviations
- CTDIvol :
-
Volumetric computed tomography dose index
- DLP:
-
Dose length product
- DRLs:
-
Diagnostic reference levels
- D W :
-
Water-equivalent diameter
- ML:
-
Machine learning
- QA:
-
Quality assurance
- RMSE:
-
Root mean-squared error
References
Butler PF, Kanal KM (2018) Diagnostic reference levels for adult patients in the United States. J Am Coll Radiol 15:932–933. https://doi.org/10.1016/j.jacr.2017.12.012
Alhailiy AB, Ekpo EU, Ryan EA, Kench PL, Brennan PC, McEntee MF (2018) Diagnostic reference levels for cardiac CT angiography in Australia. Radiat Prot Dosimetry. https://doi.org/10.1093/rpd/ncy112
Appel E, Kröpil P, Bethge OT et al (2018) Quality assurance in CT: implementation of the updated national diagnostic reference levels using an automated CT dose monitoring system. Clin Radiol 73:677.e13–677.e20. https://doi.org/10.1016/j.crad.2018.02.012
Roch P, Célier D, Dessaud C, Etard C (2018) Using diagnostic reference levels to evaluate the improvement of patient dose optimisation and the influence of recent technologies in radiography and computed tomography. Eur J Radiol 98:68–74. https://doi.org/10.1016/j.ejrad.2017.11.002
Klosterkemper Y, Appel E, Thomas C et al (2018) Tailoring CT dose to patient size: implementation of the updated 2017 ACR size-specific diagnostic reference levels. Acad Radiol 25:1624–1631. https://doi.org/10.1016/j.acra.2018.03.005
Boere H, Eijsvoogel NG, Sailer AM et al (2018) Implementation of size-dependent local diagnostic reference levels for CT angiography. AJR Am J Roentgenol 210:W226–W233. https://doi.org/10.2214/AJR.17.18566
MacGregor K, Li I, Dowdell T, Gray BG (2015) Identifying institutional diagnostic reference levels for CT with radiation dose index monitoring software. Radiology 276:507–517. https://doi.org/10.1148/radiol.2015141520
Smith-Bindman R, Wang Y, Yellen-Nelson TR et al (2016) Predictors of CT radiation dose and their effect on patient care: a comprehensive analysis using automated data. Radiology 282:182–193. https://doi.org/10.1148/radiol.2016151391
Demb J, Chu P, Nelson T et al (2017) Optimizing radiation doses for computed tomography across institutions: dose auditing and best practices. JAMA Intern Med 177:810–817. https://doi.org/10.1001/jamainternmed.2017.0445
Boos J, Thomas C, Appel E et al (2018) Institutional computed tomography diagnostic reference levels based on water-equivalent diameter and size-specific dose estimates. J Radiol Prot 38:536–548. https://doi.org/10.1088/1361-6498/aaa32c
Rubbert C, Patil KR, Beseoglu K et al (2018) Prediction of outcome after aneurysmal subarachnoid haemorrhage using data from patient admission. Eur Radiol 28:4949–4958. https://doi.org/10.1007/s00330-018-5505-0
Kim GB, Jung KH, Lee Y et al (2018) Comparison of shallow and deep learning methods on classifying the regional pattern of diffuse lung disease. J Digit Imaging 31:415–424. https://doi.org/10.1007/s10278-017-0028-9
Dreyer KJ, Geis JR (2017) When machines think: radiology’s next frontier. Radiology 285:713–718. https://doi.org/10.1148/radiol.2017171183
Nishio M, Sugiyama O, Yakami M et al (2018) Computer-aided diagnosis of lung nodule classification between benign nodule, primary lung cancer, and metastatic lung cancer at different image size using deep convolutional neural network with transfer learning. PLoS One 13:e0200721. https://doi.org/10.1371/journal.pone.0200721
Chen B, Xiang K, Gong Z, Wang J, Tan S (2018) Statistical iterative CBCT reconstruction based on neural network. IEEE Trans Med Imaging 37:1511–1521. https://doi.org/10.1109/TMI.2018.2829896
Anam C, Haryanto F, Widita R, Arif I, Dougherty G (2016) Automated calculation of water-equivalent diameter (DW) based on AAPM task group 220. J Appl Clin Med Phys 17:6171
Boos J, Kröpil P, Bethge OT et al (2018) Accuracy of size-specific dose estimate calculation from center slice in computed tomography. Radiat Prot Dosimetry 178(1):8–19
Bergstra J, Bengio Y (2012) Random search for hyper-parameter optimization. J Mach Learn Res 13:281–305
Gevrey M, Dimopoulos I, Lek S (2003) Review and comparison of methods to study the contribution of variables in artificial neural network models. Ecol Modell 160:249–264. https://doi.org/10.1016/S0304-3800(02)00257-0
BfS - diagnostic reference levels. http://www.bfs.de/EN/topics/ion/medicine/diagnostics/reference-levels/reference-levels_node.html;jsessionid=F14BA4ABEE11D08B75B07F8720579B3C.1_cid339. Accessed 16 Aug 2018
Feinstein S (1975) The accuracy of diver sound localization by pointing. Undersea Biomed Res 2:173–184
Zhang X, Yuan Z, Ji J, Li H, Xue F (2016) Network or regression-based methods for disease discrimination: a comparison study. BMC Med Res Methodol 16:100. https://doi.org/10.1186/s12874-016-0207-2
Kadir T, Gleeson F (2018) Lung cancer prediction using machine learning and advanced imaging techniques. Transl Lung Cancer Res 7:304–312. https://doi.org/10.21037/tlcr.2018.05.15
Saha A, Harowicz MR, Grimm LJ et al (2018) A machine learning approach to radiogenomics of breast cancer: a study of 922 subjects and 529 DCE-MRI features. Br J Cancer. https://doi.org/10.1038/s41416-018-0185-8
Moradi E, Pepe A, Gaser C, Huttunen H, Tohka J (2015) Machine learning framework for early MRI-based Alzheimer’s conversion prediction in MCI subjects. Neuroimage 104:398–412. https://doi.org/10.1016/j.neuroimage.2014.10.002
Coenen A, Kim YH, Kruk M et al (2018) Diagnostic accuracy of a machine-learning approach to coronary computed tomographic angiography-based fractional flow reserve: result from the MACHINE Consortium. Circ Cardiovasc Imaging 11:e007217. https://doi.org/10.1161/CIRCIMAGING.117.007217
Lakhani P, Prater AB, Hutson RK et al (2018) Machine learning in radiology: applications beyond image interpretation. J Am Coll Radiol 15:350–359. https://doi.org/10.1016/j.jacr.2017.09.044
Küstner T, Gatidis S, Liebgott A et al (2018) A machine-learning framework for automatic reference-free quality assessment in MRI. Magn Reson Imaging. https://doi.org/10.1016/j.mri.2018.07.003
Parakh A, Kortesniemi M, Schindera ST (2016) CT radiation dose management: a comprehensive optimization process for improving patient safety. Radiology 280:663–673. https://doi.org/10.1148/radiol.2016151173
Guberina N, Dietrich U, Radbruch A et al (2018) Detection of early infarction signs with machine learning-based diagnosis by means of the Alberta Stroke Program Early CT Score (ASPECTS) in the clinical routine. Neuroradiology. https://doi.org/10.1007/s00234-018-2066-5
Brink JA, Miller DL (2015) U.S. national diagnostic reference levels: closing the gap. Radiology 277:3–6. https://doi.org/10.1148/radiol.2015150971
Bataineh M, Marler T (2017) Neural network for regression problems with reduced training sets. Neural Netw 95:1–9. https://doi.org/10.1016/j.neunet.2017.07.018
Funding
The authors state that this work has not received any funding.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Guarantor
The scientific guarantor of this publication is Johannes Boos.
Conflict of interest
The authors of this manuscript declare relationships with the following companies: AM—Cerner HS Deutschland GmbH (employee) and Pulmokard GmbH (consultant).
Statistics and biometry
One of the authors has significant statistical expertise.
Informed consent
Written informed consent was waived by the Institutional Review Board.
Ethical approval
Institutional Review Board approval was obtained.
Methodology
• retrospective
• cross-sectional study
• performed at one institution
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Meineke, A., Rubbert, C., Sawicki, L.M. et al. Potential of a machine-learning model for dose optimization in CT quality assurance. Eur Radiol 29, 3705–3713 (2019). https://doi.org/10.1007/s00330-019-6013-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00330-019-6013-6