Abstract
To develop a convolutional neural network-based method for the subjective evaluation of computed tomography (CT) images having low-contrast resolution due to imaging conditions and nonlinear image processing. Four radiological technologists visually evaluated CT images that were reconstructed using three nonlinear noise reduction processes (AIDR 3D, AIDR 3D Enhanced, AiCE) on a CT system manufactured by CANON. The visual evaluation consisted of two items: low contrast detectability (score: 0–9) and texture pattern (score: 1–5). Four AI models with different convolutional and max pooling layers were constructed and trained on pairs of CANON CT images and average visual assessment scores of four radiological technologists. CANON CT images not used for training were used to evaluate prediction performance. In addition, CT images scanned with a SIEMENS CT system were input to each AI model for external validation. The mean absolute error and correlation coefficients were used as evaluation metrics. Our proposed AI model can evaluate low-contrast detectability and texture patterns with high accuracy, which varies with the dose administered and the nonlinear noise reduction process. The proposed AI model is also expected to be suitable for upcoming reconstruction algorithms that will be released in the future.
Similar content being viewed by others
Data availability
The datasets during and/or analyzed during the current study available from the corresponding author on reasonable request.
Code availability
The codes we write during the current study available from the corresponding author on reasonable request.
References
Singh S, Kalra MK, Hsieh J, Licato PE, Do S, Pien HH, Blake MA (2010) Abdominal CT: comparison of adaptive statistical iterative and filtered back projection reconstruction techniques. Radiology 257(2):373–383. https://doi.org/10.1148/radiol.10092212
Löve A, Olsson ML, Siemund R, Stålhammar F, Björkman-Burtscher IM, Söderberg M (2013) Six iterative reconstruction algorithms in brain CT: a phantom study on image quality at different radiation dose levels. Br J Radiol 86(1031):388. https://doi.org/10.1259/bjr.20130388
Willemink MJ, de Jong PA, Leiner T, de Heer LM, Nievelstein RA, Budde RP, Schilham AM (2013) Iterative reconstruction techniques for computed tomography. Part 1: technical principles. Eur Radiol 23(6):1623–1631. https://doi.org/10.1007/s00330-012-2765-y
Richard S, Husarik DB, Yadava G, Murphy SN, Samei E (2012) Towards task-based assessment of CT performance: system and object MTF across different reconstruction algorithms. Med Phys 39(7):4115–4122. https://doi.org/10.1118/1.4725171
Schindera ST, Odedra D, Raza SA, Kim TK, Jang HJ, Szucs-Farkas Z, Rogalla P (2013) Iterative reconstruction algorithm for CT: can radiation dose be decreased while low-contrast detectability is preserved? Radiology 269(2):511–518. https://doi.org/10.1148/radiol.13122349
Schindera ST, Odedra D, Mercer D, Thipphavong S, Chou P, Szucs-Farkas Z, Rogalla P (2014) Hybrid iterative reconstruction technique for abdominal CT protocols in obese patients: assessment of image quality, radiation dose, and low-contrast detectability in a phantom. Am J Roentgenol 202(2):W146-152. https://doi.org/10.2214/AJR.12.10513
Baker ME, Dong F, Primak A, Obuchowski NA, Einstein D, Gandhi N, Herts BR, Purysko A, Remer E, Vachhani N (2012) Contrast-to-noise ratio and low-contrast object resolution on full- and low-dose MDCT: SAFIRE versus filtered back projection in a low-contrast object phantom and in the liver. AJR 199(1):8–18. https://doi.org/10.2214/AJR.11.7421
Kondo M, Hatakenaka M, Higuchi K, Fujioka T, Shirasaka T, Nakamura Y, Nakamura K, Yoshiura T, Honda H (2013) Feasibility of low-radiation-dose CT for abdominal examinations with hybrid iterative reconstruction algorithm: low-contrast phantom study. Radiol Phys Technol 6(2):287–292. https://doi.org/10.1007/s12194-012-0197-7
Yang WJ, Yan FH, Liu B, Pang LF, Hou L, Zhang H, Pan ZL, Chen KM (2013) Can sinogram-affirmed iterative (SAFIRE) reconstruction improve imaging quality on low-dose lung CT screening compared with traditional filtered back projection (FBP) reconstruction? J Comput Assist Tomogr 37(2):301–305. https://doi.org/10.1097/RCT.0b013e31827b8c66
Mileto A, Zamora DA, Alessio AM, Pereira C, Liu J, Bhargava P, Carnell J, Cowan SM, Dighe MK, Gunn ML, Kim S, Kolokythas O, Lee JH, Maki JH, Moshiri M, Nasrullah A, O’Malley RB, Schmiedl UP, Soloff EV, Toia GV, Wang CL, Kanal KM (2018) CT Detectability of small low-contrast hypoattenuating focal lesions: iterative reconstructions versus filtered back projection. Radiology 289(2):443–454. https://doi.org/10.1148/radiol.2018180137
LeCun Y, Bengio Y, Hinton GE (2015) Deep learning. Nature 512:436–444
Matsubara N, Teramoto A, Saito K, Fujita H (2020) Bone suppression for chest X-ray image using a convolutional neural filter. Phys Eng Sci Med 43:97–108. https://doi.org/10.1007/s13246-019-00822-w
Kim J and Lee S (2017) Deep learning of human visual sensitivity in image quality assessment framework. In: 2017 IEEE conference on computer vision and pattern recognition (CVPR), Honolulu, pp 1969–1977. https://doi.org/10.1109/CVPR.2017.213
Hou W, Gao X, Tao D, Li X (2015) Blind image quality assessment via deep learning. IEEE Trans Neural Netw Learn Syst 26(6):1275–1286. https://doi.org/10.1109/TNNLS.2014.2336852
Lauermann JL, Treder M, Alnawaiseh M, Clemens CR, Eter N, Alten F (2019) Automated OCT angiography image quality assessment using a deep learning algorithm. Graefes Arch Clin Exp Ophthalmol 257(8):1641–1648. https://doi.org/10.1007/s00417-019-04338-7
Sujit SJ, Coronado I, Kamali A, Narayana PA, Gabr RE (2019) Automated image quality evaluation of structural brain MRI using an ensemble of deep learning networks. J Magn Reson Imaging 50(4):1260–1267. https://doi.org/10.1002/jmri.26693
Kim HY, Lee K, Chang W, Kim Y, Lee S, Oh DY, Sunwoo L, Lee YJ, Kim YH (2021) Robustness of deep learning algorithm to varying imaging conditions in detecting low contrast objects in computed tomography phantom images: in comparison to 12 radiologists. Diagnostics (Basel, Switzerland) 11(3):410. https://doi.org/10.3390/diagnostics11030410
Singh S, Kalra MK, Gilman MD, Hsieh J, Pien HH, Digumarthy SR, Shepard JA (2011) Adaptive statistical iterative reconstruction technique for radiation dose reduction in chest CT: a pilot study. Radiology 259(2):565–573. https://doi.org/10.1148/radiol.11101450
Abadi M, Barham P, Chen J et al (2016) Tensorflow: a system for large-scale machine learning. OSDI 16:265–283
Chollet F et al (2018) Keras: the python deep learning library. Astrophys Source Code Libr.
Samei E, Bakalyar D, Boedeker KL, Brady S, Fan J, Leng S, Myers KJ, Popescu LM, Ramirez Giraldo JC, Ranallo F, Solomon J, Vaishnav J, Wang J (2019) Performance evaluation of computed tomography systems: summary of AAPM Task Group 233. Med Phys 46(11):735–756. https://doi.org/10.1002/mp.13763
Rizzo S, Kalra M, Schmidt B, Dalal T, Suess C, Flohr T, Blake M, Saini S (2006) Comparison of angular and combined automatic tube current modulation techniques with constant tube current CT of the abdomen and pelvis. Am J Roentgenol 186(3):673–679. https://doi.org/10.2214/AJR.04.1513
Diel J, Perlmutter S, Venkataramanan N, Mueller R, Lane MJ, Katz DS (2000) Unenhanced helical CT using increased pitch for suspected renal colic: an effective technique for radiation dose reduction? J Comput Assist Tomogr 24(5):795–801. https://doi.org/10.1097/00004728-200009000-00023
Heyer CM, Mohr PS, Lemburg SP, Peters SA, Nicolas V (2007) Image quality and radiation exposure at pulmonary CT angiography with 100- or 120-kVp protocol: prospective randomized study. Radiology 245(2):577–583. https://doi.org/10.1148/radiol.2452061919
Acknowledgements
We would like to thank Mr. Matsumoto for useful discussions. We would like to express the deepest appreciation to Mr. Mori and Mr. Yamamoto, department of radiology, Toyama university hospital. We would like to thank Editage (www.editage.com) for English language editing.
Funding
The authors did not receive support from any organization for the submitted work.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
This study was supported in part by the research grant from A-Line corporation, Osaka, Japan.
Ethical approval
This article does not contain any studies with human participants or animals performed by any of the authors.
Consent to participate
This article does not contain patient data.
Consent for publication
Not applicable.
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
Doi, Y., Teramoto, A., Yamada, A. et al. Estimating subjective evaluation of low-contrast resolution using convolutional neural networks. Phys Eng Sci Med 44, 1285–1296 (2021). https://doi.org/10.1007/s13246-021-01062-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13246-021-01062-7