Display options
Cite
Piccini D, Demesmaeker R, Heerfordt J, et al. Deep Learning to Automate Reference-Free Image Quality Assessment of Whole-Heart MR Images. Radiol Artif Intell. 2020;2(3):e190123doi: 10.1148/ryai.2020190123.
Piccini, D., Demesmaeker, R., Heerfordt, J., Yerly, J., Di Sopra, L., Masci, P. G., Schwitter, J., Van De Ville, D., Richiardi, J., Kober, T., & Stuber, M. (2020). Deep Learning to Automate Reference-Free Image Quality Assessment of Whole-Heart MR Images. Radiology. Artificial intelligence, 2(3), e190123. https://doi.org/10.1148/ryai.2020190123
Piccini, Davide, et al. "Deep Learning to Automate Reference-Free Image Quality Assessment of Whole-Heart MR Images." Radiology. Artificial intelligence vol. 2,3 (2020): e190123. doi: https://doi.org/10.1148/ryai.2020190123
Piccini D, Demesmaeker R, Heerfordt J, Yerly J, Di Sopra L, Masci PG, Schwitter J, Van De Ville D, Richiardi J, Kober T, Stuber M. Deep Learning to Automate Reference-Free Image Quality Assessment of Whole-Heart MR Images. Radiol Artif Intell. 2020 May 27;2(3):e190123. doi: 10.1148/ryai.2020190123. eCollection 2020 May. PMID: 33937825; PMCID: PMC8082371.
Copy Download .nbib Format: NLM
Share it on

Radiol Artif Intell. 2020 May 27;2(3):e190123. doi: 10.1148/ryai.2020190123. eCollection 2020 May.

Deep Learning to Automate Reference-Free Image Quality Assessment of Whole-Heart MR Images.

Radiology. Artificial intelligence

Davide Piccini, Robin Demesmaeker, John Heerfordt, Jérôme Yerly, Lorenzo Di Sopra, Pier Giorgio Masci, Juerg Schwitter, Dimitri Van De Ville, Jonas Richiardi, Tobias Kober, Matthias Stuber

Affiliations

  1. Advanced Clinical Imaging Technology, Siemens Healthcare, Lausanne, Switzerland (D.P., R.D., J.H., J.R., T.K.); Department of Diagnostic and Interventional Radiology, Lausanne University Hospital and University of Lausanne, Rue de Bugnon 46, BH 8.80, 1011 Lausanne, Switzerland (D.P., J.H., J.Y., L.D.S., J.R., T.K., M.S.); LTS5, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland (D.P., J.R., T.K.); Institute of Electrical Engineering, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland (R.D.); Institute of Bioengineering/Center for Neuroprosthetics, École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland (R.D., D.V.D.V.); Center for Biomedical Imaging (CIBM), Lausanne, Switzerland (J.Y., M.S.); Division of Cardiology and Cardiac MR Center, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland (P.G.M., J.S.); and Department of Radiology and Medical Informatics, University Hospital of Geneva (HUG), Geneva, Switzerland (D.V.D.V.).

PMID: 33937825 PMCID: PMC8082371 DOI: 10.1148/ryai.2020190123

Abstract

PURPOSE: To develop and characterize an algorithm that mimics human expert visual assessment to quantitatively determine the quality of three-dimensional (3D) whole-heart MR images.

MATERIALS AND METHODS: In this study, 3D whole-heart cardiac MRI scans from 424 participants (average age, 57 years ± 18 [standard deviation]; 66.5% men) were used to generate an image quality assessment algorithm. A deep convolutional neural network for image quality assessment (IQ-DCNN) was designed, trained, optimized, and cross-validated on a clinical database of 324 (training set) scans. On a separate test set (100 scans), two hypotheses were tested:

RESULTS: Regression performance of the IQ-DCNN was within the range of human intra- and interobserver agreement and in very good agreement with the human expert (

CONCLUSION: The proposed IQ-DCNN was trained to mimic expert visual image quality assessment of 3D whole-heart MR images. The results from the IQ-DCNN were in good agreement with human expert reading, and the network was capable of automatically comparing different reconstructed volumes.

2020 by the Radiological Society of North America, Inc.

Conflict of interest statement

Disclosures of Conflicts of Interest: D.P. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: employed by and has stock opti

References

  1. MAGMA. 2018 Apr;31(2):243-256 - PubMed
  2. Eur Radiol. 2018 Mar;28(3):1267-1275 - PubMed
  3. IEEE Trans Image Process. 2018 Jan;27(1):206-219 - PubMed
  4. Int J Cardiovasc Imaging. 2016 Dec;32(12):1735-1744 - PubMed
  5. J Cardiovasc Magn Reson. 2015 Jul 09;17:55 - PubMed
  6. Atten Percept Psychophys. 2010 Jul;72(5):1205-17 - PubMed
  7. Theriogenology. 2010 Jun;73(9):1167-79 - PubMed
  8. Radiology. 2003 Aug;228(2):303-8 - PubMed
  9. Magn Reson Med. 2012 Aug;68(2):571-9 - PubMed
  10. AJR Am J Roentgenol. 1997 May;168(5):1369-75 - PubMed
  11. J Magn Reson Imaging. 2018 Mar;47(3):723-728 - PubMed
  12. Circ Cardiovasc Imaging. 2016 May;9(5): - PubMed
  13. Nature. 2015 May 28;521(7553):436-44 - PubMed
  14. Magn Reson Med. 2002 Oct;48(4):658-66 - PubMed
  15. Magn Reson Med. 2018 Feb;79(2):826-838 - PubMed
  16. J Cardiovasc Magn Reson. 2013 Jun 20;15:55 - PubMed
  17. Heart Rhythm. 2013 Jun;10(6):864-72 - PubMed
  18. Magn Reson Med. 2016 Apr;75(4):1594-604 - PubMed
  19. J Digit Imaging. 2018 Dec;31(6):869-878 - PubMed
  20. Magn Reson Med. 2016 Feb;75(2):775-88 - PubMed
  21. Appl Radiat Isot. 1999 Jan;50(1):21-38 - PubMed
  22. Radiology. 2014 Feb;270(2):378-86 - PubMed
  23. IEEE Trans Image Process. 2004 Apr;13(4):600-12 - PubMed
  24. JACC Cardiovasc Imaging. 2014 Sep;7(9):882-92 - PubMed
  25. Magn Reson Med. 2017 Apr;77(4):1473-1484 - PubMed

Publication Types