Pre-examinations Improve Automated Metastases Detection on Cranial MRI.

Szczegóły
Abstrakt

Tytuł:: Pre-examinations Improve Automated Metastases Detection on Cranial MRI.
Autorzy:: Deike-Hofmann K
Dancs D; From the Department of Radiology, German Cancer Research Center, Heidelberg.
Paech D; From the Department of Radiology, German Cancer Research Center, Heidelberg.
Schlemmer HP; From the Department of Radiology, German Cancer Research Center, Heidelberg.
Maier-Hein K; Department for Medical Image Computing, German Cancer Research Center, Heidelberg, Germany.
Bäumer P; From the Department of Radiology, German Cancer Research Center, Heidelberg.
Radbruch A; Department of Neuroradiology, Bonn University Clinic, Bonn.
Götz M; Department for Medical Image Computing, German Cancer Research Center, Heidelberg, Germany.
Źródło:: Investigative radiology [Invest Radiol] 2021 May 01; Vol. 56 (5), pp. 320-327.
Typ publikacji:: Journal Article; Research Support, Non-U.S. Gov't
Język:: English
Imprint Name(s):: Publication: 1998- : Hagerstown, MD : Lippincott Williams & Wilkins
Original Publication: Philadelphia.
MeSH Terms:: Contrast Media*
Magnetic Resonance Imaging*
Artifacts ; Retrospective Studies ; Sensitivity and Specificity
References:: Matthews NH, Li W-Q, Qureshi AA, et al. Epidemiology of melanoma. In: Ward WH, Farma JM, eds. Cutaneous Melanoma: Etiology and Therapy . 1st ed. Brisbane, Australia: Codon Publications; 2017:3–22. doi:10.15586/codon.cutaneousmelanoma.2017.ch1.
Postow MA, Chesney J, Pavlick AC, et al. Nivolumab and ipilimumab versus ipilimumab in untreated melanoma. N Engl J Med . 2015;372:2006–2017. doi: 10.1056/NEJMoa1414428. (PMID: 10.1056/NEJMoa1414428)
Wolchok JD, Chiarion-Sileni V, Gonzalez R, et al. Overall survival with combined nivolumab and ipilimumab in advanced melanoma. N Engl J Med . 2017;377:1345–1356.
Robert C, Schachter J, Long GV, et al. Pembrolizumab versus ipilimumab in advanced melanoma. N Engl J Med . 2015;372:2521–2532.
Bickelhaupt S, Jaeger PF, Laun FB, et al. Radiomics based on adapted diffusion kurtosis imaging helps to clarify most mammographic findings suspicious for cancer. Radiology . 2018;287:761–770.
Schelb P, Kohl S, Radtke JP, et al. Classification of cancer at prostate MRI: deep learning versus clinical PI-RADS assessment. Radiology . 2019;293:607–617.
Niemeyer F, Galbusera F, Tao Y, et al. A deep learning model for the accurate and reliable classification of disc degeneration based on MRI data. Invest Radiol . 2020. doi:10.1097/RLI.0000000000000709. (PMID: 10.1097/RLI.0000000000000709)
Perl RM, Grimmer R, Hepp T, et al. Can a novel deep neural network improve the computer-aided detection of solid pulmonary nodules and the rate of false-positive findings in comparison to an established machine learning computer-aided detection? Invest Radiol . 2020. doi:10.1097/RLI.0000000000000713. (PMID: 10.1097/RLI.0000000000000713)
Wichmann JL, Willemink MJ, De Cecco CN. Artificial intelligence and machine learning in radiology: current state and considerations for routine clinical implementation. Invest Radiol . 2020;55:619–627.
Ambrosini RD, Wang P, O'Dell WG. Computer-aided detection of metastatic brain tumors using automated three-dimensional template matching. J Magn Reson Imaging . 2010;31:85–93.
Farjam R, Parmar HA, Noll DC, et al. An approach for computer-aided detection of brain metastases in post-Gd T1-W MRI. Magn Reson Imaging . 2012;30:824–836.
Pérez-Ramírez Ú, Arana E, Moratal D. Brain metastases detection on MR by means of three-dimensional tumor-appearance template matching. J Magn Reson Imaging . 2016;44:642–652.
Sunwoo L, Kim YJ, Choi SH, et al. Computer-aided detection of brain metastasis on 3D MR imaging: observer performance study. PLoS One . 2017;12:e0178265.
Charron O, Lallement A, Jarnet D, et al. Automatic detection and segmentation of brain metastases on multimodal MR images with a deep convolutional neural network. Comput Biol Med . 2018;95:43–54.
Perkuhn M, Stavrinou P, Thiele F, et al. Clinical evaluation of a multiparametric deep learning model for glioblastoma segmentation using heterogeneous magnetic resonance imaging data from clinical routine. Invest Radiol . 2018;53:647–654.
Finck T, Li H, Grundl L, et al. Deep-learning generated synthetic double inversion recovery images improve multiple sclerosis lesion detection. Invest Radiol . 2020;55:318–323.
Grøvik E, Yi D, Iv M, et al. Deep learning enables automatic detection and segmentation of brain metastases on multisequence MRI. J Magn Reson Imaging . 2020;51:175–182.
Federau C, Christensen S, Scherrer N, et al. Improved segmentation and detection sensitivity of diffusion-weighted stroke lesions with synthetically. Radiol Artif Intell . 2020;2:e190217. doi: 10.1148/ryai.2020190217. (PMID: 10.1148/ryai.2020190217)
Kickingereder P, Isensee F, Tursunova I, et al. Automated quantitative tumour response assessment of MRI in neuro-oncology with artificial neural networks: a multicentre, retrospective study. Lancet Oncol . 2019;20:728–740.
Chopra S, Hadsell R, Lecun Y. Learning a similarity metric discriminatively, with application to face verification. In: Conference on Computer Vision and Pattern Recognition (CVPR) . San Diego, CA, USA: IEEE Computer Society Conference; 2005:539–546. doi:10.1109/CVPR.2005.202.
Zagoruyko S, Komodakis N. Learning to compare image patches via convolutional neural networks. In: Conference on Computer Vision and Pattern Recognition (CVPR) . Boston, MA, USA: IEEE Computer Society Conference; 2015:4353–4361. doi: 10.1109/CVPR.2015.7299064. arXiv:1504.03641v1.
Stent S, Gherardi R, Stenger B, et al. Detecting change for multi-view, long-term surface inspection. BMVC . 2015. Available at: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.701.216&rep=rep1&type=pdf .
Floca R. MatchPoint: on bridging the innovation gap between algorithmic research and clinical use in image registration. In: World Congress on Medical Physics and Biomedical Engineering . Munich, Germany: Springer Berlin, Heidelberg; 2010:1105–1108.
Götz M, Nolden M, Maier-Hein K. MITK Phenotyping: an open-source toolchain for image-based personalized medicine with radiomics. Radiother Oncol . 2019;131:108–111.
Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation. In: Conference on Computer Vision and Pattern Recognition (CVPR) . Boston, MA, USA: IEEE Computer Society; 2015:3431–3440. doi:10.1109/CVPR.2015.7298965.
Daudt RC, Le Saux B, Boulch A. Fully Convolutional Siamese Networks for Change Detection. October 2018. 25th IEEE International Conference on Image Processing (ICIP), Athens, 2018;4063–4067. doi: 10.1109/ICIP.2018.8451652.
Isensee F, Petersen J, Klein A, et al. nnU-Net: self-adapting framework for U-net-based medical image segmentation. arXiv:1809.10486v1 . 2018;1–11.
Petersen J, Heiland S, Bendszus M, et al. Leveraging open source software to close translational gaps in medical image computing. In: Maier A, Deserno T, Handels H, et al., eds. Bildverarbeitung Für Die Medizin . Berlin/Heidelberg, Germany: Springer; 2018. https://doi.org/10.1007/978-3-662-56537-7_18 .
Van Rijsbergen CJ. Information Retrieval . 2nd ed. London, United Kingdom: Butterworth-Heinemann Ltd; 1979.
Deike-Hofmann K, Thünemann D, Breckwoldt MO, et al. Sensitivity of different MRI sequences in the early detection of melanoma brain metastases. PLoS One . 2018;13:e0193946.
Schwarz D, Niederle T, Münch P, et al. Susceptibility-weighted imaging in malignant melanoma brain metastasis. J Magn Reson Imaging . 2019;50:1251–1259.
Campanella G, Hanna MG, Geneslaw L, et al. Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med . 2019;25:1301–1309.
Runge VM. Critical questions regarding gadolinium deposition in the brain and body after injections of the gadolinium-based contrast agents, safety, and clinical recommendations in consideration of the EMA's Pharmacovigilance and Risk Assessment Committee recommendation for suspension of the marketing authorizations for 4 linear agents. Invest Radiol . 2017;52:317–323.
Bower DV, Richter JK, Von Tengg-Kobligk H, et al. Gadolinium-based MRI contrast agents induce mitochondrial toxicity and cell death in human neurons, and toxicity increases with reduced kinetic stability of the agent. Invest Radiol . 2019;54:453–463.
Radbruch A, Richter H, Bücker P, et al. Is small fiber neuropathy induced by gadolinium-based contrast agents? Invest Radiol . 2020;55:473–480.
Radbruch A, Richter H, Fingerhut S, et al. Gadolinium deposition in the brain in a large animal model. Invest Radiol . 2019;54:531–536.
Bertinetto L, Valmadre J, Henriques JF, et al. Fully-Convolutional Siamese Networks for Object Tracking. In: Computer Vision - ECCV 2016 Workshops, 2016, Springer International Publishing. DOI: 10.1007/978-3-319-48881-3_56.
Daudt RC, Le Saux B, Boulch A, et al. Urban change detection for multispectral earth observation using convolutional neural networks. Int Geosci Remote Sens Symp . 2018;2115–2118. doi:10.1109/IGARSS.2018.8518015. (PMID: 10.1109/IGARSS.2018.8518015)
Substance Nomenclature:: 0 (Contrast Media)
Entry Date(s):: Date Created: 20201201 Date Completed: 20211015 Latest Revision: 20230926
Update Code:: 20240105
DOI:: 10.1097/RLI.0000000000000745
PMID:: 33259442
: Czasopismo naukowe

Materials and Methods: Our local ethics committee approved this retrospective monocenter study.First, a dual-time approach was assessed, for which the CNN was provided sequences of the MRI that initially depicted new MM (diagnosis MRI) as well as of a prediagnosis MRI: inclusion of only contrast-enhanced T1-weighted images (CNNdual_ce) was compared with inclusion of also the native T1-weighted images, T2-weighted images, and FLAIR sequences of both time points (CNNdual_all).Second, results were compared with the corresponding single time approaches, in which the CNN was provided exclusively the respective sequences of the diagnosis MRI.Casewise diagnostic performance parameters were calculated from 5-fold cross-validation.
Results: In total, 94 cases with 494 MMs were included. Overall, the highest diagnostic performance was achieved by inclusion of only the contrast-enhanced T1-weighted images of the diagnosis and of a prediagnosis MRI (CNNdual_ce, sensitivity = 73%, PPV = 25%, F1-score = 36%). Using exclusively contrast-enhanced T1-weighted images as input resulted in significantly less false-positives (FPs) compared with inclusion of further sequences beyond contrast-enhanced T1-weighted images (FPs = 5/7 for CNNdual_ce/CNNdual_all, P < 1e-5). Comparison of contrast-enhanced dual and mono time approaches revealed that exclusion of prediagnosis MRI significantly increased FPs (FPs = 5/10 for CNNdual_ce/CNNce, P < 1e-9).Approaches with only native sequences were clearly inferior to CNNs that were provided contrast-enhanced sequences.
Conclusions: Automated MM detection on contrast-enhanced T1-weighted images performed with high sensitivity. Frequent FPs due to artifacts and vessels were significantly reduced by additional inclusion of prediagnosis MRI, but not by inclusion of further sequences beyond contrast-enhanced T1-weighted images. Future studies might investigate different change detection architectures for computer-aided detection.
Competing Interests: Conflicts of interest and sources of funding: A.R. received personal fees for consulting and talks (within the last 3 years) from Bayer, Guerbet, and Novartis, and financial study support from Bayer and Guerbet. K.D.H. received personal fees for talks from GE and financial study support from Bayer and Guerbet. The other authors declare no conflicts of interest.
(Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.)

Informacja