Automated Neural Architecture Search for Cardiac Amyloidosis Classification from [18F]-Florbetaben PET Images.
AutoML
Cardiac amyloidosis
Neural architecture search
Nuclear medicine
[18-F]-Florbetaben
Journal
Journal of imaging informatics in medicine
ISSN: 2948-2933
Titre abrégé: J Imaging Inform Med
Pays: Switzerland
ID NLM: 9918663679206676
Informations de publication
Date de publication:
02 Oct 2024
02 Oct 2024
Historique:
received:
18
06
2024
accepted:
08
09
2024
revised:
30
08
2024
medline:
2
10
2024
pubmed:
2
10
2024
entrez:
2
10
2024
Statut:
aheadofprint
Résumé
Medical image classification using convolutional neural networks (CNNs) is promising but often requires extensive manual tuning for optimal model definition. Neural architecture search (NAS) automates this process, reducing human intervention significantly. This study applies NAS to [18F]-Florbetaben PET cardiac images for classifying cardiac amyloidosis (CA) sub-types (amyloid light chain (AL) and transthyretin amyloid (ATTR)) and controls. Following data preprocessing and augmentation, an evolutionary cell-based NAS approach with a fixed network macro-structure is employed, automatically deriving cells' micro-structure. The algorithm is executed five times, evaluating 100 mutating architectures per run on an augmented dataset of 4048 images (originally 597), totaling 5000 architectures evaluated. The best network (NAS-Net) achieves 76.95% overall accuracy. K-fold analysis yields mean ± SD percentages of sensitivity, specificity, and accuracy on the test dataset: AL subjects (98.7 ± 2.9, 99.3 ± 1.1, 99.7 ± 0.7), ATTR-CA subjects (93.3 ± 7.8, 78.0 ± 2.9, 70.9 ± 3.7), and controls (35.8 ± 14.6, 77.1 ± 2.0, 96.7 ± 4.4). NAS-derived network performance rivals manually determined networks in the literature while using fewer parameters, validating its automatic approach's efficacy.
Identifiants
pubmed: 39356368
doi: 10.1007/s10278-024-01275-8
pii: 10.1007/s10278-024-01275-8
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Informations de copyright
© 2024. The Author(s).
Références
D. Shen, G. Wu, H. Il Suk, Deep Learning in Medical Image Analysis, Annu. Rev. Biomed. Eng. 19 (2017) 221–248. https://doi.org/10.1146/annurev-bioeng-071516-044442 .
K. Suzuki, Overview of deep learning in medical imaging, Radiol. Phys. Technol. 10 (2017) 257–273. https://doi.org/10.1007/s12194-017-0406-5 .
doi: 10.1007/s12194-017-0406-5
P. Ren, Y. Xiao, X. Chang, P.Y. Huang, Z. Li, X. Chen, X. Wang, A comprehensive survey of neural architecture search: Challenges and solutions, ACM Comput. Surv. 54 (2021). https://doi.org/10.1145/3447582 .
Z. Zhao, L. Alzubaidi, J. Zhang, Y. Duan, Y. Gu, A comparison review of transfer learning and self-supervised learning: Definitions, applications, advantages and limitations, Expert Systems with Applications, (2024) 242: 122807, https://doi.org/10.1016/j.eswa.2023.122807 .
doi: 10.1016/j.eswa.2023.122807
T. Elsken, J.H. Metzen, F. Hutter, Neural Architecture Search: A Survey, (2018). http://arxiv.org/abs/1808.05377 .
B. Baker, O. Gupta, N. Naik, R. Raskar, Designing Neural Network Architectures using Reinforcement Learning, (2016). http://arxiv.org/abs/1611.02167 . in Proc. Int. Conf. Learn. Represent. (ICLR), Mar. 2017, pp. 1–18. doi: 1611.02167.
E. Real, S. Moore, A. Selle, S. Saxena, Y.L. Suematsu, J. Tan, Q. Le, A. Kurakin, Large-Scale Evolution of Image Classifiers, (2017). http://arxiv.org/abs/1703.01041 .
A. Kwasigroch, M. Grochowski, A. Mikolajczyk, Neural architecture search for skin lesion classification, IEEE Access. 8 (2020) 9061–9071. https://doi.org/10.1109/ACCESS.2020.2964424 .
doi: 10.1109/ACCESS.2020.2964424
Y. Weng, T. Zhou, Y. Li, X. Qiu, NAS-Unet: Neural architecture search for medical image segmentation, IEEE Access. 7 (2019) 44247–44257. https://doi.org/10.1109/ACCESS.2019.2908991 .
doi: 10.1109/ACCESS.2019.2908991
M.F. Santarelli, D. Genovesi, V. Positano, M. Scipioni, G. Vergaro, B. Favilli, A. Giorgetti, M. Emdin, L. Landini, P. Marzullo, Deep-learning-based cardiac amyloidosis classification from early acquired pet images, Int. J. Cardiovasc. Imaging. 37 (2021) 2327–2335. https://doi.org/10.1007/s10554-021-02190-7 .
doi: 10.1007/s10554-021-02190-7
pubmed: 33591476
M. Wistuba, A. Rawat, T. Pedapati, A Survey on Neural Architecture Search, (2019). http://arxiv.org/abs/1905.01392 .
B. Zoph, V. Vasudevan, J. Shlens, Q. V. Le, Learning Transferable Architectures for Scalable Image Recognition, Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. (2018) 8697–8710. https://doi.org/10.1109/CVPR.2018.00907 .
S. Alarie, C. Audet, A. E. Gheribi, M. Kokkolaras, S. Le Digabel, Two decades of blackbox optimization applications, EURO Journal on Computational Optimization, (2021), 9:100011, https://doi.org/10.1016/j.ejco.2021.100011
doi: 10.1016/j.ejco.2021.100011
M.G. Fernandez-Godino, Review of multi-fidelity models, Advances in Computational Science and Engineering (2023), (1)4;351-400. https://doi.org/10.3934/acse.2023015 .
doi: 10.3934/acse.2023015
G. Bender, P.-J. Kindermans, B. Zoph, V. Vasudevan, and Q. Le, Understanding and simplifying one-shot architecture search, (2018) Proceedings of Machine Learning Research, 80:550–559.
B. Zoph, Q. V. Le, Neural architecture search with reinforcement learning, 5th Int. Conf. Learn. Represent. ICLR 2017 - Conf. Track Proc. (2017) 1–16.
H. Liu, K. Simonyan, Y. Yang, DARTS: Differentiable architecture search, 7th Int. Conf. Learn. Represent. ICLR 2019. (2019) 1–13.
H. Liu, K. Simonyan, O. Vinyals, C. Fernando, K. Kavukcuoglu, Hierarchical representations for efficient architecture search, 6th Int. Conf. Learn. Represent. ICLR 2018 - Conf. Track Proc. (2018) 1–13.
E. Real, A. Aggarwal, Y. Huang, Q. V. Le, Regularized evolution for image classifier architecture search, 33rd AAAI Conf. Artif. Intell. AAAI 2019, 31st Innov. Appl. Artif. Intell. Conf. IAAI 2019 9th AAAI Symp. Educ. Adv. Artif. Intell. EAAI 2019. (2019) 4780–4789. https://doi.org/10.1609/aaai.v33i01.33014780 .
T. Domhan, T. Springenberg, F. Hutter, Speeding Up Automatic Hyperparameter Optimization of.pdf, Twenty-Fourth Int. Jt. Conf. Artif. Intell. (. (2015) 3460–3468.
A. Klein, S. Falkner, J.T. Springenberg, F. Hutter, Learning curve prediction with Bayesian neural networks, 5th Int. Conf. Learn. Represent. ICLR 2017 - Conf. Track Proc. (2017).
R.H. Falk, S.W. Dubrey, Amyloid Heart Disease, Prog. Cardiovasc. Dis. 52 (2010) 347–361. https://doi.org/10.1016/j.pcad.2009.11.007 .
doi: 10.1016/j.pcad.2009.11.007
pubmed: 20109604
M. Skinner, V. Sanchorawala, D.C. Seldin, L.M. Dember, R.H. Falk, J.L. Berk, J.J. Anderson, C. O'Hara, K.T. Finn, C.A. Libbey, J. Wiesman, K. Quillen, N. Swan, D.G. Wright, High-Dose Melphalan and Autologous Stem-Cell Transplantation in Patients with AL Amyloidosis: An 8-Year Study, Ann. Intern. Med. 140 (2004) 85. https://doi.org/10.7326/0003-4819-140-2-200401200-00008 .
doi: 10.7326/0003-4819-140-2-200401200-00008
pubmed: 14734330
F.L. Ruberg, J.L. Berk, Transthyretin (TTR) cardiac amyloidosis, Circulation. 126 (2012) 1286–1300. https://doi.org/10.1161/CIRCULATIONAHA.111.078915 .
doi: 10.1161/CIRCULATIONAHA.111.078915
pubmed: 22949539
pmcid: 3501197
P. Mollee, P. Renaut, D. Gottlieb, H. Goodman, How to diagnose amyloidosis, Intern. Med. J. 44 (2014) 7–17. https://doi.org/10.1111/imj.12288 .
M.F. Santarelli, M. Scipioni, D. Genovesi, A. Giorgetti, P. Marzullo, L. Landini, Imaging Techniques as an Aid in the Early Detection of Cardiac Amyloidosis, Curr. Pharm. Des. 27 (2020) 1878–1889. https://doi.org/10.2174/1381612826666200813133557 .
doi: 10.2174/1381612826666200813133557
S.P. Lee, J.B. Park, H.K. Kim, Y.J. Kim, M. Grogan, D.W. Sohn, Contemporary imaging diagnosis of cardiac amyloidosis, J. Cardiovasc. Imaging. 27 (2019) 1–10. https://doi.org/10.4250/jcvi.2019.27.e9 .
doi: 10.4250/jcvi.2019.27.e9
pubmed: 30701710
pmcid: 6358430
M. Kircher, S. Ihne, J. Brumberg, C. Morbach, S. Knop, K.M. Kortüm, S. Störk, A.K. Buck, T. Reiter, W.R. Bauer, C. Lapa, Detection of cardiac amyloidosis with 18F-Florbetaben-PET/CT in comparison to echocardiography, cardiac MRI and DPD-scintigraphy, Eur. J. Nucl. Med. Mol. Imaging. 46 (2019) 1407–1416. https://doi.org/10.1007/s00259-019-04290-y .
doi: 10.1007/s00259-019-04290-y
pubmed: 30798427
Y Y.J. Kim, S. Ha, Y. il Kim, Cardiac amyloidosis imaging with amyloid positron emission tomography: A systematic review and meta-analysis, J. Nucl. Cardiol. 27 (2020) 123–132. https://doi.org/10.1007/s12350-018-1365-x .
D. Genovesi, G. Vergaro, A. Giorgetti, P. Marzullo, M. Scipioni, M.F. Santarelli, A. Pucci, G. Buda, E. Volpi, M. Emdin, [18F]-Florbetaben PET/CT for Differential Diagnosis Among Cardiac Immunoglobulin Light Chain, Transthyretin Amyloidosis, and Mimicking Conditions, JACC Cardiovasc. Imaging. 14 (2021) 246–255. https://doi.org/10.1016/j.jcmg.2020.05.031 .
doi: 10.1016/j.jcmg.2020.05.031
pubmed: 32771577
M.F. Santarelli, D. Genovesi, M. Scipioni, V. Positano, B. Favilli, A. Giorgetti, G. Vergaro, L. Landini, M. Emdin, P. Marzullo, Cardiac amyloidosis characterization by kinetic model fitting on [18F]florbetaben PET images, J. Nucl. Cardiol. 29 (2022) 1919–1932. https://doi.org/10.1007/s12350-021-02608-8 .
doi: 10.1007/s12350-021-02608-8
pubmed: 33864226
J.D. Gillmore, A. Wechalekar, J. Bird, J. Cavenagh, S. Hawkins, M. Kazmi, H.J. Lachmann, P.N. Hawkins, G. Pratt, Guidelines on the diagnosis and investigation of AL amyloidosis, Br. J. Haematol. 168 (2015) 207–218. https://doi.org/10.1111/bjh.13156 .
doi: 10.1111/bjh.13156
pubmed: 25312307
J.D. Gillmore, M.S. Maurer, R.H. Falk, G. Merlini, T. Damy, A. Dispenzieri, A.D. Wechalekar, J.L. Berk, C.C. Quarta, M. Grogan, H.J. Lachmann, S. Bokhari, A. Castano, S. Dorbala, G.B. Johnson, A.W.J.M. Glaudemans, T. Rezk, M. Fontana, G. Palladini, P. Milani, P.L. Guidalotti, K. Flatman, T. Lane, F.W. Vonberg, C.J. Whelan, J.C. Moon, F.L. Ruberg, E.J. Miller, D.F. Hutt, B.P. Hazenberg, C. Rapezzi, P.N. Hawkins, Nonbiopsy diagnosis of cardiac transthyretin amyloidosis, Circulation. 133 (2016) 2404–2412. https://doi.org/10.1161/CIRCULATIONAHA.116.021612 .
doi: 10.1161/CIRCULATIONAHA.116.021612
pubmed: 27143678
E. Goceri, Medical image data augmentation: techniques, comparisons and interpretations. Artificial Intelligence Review (2023) 56:12561–12605. https://doi.org/10.1007/s10462-023-10453-z
S. Ioffe, C. Szegedy, Batch normalization: Accelerating deep network training by reducing internal covariate shift, 32nd Int. Conf. Mach. Learn. ICML 2015. 1 (2015) 448–456.