Training high-performance deep learning classifier for diagnosis in oral cytology using diverse annotations.
Classification
Convolutional neural network
Deep learning
Oral cytology
Probabilistic labeling
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
30 Jul 2024
30 Jul 2024
Historique:
received:
02
04
2024
accepted:
17
07
2024
medline:
31
7
2024
pubmed:
31
7
2024
entrez:
30
7
2024
Statut:
epublish
Résumé
The uncertainty of true labels in medical images hinders diagnosis owing to the variability across professionals when applying deep learning models. We used deep learning to obtain an optimal convolutional neural network (CNN) by adequately annotating data for oral exfoliative cytology considering labels from multiple oral pathologists. Six whole-slide images were processed using QuPath for segmenting them into tiles. The images were labeled by three oral pathologists, resulting in 14,535 images with the corresponding pathologists' annotations. Data from three pathologists who provided the same diagnosis were labeled as ground truth (GT) and used for testing. We investigated six models trained using the annotations of (1) pathologist A, (2) pathologist B, (3) pathologist C, (4) GT, (5) majority voting, and (6) a probabilistic model. We divided the test by cross-validation per slide dataset and examined the classification performance of the CNN with a ResNet50 baseline. Statistical evaluation was performed repeatedly and independently using every slide 10 times as test data. For the area under the curve, three cases showed the highest values (0.861, 0.955, and 0.991) for the probabilistic model. Regarding accuracy, two cases showed the highest values (0.988 and 0.967). For the models using the pathologists and GT annotations, many slides showed very low accuracy and large variations across tests. Hence, the classifier trained with probabilistic labels provided the optimal CNN for oral exfoliative cytology considering diagnoses from multiple pathologists. These results may lead to trusted medical artificial intelligence solutions that reflect diverse diagnoses of various professionals.
Identifiants
pubmed: 39080384
doi: 10.1038/s41598-024-67879-w
pii: 10.1038/s41598-024-67879-w
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
17591Informations de copyright
© 2024. The Author(s).
Références
Warnakulasuriya, S. et al. An appraisal of oral cancer and pre-cancer screening programmes in Europe: A systematic review. J. Oral Pathol. Med 44, 559–570 (2015).
doi: 10.1111/jop.12267
pubmed: 25346441
Sekine, J., Nakatani, E., Hideshima, K., Iwahashi, T. & Sasaki, H. Diagnostic accuracy of oral cancer cytology in a pilot study. Diagn. Pathol. 12, (2017).
Sukegawa, S. et al. Clinical study on primary screening of oral cancer and precancerous lesions by oral cytology. Diagn. Pathol. 15, (2020).
Sukegawa, S. et al. Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists. Sci. Rep. 13, (2023).
Sukegawa, S. et al. Effective deep learning for oral exfoliative cytology classification. Sci. Rep. 12, (2022).
Vega, R. et al. Sample efficient learning of image-based diagnostic classifiers using probabilistic labels. Proc. Mach. Learn. Res. 130, 739–747 (2021).
Iglesias, J. E. et al. An algorithm for optimal fusion of atlases with different labeling protocols. Neuroimage 106, 451–463 (2015).
doi: 10.1016/j.neuroimage.2014.11.031
pubmed: 25463466
Bhardwaj, P., Gupta, P., Guhan, T. & Srinivasan, K. Early diagnosis of retinal blood vessel damage via deep learning-powered collective intelligence models. Comput. Math. Methods Med. 2022, 3571364 (2022).
doi: 10.1155/2022/3571364
pubmed: 35785142
pmcid: 9246601
Popescu, D., El-Khatib, M. & Ichim, L. Skin lesion classification using collective intelligence of multiple neural networks. Sensors 22, 4399 (2022).
doi: 10.3390/s22124399
pubmed: 35746180
pmcid: 9231065
He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. 2016-December, 770–778 (2015).
Gitman, I., Lang, H., Zhang, P. & Xiao, L. Understanding the role of momentum in stochastic gradient ethods. Adv. Neural Inf. Process. Syst. 32, (2019).
Selvaraju, R. R. et al. Grad-cam: Visual explanations from deep networks via gradient-based localization. Int. J. Comput. Vis. 128, 336–359 (2016).
doi: 10.1007/s11263-019-01228-7
Hoang Duc, A. K. et al. Using manifold learning for atlas selection in multi-atlas segmentation. PLoS One 8, (2013).
Li, J., Baba, Y. & Kashima, H. Hyper questions: Unsupervised targeting of a few experts in crowdsourcing. Int. Conf. Inf. Knowl. Manag. Proc. Part F131841, 1069–1078 (2017).
Zhang, Y., Liao, Q. V. & Bellamy, R. K. E. Effect of Confidence and Explanation on Accuracy and Trust Calibration in AI-Assisted Decision Making. In FAT* 2020 - Proc. 2020 Conf. Fairness, Accountability, Transpar. 295–305 (2020). https://doi.org/10.1145/3351095.3372852
Sukegawa, S. et al. Identification of osteoporosis using ensemble deep learning model with panoramic radiographs and clinical covariates. Sci. Rep. 12, (2022).
Huang, S. C., Pareek, A., Seyyedi, S., Banerjee, I. & Lungren, M. P. Fusion of medical imaging and electronic health records using deep learning: A systematic review and implementation guidelines. NPJ Digit. Med. 3, 1–9 (2020).
doi: 10.1038/s41746-020-00341-z