Application of deep learning algorithm in the recognition of cryptococcosis and talaromycosis skin lesions
Mycoses · 2023
DOI: 10.1111/myc.13598 ↗Kurz gefasst
Fachartikel in der Zeitschrift Mycoses, 2023. Verfasst von Wudi Wei und 10 weiteren Autor:innen. Beteiligte Einrichtung: Guangxi Key Laboratory of AIDS Prevention and Treatment, School of Public Health, Guangxi Medical University, Nanning, China. Hauptthemen laut MEDLINE: Cryptococcosis, Deep Learning, Skin Diseases. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 46 Literaturangaben. Gefördert von: Guangxi Major Science and Technology Project, Guangxi Youth Science Fund Project, Guangxi Medical University Training Program for Distinguished Young Scholars.
Abstract (Originalsprache)
Background
Cryptococcosis and talaromycosis are known as ‘neglected epidemics’ due to their high case fatality rates and low concern. Clinically, the skin lesions of the two fungal diseases are similar and easily misdiagnosed. Therefore, this study aims to develop an algorithm to identify cryptococcosis/talaromycosis skin lesions.
Methods
Skin images of tararomiasis and cryptococcosis were collected from published articles and augmented using the Python Imaging Library (PIL). Then, five deep artificial intelligence models, VGG19, MobileNet, InceptionV3, Incept ResNetV2 and DenseNet201, were developed based on the collected datasets using transfer learning technology. Finally, the performance of the models was evaluated using sensitivity, specificity, F1 score, accuracy, AUC and ROC curve.
Results
In total, 159 articles (79 for cryptococcosis and 80 for talaromycosis), including 101 cryptococcosis skin lesion images and 133 talaromycosis skin lesion images, were collected for further mode construction. Five methods showed good performance for prediction but did not yield satisfactory results for all cases. Among them, DenseNet201 performed best in the validation set, followed by InceptionV3. However, InceptionV3 showed the highest sensitivity, accuracy, F1 score and AUC values in the training set, followed by DenseNet201. The specificity of DenseNet201 in the training set is better than that of InceptionV3.
Conclusions
DenseNet201 and InceptionV3 are equivalent to the optimal model in these conditions and can be used in clinical settings as decision support tools for the identification and classification of skin lesions of cryptococcus/talaromycosis.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Guangxi Key Laboratory of AIDS Prevention and Treatment, School of Public Health, Guangxi Medical University, Nanning, China.
- Guangxi-ASEAN Collaborative Innovation Center for Major Disease Prevention and Treatment, Life Sciences Institute, Guangxi Medical University, Nanning, China.
- Guangxi Biobank, Life Sciences Institute, Guangxi Medical University, Nanning, China.
ORCID-Profile: Wudi Wei
Förderung
- Guangxi Major Science and Technology Project
- Guangxi Youth Science Fund Project
- Guangxi Medical University Training Program for Distinguished Young Scholars
- National Natural Science Foundation of China
Zitieren
APA
Wei, W., He, X., Bao, X., Wang, G., Luo, Q., Chen, L., Zhan, B., Lai, J., Jiang, J., Ye, L., & Liang, H. (2023). Application of deep learning algorithm in the recognition of cryptococcosis and talaromycosis skin lesions. Mycoses. https://doi.org/10.1111/myc.13598
Vancouver
Wei W, He X, Bao X, Wang G, Luo Q, Chen L, et al. Application of deep learning algorithm in the recognition of cryptococcosis and talaromycosis skin lesions. Mycoses. 2023. doi:10.1111/myc.13598
BibTeX
@article{Wei2023,
author = {Wei, Wudi and He, Xiaotao and Bao, Xiuli and Wang, Gang and Luo, Qiang and Chen, Lixiang and Zhan, Baili and Lai, Jingzhen and Jiang, Junjun and Ye, Li and Liang, Hao},
title = {Application of deep learning algorithm in the recognition of cryptococcosis and talaromycosis skin lesions},
journal = {Mycoses},
year = {2023},
doi = {10.1111/myc.13598}
}
EUVIMED