A Multi-dataset Approach for DME Risk Detection in Eye Fundus Images

• Type: Article in International Conference Proceedings Book
• Date: 2020

Authors

Catarina Carvalho (Author) Other; João Pedrosa (Author) Other; Maia, C (Author) Other; Susana Penas (Author) Other; Ângela Carneiro (Author) Other; Luís Mendonça (Author) Other; Ana Maria Mendonça (Author) FEUP; Aurélio Campilho (Author) FEUP

Conference proceedings International

Title: Image Analysis and Recognition - 17th International Conference, ICIAR 2020, Póvoa de Varzim, Portugal, June 24-26, 2020, Proceedings, Part II

Pages: 285-298

17th International Conference on Image Analysis and Recognition, ICIAR 2020

24 June 2020 through 26 June 2020

Indexing

Scopus - 2 Citations

INSPEC

Other information

Authenticus ID: P-00S-CVH

DOI: 10.1007/978-3-030-50516-5_25

Abstract (EN): Diabetic macular edema is a leading cause of visual loss for patients with diabetes. While diagnosis can only be performed by optical coherence tomography, diabetic macular edema risk assessment is often performed in eye fundus images in screening scenarios through the detection of hard exudates. Such screening scenarios are often associated with large amounts of data, high costs and high burden on specialists, motivating then the development of methodologies for automatic diabetic macular edema risk prediction. Nevertheless, significant dataset domain bias, due to different acquisition equipment, protocols and/or different populations can have significantly detrimental impact on the performance of automatic methods when transitioning to a new dataset, center or scenario. As such, in this study, a method based on residual neural networks is proposed for the classification of diabetic macular edema risk. This method is then validated across multiple public datasets, simulating the deployment in a multi-center setting and thereby studying the method¿s generalization capability and existing dataset domain bias. Furthermore, the method is tested on a private dataset which more closely represents a realistic screening scenario. An average area under the curve across all public datasets of 0.891 ± 0.013 was obtained with a ResNet50 architecture trained on a limited amount of images from a single public dataset (IDRiD). It is also shown that screening scenarios are significantly more challenging and that training across multiple datasets leads to an improvement of performance (area under the curve of 0.911 ± 0.009). © Springer Nature Switzerland AG 2020.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 14