FEEDBACK ON A PUBLICLY DISTRIBUTED IMAGE DATABASE: THE MESSIDOR DATABASE

By Etienne Decencière, Xiwei Zhang, G. Cazuguel, B. Lay, Béatrice Cochener, Caroline Trone, P. Gain, John-Richard Ordóñez-Varela, Pascale Massin, Ali Erginay and 2 more. 2014

1,314 citations recorded by the source corpus.

The Messidor database, which contains hundreds of eye fundus images, has been publicly distributed since 2008. It was created by the Messidor project in order to evaluate automatic lesion segmentation and diabetic retinopathy grading methods. Designing, producing and maintaining such a database entails significant costs. By publicly sharing it, one hopes to bring a valuable resource to the public research community. However, the real interest and benefit of the research community is not easy to quantify. We analyse here the feedback on the Messidor database, after more than 6 years of diffusion. This analysis should apply to other similar research databases.

Subfields: Ophthalmology, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ophthalmology

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Convolutional neural networks based transfer learning for diabetic retinopathy fundus image classification (2017, background): “2) MESSIDOR Dataset: MESSIDOR is also a publicly available dataset [13].”
CANet: Cross-Disease Attention Network for Joint Diabetic Retinopathy and Diabetic Macular Edema Grading (2019, methodology): “We evaluate the effectiveness of our method by comparing it against existing works on Messidor dataset [13]1 and 2018 ISBI IDRiD challenge dataset [12]2.”
IDRiD: Diabetic Retinopathy - Segmentation and Grading Challenge. (2020, methodology): “Some of the common approaches involved for lesion segmentation are mathematical morphology ( Joshi and Karule, 2019; Hatanaka et al., 2008; Zhang et al., 2014 ), region growing ( Fleming et al., 2006; Li and Chutatape, 2004 ), and supervised methods ( Wu et al., 2017; Zhou et al., 2017; Garcia et…”
CABNet: Category Attention Block for Imbalanced Diabetic Retinopathy Grading (2020, methodology): “Messidor DataSet [34]: There are 1,200 fundus images in this dataset and DR is graded into four classes in terms of the severity, which is inconsistent with international standards.”
Transformers in Medical Imaging: A Survey (2022, background): “LAT [260] Eye Fundus 2D Messidor-1 [261] Messidor-2 [261] EyePACS [262] AuC Kappa Hybrid Formulate lesion discovery as a weakly supervised lesion localization problem via a transformer decoder.”
Zoom-in-Net: Deep Mining Lesions for Diabetic Retinopathy Detection (2017, methodology): “The Messidor dataset is a public dataset provided by the Messidor program partners [7].”
REFUGE Challenge: A Unified Framework for Evaluating Automated Methods for Glaucoma Assessment from Fundus Photographs (2019, neutral): “, 2018) Kaggle (Kaggle, 2015), MESSIDOR (Decencire et al., 2014) and IDRiD (Porwal et al.”
CE-Net: Context Encoder Network for 2D Medical Image Segmentation (2019, methodology): “Messidor dataset is a public dataset provided by the Messidor program partners.”
End-to-End Adversarial Retinal Image Synthesis (2018, methodology): “Then, the model as trained on DRIVE was used to segment images from the Messidor-1 dataset [19].”
Medical image retrieval using deep convolutional neural network (2017, background): “The other two classes contain data from Messidor [46] and an open access website for knee images [47].”