Dataset for OCT Classification (50 Normal, 48 AMD & 50 DME)


Dataset for OCT Classification (50 Normal, 48 AMD & 50 DME)

This dataset is acquired at Noor Eye Hospital in Tehran and is consisting of 50 normal, 48 dry AMD, and 50 DME OCTs.

This dataset is acquired at Noor Eye Hospital in Tehran and is consisting of 50 normal, 48 dry AMD, and 50 DME OCTs. For this dataset, the axial resolution is 3:5.m with the scan-dimension of 8:9.7:4 mm2, but the lateral and azimuthal resolutions are not consistent for all patients. So, the number of A-scans varies among 512 or 768 scans where 19, 25, 31, and 61 B-scans per volume are acquired from different patients. The dataset can be downloaded from HERE. The password is MCME2017 and the label of B-scans can be found in DME Labels and AMD Labels (50% shows the suspicious B-scans).

Reza Rasti, Hossein Rabbani*, Alireza Mehri, Fedra Hajizadeh, "Macular OCT Classification using a Multi-Scale Convolutional Neural Network Ensemble”, IEEE Transactions on Medical Imaging, vol. 37, no. 4, pp. 1024-1034, April 2018.

Please reference the above paper if you would like to use any part of datasets and this method.



This site was built using