【This is a TEMPORARY host of iChallenge-AMD】

The ONLINE iChallenges moved to Baidu Research Open-Access Dataset (BROAD) for the better experience. http://ai.baidu.com/broad/introduction
The 1st two iChallenges are iChallenge-GON (REFUGE) and iChallenge-AMD.

FULL iChallenge will be released soon with

1) a brand NEW and NICE-looking website;

2) FREE online computation rescoure (on Baidu AI Studio);

3) automated performance evaluation, limited to 2 submissions per day per account;

4) full-time customer SERVICE.

But we need to

1) make sure the annotations are ACCURACT;

2) DEBUG the NEW systems.

Please also keep an eye on the challenge-PM, which has an on-site version named 'PALM' together with ISBI 2019 (Venice Italy, Apr 8-11) 

Challenge 9: PALM: PathologicAL Myopia detection from retinal images

Yanwu (Frank) Xu <xuyanwu@baidu.com>
Hrvoje Bogunovic <hrvoje.bogunovic@meduniwien.ac.at>

Challenge websitehttps://palm.grand-challenge.org/

The PALM challenge focuses on the investigation and development of algorithms associated with diagnosis of Pathologic Myopia (PM) and segmentation of lesions in fundus photos from PM patients. Myopia is currently the ocular disease with the highest morbidity. About 2 billion people have myopia in the world, 35% of which are high myopia. High myopia leads to elongation of axial length and thinning of retinal structures. With progression of the disease into PM, macular retinoschisis, retinal atrophy and even retinal detachment may occur, causing irreversible impairment to visual acuity. There are typical signs of PM in fundus photos, including atrophy, lacquer crack, etc. Neural networks can be trained to help screening and grading PM in large populations.


Clinical Background

Age-related macular degeneration, abbreviated as AMD, is a degenerative disorder in the macular region. It mainly occurs in people older than 45 years old and its incidence rate is even higher than diabetic retinopathy in the elderly.  

The etiology of AMD is not fully understood, which could be related to multiple factors, including genetics, chronic photodestruction effect, and nutritional disorder. AMD is classified into Dry AMD and Wet AMD. Dry AMD (also called nonexudative AMD) is not neovascular. It is characterized by progressive atrophy of retinal pigment epithelium (RPE). In the late stage, drusen and the large area of atrophy could be observed under ophthalmoscopy. Wet AMD (also called neovascular or exudative AMD), is characterized by active neovascularization under RPE, subsequently causing exudation, hemorrhage, and scarring, and will eventually cause irreversible damage to the photoreceptors and rapid vision loss if left untreated.

An early diagnosis of AMD is crucial to treatment and prognosis. Fundus photo is one of the basic examinations. The current dataset is composed of AMD and non-AMD (myopia, normal control, etc.) photos. Typical signs of AMD that can be found in these photos include drusen, exudation, hemorrhage, etc. 


Task 1: Classification of AMD and non-AMD fundus images.

DescriptionThe reference standard for AMD presence obtained from the health records, which is not based on fundus image ONLY, but also take OCT, Visual Field, and other facts into consideration. For training data, AMD and non-AMD labels (a.k.a. the reference standard) are reflected in the image folder and file names.

Data: Released on Oct 20 

The classification results should be provided in a single CSV file, named “classification_results.csv”, with the first column corresponding to the filename of the test fundus image (including the extension “.jpg”) and the second column containing the estimated classification probability/risk of the image belonging to a patient diagnosed with AMD (value from 0.0 to 1.0).

Task 2: Localization of disc and fovea.

DescriptionManual pixel-wise annotations of the optic disc and fovea were obtained by SEVEN independent  OPHTHALMOLOGISTS from Zhongshan Ophthalmic Center, Sun Yat-Sen University, China. The reference standard for the segmentation task was created from the seven annotations, which were merged into single annotation by another SENIOR SPECIALIST. It is stored as a BMP image with the same size as the corresponding fundus image with the following labels:

0: Optic Disc (Black color)
255: Others (White color)


Image w Annotations

Annotation Masks

Data: Released on Nov 21 

  • The segmentation results should be provided in a “disc segmentation” folder, as one image per test image, with the segmented pixels labeled in the same way as in the reference standard (bmp (8-bit) files with 0: optic disc, 255: elsewhere). Please, make sure that your submitted segmentation files are named according to the original image names and with the same extension.
  • The localization results should be provided in a single CSV file, named “fovea_location_results.csv”, with the first column corresponding to the filename of the test fundus image (including the extension “.jpg”), the second column containing the X-coordinate and the third column containing the Y-coordinate. Please, make sure that your submitted segmentation files are named according to the original image names and with the same extension.

Task 3: Detection & Segmentation of lesions (drusen, exudate, hemorrhage, scar, and others) from fundus images. 

Description: Four typical kinds of lesions related to AMD are annotated on each image. (more details will be added soon)

Image w Lesion Annotations


: Released by Jan 1, 2019 



Images and AMD labels (released on Oct 20)

-Link for Mainland China PWD: mb36

Disc and fovea annotations (released on Nov 21)

-Link for Mainland China PWD: km0n

Lesions annotations  (released on Jan 1)

-Link for Mainland China (released on Jan 1)


 Images ( to be released on Jan 31)





Evaluation Framework

This challenge evaluates the performance of the algorithms for 3 types of general tasks: (1) binary classification, (2) localization, and (3) segmentation. More details will be added before results submission open. 

Leaderboards to be open by Jan 15 (Partially - Task 1&2)


Please email redkisses121@gmail for any inquiry.