Papers › Deep Residual Network based Automatic Image Grading for Diabetic Macular Edema

Deep Residual Network based Automatic Image Grading for Diabetic Macular Edema

1 Jul 2018Engineering in Medicine and Biology Society (EMBC), 2018 40th Annual International Conference of the IEEE 2018 7archive 2025-07-28

Santhosh Kumar Sukumar, Kamalakkannan Ravi, Supriti Mulay, Keerthi Ram, Mohanasankar Sivaprakasam

Diabetic Macular Edema (DME) is an advanced symptom of diabetic retinopathy that affects central vision of diabetes patients. An automated system for early detection of DME symptom has been proposed herein to elude vision impairment and assist in effective treatment. Transfer learning based on Deep Residual Networks (ResNets) which has proven to be a very successful model in many image classification applications and is used in the proposed system for automatic grading of DME images. Validation of the developed system on Indian Diabetic Retinopathy Image Dataset (IDRID 2018) results in 86.56 % detection accuracy.

PaperPDFCode

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Fovea DetectionImage ClassificationMedical Image AnalysisMedical Image ClassificationOptic Disc DetectionTransfer Learningimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Medical Image Classification IDRiD ResNet-152 Accuracy (% ) 86.56 #2 of 2 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections