Skip to content

kaylaque/glauco-scan

Repository files navigation

Glauco-scan

Glauco-scan is part of our work in an internal event 'Hackathon Penelitian Teknik Biomedis 2021' held by Department of Electrical Engineering and Technological Information, University of Gadjah Mada. Our team called 'GumiGummy' consisted of myself (Kayla Queenazima S), Arin Yusianti, Hilmi Arisanti R, and Priscillia Sonia Putri L from Biomedical Engineering, UGM. We built Glauco-Scan as a simple and goto web-app to easily process OCT scan result and show the segmentation result of cup and disc, the Cup to Disc Ratio (CDR), prediction of Glaucoma, and QR code result so user can save easily the results. We built this project in 2 weeks and helped by the mentors from the department.

Segmentation Process

Built in from the project of Seva, we use the same dataset as him but using Segmentation API to built the architecture of Unet with transfer learning of EfficientNet

Cup-to-Disc Ratio Calculation

The CDR calculation editted from Nupur Bhaisare's work

Prediction of Glaucoma

From the CDR result, based on dataset CDR in kaggle, we built algorithm to classify glaucoma using SVM

The model weights are here. The final webapp should be added some preprocessing of bounding box from segmentation of disc result, but not yet finished.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published