{"id":15650040,"url":"https://github.com/koriavinash1/optic-disk-cup-segmentation","last_synced_at":"2025-08-16T12:34:25.290Z","repository":{"id":39912182,"uuid":"141914523","full_name":"koriavinash1/Optic-Disk-Cup-Segmentation","owner":"koriavinash1","description":"Optic Disc and Optic Cup Segmentation using 57 layered deep convolutional neural network","archived":false,"fork":false,"pushed_at":"2018-12-12T14:19:00.000Z","size":1142,"stargazers_count":52,"open_issues_count":2,"forks_count":23,"subscribers_count":4,"default_branch":"master","last_synced_at":"2025-04-19T07:47:02.359Z","etag":null,"topics":["ai","artificial-intelligence","cdr-prediction","deep-convolutional-neural-networks","deep-learning","fundus-image-analysis","glaucoma-detection","medical-image-analysis","opticdisk-segmentation","segmentation"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/koriavinash1.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2018-07-22T16:59:03.000Z","updated_at":"2025-01-23T10:49:36.000Z","dependencies_parsed_at":"2022-09-19T05:40:13.183Z","dependency_job_id":null,"html_url":"https://github.com/koriavinash1/Optic-Disk-Cup-Segmentation","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/koriavinash1%2FOptic-Disk-Cup-Segmentation","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/koriavinash1%2FOptic-Disk-Cup-Segmentation/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/koriavinash1%2FOptic-Disk-Cup-Segmentation/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/koriavinash1%2FOptic-Disk-Cup-Segmentation/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/koriavinash1","download_url":"https://codeload.github.com/koriavinash1/Optic-Disk-Cup-Segmentation/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":251749113,"owners_count":21637456,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai","artificial-intelligence","cdr-prediction","deep-convolutional-neural-networks","deep-learning","fundus-image-analysis","glaucoma-detection","medical-image-analysis","opticdisk-segmentation","segmentation"],"created_at":"2024-10-03T12:33:05.977Z","updated_at":"2025-04-30T17:21:16.905Z","avatar_url":"https://github.com/koriavinash1.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Optic-Disk-Cup-Segmentation and Glaucoma Screening\n\n## Introduction\n\nThis repository contains the implementation of convolutional neural network for optic disk and cup segmentation and glaucoma screening from given fundus images\n\n\u003chr\u003e\n\n# Segmentation Network\n\n## Preprocesing\n\nImages were cropped to nearest square size and re-sized to a dimension of (512, 512). The different lighting conditions and intensity variations among images across various databases were circumvented by perform-ing normalization of the histogram using Contrast Limited Adaptive HistogramEqualization (CLAHE). 2 different images were generated by varying parameters such as clip value \u0026 window level while performing CLAHE. Along with CLAHE, spatial co-ordinates information were also provided to thenetwork. This additional information aided in learning relative features (i.e. disklocation with respect to fovea)\n\n\u003chr\u003e\n\n## Network Architecture\n\n57 layered deep network was used for segmentation of optic disk and cup. Network architecture is illustrated in figure below...\n![pipeline](./images/segnet.png)\n\n\u003chr\u003e\n\n## Results\n\n### Model predictions\n\nMask generation used for reducing false positives predicted by network...\n![postprocessing](./images/maskgen.png)\n\n\n![prediction](./images/Selection_007.png)\nImage on left shows raw data and image on left shows model predictions...\n\n\u003chr\u003e\n\n# Classification Network\n\n## Preprocessing\n\nThe pixel level segmentation of the optic disk and\noptic cup was utilized to generate images of dimension (550, 550) centered around the optic disk. 6 different images were generated by varying parameters such as clip  value  \u0026  window  level  while  performing  CLAHE. \n\n## Network Architecture\n\nA  DenseNet201 \u0026  ResNet18 pre-trained  on natural images forms the ensemble. The hindmost layer in the network i.e. the classification layer was modified to have 2 neurons. An additional convolutional layer was appended before both the pre-trained models  to  convert  out  21  channel  input  to  3  channels.  To  make  the  network images accept inputs of variable dimension, the global average pooling layer was substituted with an adaptive average pooling layer.\n\n## Results\n\nThe proposed classification network achieved a sensitivity of 0.75 at a specificity of 0.85 and 0.856 area under the ROC curve.\n\n## How to use?\n\n~~~~\n\ngit clone https://github.com/koriavinash1/Optic-Disk-Cup-Segmentation.git\ncd Optic-Disk-Cup-Segmentation\npip install -r requirements.txt\n\n~~~~\n\n\u003chr\u003e\n\n## Folder structure\n\n\u003e ./src consists all source codes\n\n\u003e \u003e ./src/segmentation code for all segmentation work\n\n\u003e \u003e ./src/classification code for glaucoma screening\n\n\u003e \u003e Tune parameters and run Main.py for executing task\n\n\u003chr\u003e\n\n## Publication\nOur paper is available on arXiv(https://arxiv.org/pdf/1809.05216.pdf)\n\nPlease cite with the following Bibtex code:\n~~~\n@article{agrawal2018enhanced,\n  title={Enhanced Optic Disk and Cup Segmentation with Glaucoma Screening from Fundus Images using Position encoded CNNs},\n  author={Agrawal, Vismay and Kori, Avinash and Alex, Varghese and Krishnamurthi, Ganapathy},\n  journal={arXiv preprint arXiv:1809.05216},\n  year={2018}\n}\n~~~\n\nIf any comments or issues, pull requests/issues are Welcomed....\n\nThankyou\n\n\n### Contact \n\n* Avinash Kori (avinashgkori@smail.iitm.ac.in)\n* Vismay Agrawal (vismay.iitm@gmail.com)\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkoriavinash1%2Foptic-disk-cup-segmentation","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkoriavinash1%2Foptic-disk-cup-segmentation","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkoriavinash1%2Foptic-disk-cup-segmentation/lists"}