| Idea Name: |
AI-based Ophthalmology |
| Slogan: |
Bringing Eye Care To Masses |
| Supervisor Name: |
Dr Usman Akram |
| Supervisor Designation: |
Assistant Professor |
| Supervisor School: |
CEME |
| Supervisor Department: |
Department of Computer Engineering |
| Contact number: |
Off: 051-9247550 (Ext 4170) Cell: 0333-6913920 |
| Email ID: |
[email protected] |
| Abstract: |
| We are developing a 360° AI based Ophthalmology Grading & Analysis tool that will automate the long doctor-patient cycle in Ophthalmology. It will use AI & Cloud to provide quick & accurate solutions. |
| What is the unmet need in society that your idea will fulfill ? |
| The conventional tools used in Ophthalmology are too time consuming and require a lot of expertise which can be hard to come by. The tool we are developing will provide AI based quick and accurate diagnosis of widespread problems like Diabetic Retinopathy and Glaucoma |
| Who needs it ? How many would benefit ? |
| There are about 347 mil+ diabetic patients worldwide & most of them have a high risk of developing Diabetic Retinopathy around the world. This project can be deployed at clinics, hospitals, ophthalmology centers & can be even used for self-diagnosis in some cases. It will be a lot quick & accurate |
| How will the solution works |
| The first part is the fundus photography. For that we introduce a cheap apparatus to which a smartphone can be mounted. The fundus images are then sent to a cloud-based AI model where grading of severity of the diseases takes place. Along with that, the quality of each fundus photo is examined to generate a montage. The deep learning-based models will segment out vessels, hemorrhages, micro aneurysm, exudates, OD & macula. The doctors will see the results on a web-based dashboard & give feedback |
| Who are your competitors ? How is your solution different |
| Google recently deployed an AI based ophthalmology grader in Africa which failed just after a few weeks due to sensitive quality check. Some other ones are RIVAS, ImageJ, Mona health and Peek Solutions. In our case, we are making sure that our platform is faster, more accurate and flexible. |
| Status: |
new |
| Entry Date & Time: |
2021-01-09 (1218) |
