Rethinopathy

The research is driven by a set of particular goals, such as a thorough analysis of the major molecular pathways implicated in diabetic retinopathy, the discovery of potential biomarkers for early detection, and the investigation of novel therapeutic approaches. Through these goals, the study hopes to provide practical information that can guide clinical practise, public health programmes, and ongoing research projects aimed at tackling the complex web of diabetes complications, with a focus on maintaining and restoring visual acuity. We recognise the necessity of collaboration between doctors, researchers, and policymakers as we begin this investigative journey in order to solve the many issues raised by diabetes and its effects on ocular health.

Dataset Information

• Factors:
  
  Vision problems in individuals with diabetes are influenced by a interplay of factors. Elevated blood sugar levels damage eye blood vessels in diabetic retinopathy, worsened by age and diabetes duration. Dry eyes are common and worsened by concurrent health issues like hypertension. Some diabetes medications increase the risk of cataracts and glaucoma. Genetics can predispose certain individuals to eye issues. Lifestyle choices, such as diet and exercise, impact diabetes control and eye.
  

You can find the Dataset:- Here

Models Applied

• Logistic Regression

• Decision Trees

• Gradeient Boosting Classifier

• SVM

• Ada-Boost Classifier

• KNN

• Ensembling

1. Model One:

2. Final Model:

Conclusion

We compare and analyse various machine learning models and get the best performance by ensembling Logistic Regression, Voting Classifier and SVM with accuracy of 92 %.

Authors

• Astha Jha

• Diti Jain

• Shailza Thakur

• Dev Mehta