Prediction of Diabetic using Retinopathy Diabetic retinopathy is a diabetes-related eye disease that is caused due to the damage of neurons present in the blood vessels of the eye. Due to high sugar levels in the blood, a patient can attain diabetes. When diabetes in a patient spread to the region of the eye, this disease is mentioned as Diabetic Retinopathy. This disease can lead to the progress of uncharacteristic blood vessels in the retina and also leakage and swelling of blood vessels. Symptoms of diabetic retinopathy include blurred vision, undulating vision, diminished color vision, and spots or dusky filaments. The main goal of this project is to categorize the patients as having diabetic retinopathy or not.
- 1. Data Collection: Dataset is collected by performing microaneurysms detection test, and pre-screening test to obtain suitable details for the data visualization. The data obtained from the patients contain various features such as retinal abnormality, Euclidean distance, and the diameter of the optic disc.
- 2. Data Description: This module uses the dataset to predict whether the patient has diabetic retinopathy or not in the form of binary values. It passes the dataset by binary numbers. “1” means the patient has diabetic retinopathy and “0” means the absence of the disease.
- 3. Data Visualization: Obtained data is organized for comparison with the features having major differences. Histograms are used as they allow you to easily see where a large and a little amount of the data can be found.
- 4. Split Dataset: Data is separated into training and testing sets. Most of the data is used for training, and a smaller portion of the data is used for testing. The training set contains known data from the patient's record. The testing set contains the data regarding the stages in Diabetic Retinopathy.
- 5. Model Prediction: By using the trained and tested dataset we predict the stage for the patients detected with Diabetic retinopathy. This prediction is processed by applying the SVM algorithm.
1. SVM Algorithm: Dataset is divided into a training set and a testing set. SVM algorithm is applied to the input dataset for the characterization task.
supported operating system
- 1. window
- 2. Linux
- 3. ubuntu
Cost and technology
|Project Details contact - +91 9964716807
||Prediction of Diabetic Retinopathy Using SVM algorithm using ML
||Final year project.
||₹ 12000 INR
||₹ 2000 INR
||₹ 10000 INR
||charges extra cost for documentation for all project
We have tried to construct an ensemble to predict if a patient has diabetic retinopathy using the datasets and blood sample results of the individual. After training and testing the model the accuracy we get is quite similar. Despite the shortcomings in reaching good performance results, this work provides a means to make use of and test machine learning algorithms such as SVM and try to arrive at ensemble models that would outperform individual learners.
It also allows exploring a little feature selection, feature generation, parameter selection, and ensemble selection problems and experiences the constraints in computation time when looking for possible candidate models in high combinatorial spaces, even for a small dataset as the one used. The structure of our research has been built in such a way that with proper dataset and minor alternation it can work to classify the disease in any number of categories.