A collective machine learning and deep learning prototypical to expect diabetic retinopathy

Srisuma, V. and Gurumoorthy, Sasikumar and S. M. M Yassin, S. M. Warusia Mohamed and MacHerla, Sivudu and Naik, Ashitha V (2023) A collective machine learning and deep learning prototypical to expect diabetic retinopathy. In: 2023 IEEE International Conference on Integrated Circuits and Communication Systems, ICICACS 2023, 24 February 2023 through 25 February 2023, Virtual, Online.

[img] Text
A collective machine learning and deep learning prototypical to expect diabetic retinopathy.pdf
Restricted to Repository staff only

Download (423kB)

Abstract

Type 2 diabetes mellitus (T2DM) is a degenerative condition. Beta cell dysfunction worsens with disease progression, leading to elevated blood glucose levels. Over time, untreated type 2 diabetes leads to a plethora of complications and eventually death. When type 2 diabetes (T2DM) is properly controlled, progression can be delayed or even stopped altogether, and in some cases, remission can even occur. Certain risk factors, such as those related to one's nutrition and level of physical activity, can be changed to influence one's prognosis and the rate at which their condition worsens. It is first vital to understand the elements that foretell the speed and direction of T2-DM advancement in order to build efficient intervention and management strategies. In this study, we focus on how to build an ensemble model to categorize people with heart problems. Accuracy for the proposed model is determined by summing the results of all of the learners, who each contribute to the overall accuracy of the model. The dataset chosen for investigation is the Cleveland Heart Dataset acquired from UCI Machine learning repository. Because of the accuracy of the suggested model, heart illness can be diagnosed sooner, reducing the risk of serious consequences or even death. By comparing the results with those of recently announced methods from different researchers, it was observed that the created models offered superior accuracy by 87.5%, sensitivity by 97.3%, and specificity by 98%.

Item Type: Conference or Workshop Item (Keynote)
Uncontrolled Keywords: Diabetic Retinopathy, Machine Learning, T2DM, UCI
Divisions: Faculty of Information and Communication Technology
Depositing User: Anis Suraya Nordin
Date Deposited: 20 Sep 2024 16:15
Last Modified: 20 Sep 2024 16:15
URI: http://eprints.utem.edu.my/id/eprint/27905
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item