GET THE APP

Comparison of VGG-16 and Few-shot Learning Using a Small-Siz | 108694

Journal of Research in Medical and Dental Science
eISSN No. 2347-2367 pISSN No. 2347-2545

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Comparison of VGG-16 and Few-shot Learning Using a Small-Sized Dataset of OCT Images for Neurodegeneration in Epilepsy

Author(s): Senol Dane

Abstract

Due mostly to significant improvements in efficacy, deep learning has recently gained a lot of interest. Although there has been progress, there is still much potential for development, particularly when dealing with use cases that have low data availability, as is frequently the case in the field of medical image analysis. In this study, we present a method for detecting neurodegeneration in epilepsy early on in OCT images using a minimal amount of training data. In particular, we developed a predictive model based on convolutional neural network architecture, leveraging few shots learning. Our experimental results show that our predictive model has an accuracy of 88.1% and can obtain higher levels of effectiveness than VGG-16 with an accuracy of 65.4%.

Share this article

http://sacs17.amberton.edu/