Abstract
Age-related Macular Degeneration (AMD) is the leading medical condition causing blindness for the older population due to damaged macular. This paper compares the performance of two models with differing numbers of Convolutional Neural Network (CNN) layers in Deep Learning in detecting AMD through Optical Coherence Tomography (OCT) images of the human retina. This will potentially be able to provide assistance to clinical specialists for quicker diagnosis and earlier intervention. ResNet50 with 50 CNN layers and a custom model (OwnModel) with five CNN layers were tested using fourfold cross-validation and on a separate data set. Results were then collected in two stages from the same pre-labelled public data set: 1000 images for Stage 1 on a personal machine and another 5000 for Stage 2 on the mentor’s machine. ResNet50 produced more satisfactory results than OwnModel in Stage 2 with accuracies of 98.09% ± 1.00% and 96.62% ± 2.00%, respectively, after cross-validation. Similarly, ResNet50 yielded a higher accuracy at 96.90 ± 2.00% than OwnModel’s at 90.00 ± 2.00% through testing on a separate data set. A higher number of CNN layers improved the machine’s performance in detecting AMD. Analysis of results has proven that AI has great potential to act as future diagnostic aids based on high accuracy rates comparable to their human counterparts. This reduces the workload of clinicians and allows more time for patient care, paving the way for improved patient outcomes.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.