Scribd
Upload
9 views2 pages

Abstract

The document presents a research project on the detection and classification of ocular diseases using multimodal deep learning techniques, focusing on diseases like glaucoma, cataracts, and diabetic retinopathy. It compares the performance of various deep learning models, ultimately favoring VGG-19 for its superior accuracy in classifying ocular diseases. The project aims to develop an Android application to enhance the efficiency and cost-effectiveness of ocular disease detection.

Uploaded by

Rimsha Tayyab
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views2 pages

Abstract

The document presents a research project on the detection and classification of ocular diseases using multimodal deep learning techniques, focusing on diseases like glaucoma, cataracts, and diabetic retinopathy. It compares the performance of various deep learning models, ultimately favoring VGG-19 for its superior accuracy in classifying ocular diseases. The project aims to develop an Android application to enhance the efficiency and cost-effectiveness of ocular disease detection.

Uploaded by

Rimsha Tayyab
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

OCUNET: Multimodal Deep Learning for Efficient Ocular

Disease Detection and Classification

Submitted by
Sara Fareed

AU-02-05-10228

Rimsha Tayyab

AU-02-05-10251

Hadiqa Naeem

AU-02-05-10253

Muqadus

AU-02-05-10243

Supervised by
Ms .Sidra Kiyani

Submitted in partial fulfillment for the degree of Bachelor of Computer Science.

FAZAI BILQUIS COLLEGE OF EDUCATION FOR WOMEN, PAF NUR KHAN


RAWALPINDI
(Aug, 2024)
ABSTRACT
Ocular diseases are major global health concern, affecting millions of individuals and leading to
varying degrees of visual impairment and blindness. Eye Flu commonly known as Pink eye is
also the type of ocular disease, which is rapidly spreading among people. This research project
looks into the prevalence, causes and treatment strategies for major ocular diseases, including
glaucoma, cataracts, and diabetic retinopathy. It presents a systematic use of multimodal deep
learning techniques, including Convolutional Neural Networks (CNN), along with image
processing methodologies, containing two datasets EDC and conjunctivitis, to classify and detect
ocular diseases. Conventionally, in previous researches, the cutting-edge image classification
algorithms such as Efficient-Net, MobileNetV2, and VGG-16 techniques were used on the EDC
dataset. In this research, we explored deep learning multimodal techniques, including Resnet-152,
VGG-19, Single Shot Detection (SSD) and Particle swarm optimization (PSO). A comparative
analysis will then be performed between the results obtained from the previous research and our
research resulting in our method outperforming the conventional methods. By exploring Deep
learning (CNN) models such as VGG-19 with an accuracy of 92%. The ResNet-152 with an
accuracy of 91%, the PSO achieved best fit line is 0.23 in Conjunctivitis dataset. In EDC dataset,
the VGG-19 with an accuracy of 90%, ResNet-152 with an accuracy of 89%, the PSO achieved
best fit line is 0.24. After comparing two models, ResNet-152 didn’t yield significant results,
that’s why VGG-19 is used as a final model on both datasets. The project aims to leverage a
hybrid approach that utilizes deep learning and create an android application for efficient
detection of ocular diseases. This would revolutionize the ocular disease detection domain by
utilizing cutting-edge technologies and would optimize the cost of eye disease detection. This
application serves as a starting point for further research in the eye disease detection as with the
advancement in artificial intelligence, more image processing deep-learning-based models can be
used for efficient eye disease detection
Keywords: Artificial intelligence (AI), Convolutional Neural Networks (CNN), Deep
learning, Image processing, Ocular Disease Detection, Single Shot Detection (SDD),
Particle swarm optimization (PSO)

You might also like