Computer Aided Diagnosis for Early Detection of

Glaucoma using Convolutional Neural Network

(CNN)
Yunendah Nur Fu’adah* , Sofia Sa’idah, Inung Wijayanto,
Nur Ibrahim, Syamsul Rizal, Rita Magdalena
School of Electrical Engineering, Telkom University, Bandung, Indonesia
E-mail: * [email protected]

Abstract. Glaucoma is an eye disease caused by an increase in eye pressure resulting in

damage to the optic nerve and can cause blindness. An ophthalmologist makes the diagnosis of
glaucoma by analyzing the fundus images that show the retinal structure. This manual diagnosis
requires years of expertise and prone to error. Previous studies have designed a computer-aided
diagnose (CAD) system based on Convolutional Neural Network (CNN) to detect glaucoma and
showed promising results. This study proposes the CNN method that consists of three hidden
layers that use 3x3 of the filter size with 16, 32, 64 output channels respectively, a fully connected
layer, and sigmoid activation. The experiment is conducted using the RIMONE R2 fundus
images dataset to classify normal and glaucoma conditions. From 455 fundus images, 75% are
used as the training data, while the rest is used as the validation data. From the experiment,
this study outperforms other previous studies by achieving 91% of accuracy. The glaucoma
system detection that has been developed in this research, can be helpful for ophthalmologists
to establish an initial diagnosis of glaucoma that can reduce the harmful effects of glaucoma.

1. Introduction
Glaucoma is a multifactorial neurodegenerative disease, which reduces human vision and may
cause blindness. The main cause of glaucoma is the Intra Ocular Pressure (IOP). This condition
can damage the optic nerve, which made the brain does not get image information from light
receptors [1]. The retina and optic nerve changes often occur without symptoms and are not
detected by diagnostic tests [2]. Early examination and medical treatment suggested by an
ophthalmologist can help reduce the risk of glaucoma [1]. Based on data from the WHO,
glaucoma is one of the major causes of blindness globally after cataracts. More than 82% of all
blind people are aged 50 years and over. In 2013, the number of people (40-80 years old) with
glaucoma was 64.3 million and continued to increase until 2020 [3].
Ophthalmologists diagnose Glaucoma by analyzing the retinal structures. They are followed
by analyzing ocular parameters such as Cup to Disk Ratio (CDR) and very high Rim to Disk
Ratio (RDR). This requires expertise and accuracy. In diagnosing Glaucoma, ophthalmologists
use several expensive devices such as the Heidelberg Retinal Tomography (HRT), Optical
Coherence Tomography (OCT), Including Confocal Scanning Laser Ophthalmoscopy (CSLO)
and relatively limited fundus imaging [4]. Several studies based on fundus image processing
have developed for glaucoma detection. Previous studies mentioned that glaucoma disease is
detected by localizing and segmenting fundus images to obtain the Optic Nerve Head (ONH)
sections. The features are extracted using statistical measurement and classified using K-Nearest
Neighbor (K-NN). The method achieves 95.24% accuracy, which tested in a private data set [5].
Ramani et al. [6] did a segmentation process is performed to calculate the Cup to Disc Ratio
(CDR) based on image processing. The study obtains an accuracy of 98.7% using a private
dataset. In study [7], the feature extraction process is performed using the Empirical Wavelet
Transform method, and the Support Vector Machine (SVM) classification method. The method
is tested in a private dataset combined with the RIM-ONE dataset. They achieved 98.33%
accuracy for the private data set and 81.32% for the RIM-ONE dataset.
In some previous studies, the researches carried out several important stages in designing a
glaucoma detection system, such as the preprocessing and segmentation process that must be
accurate, proper feature selection and optimization at the classification stage which greatly affect
the system performance. Some studies are starting to use the Convolutional Neural Network
(CNN) as the main method. CNN has the advantage that it can learn the deterministic features
of raw data directly and provide a promising system performance in detecting Glaucoma. Studies
by Memon et al. [8] and Bajwa et al. [9] are the examples of study which use CNN as the main
method for detecting Glaucoma. They achieve an accuracy of 85% [8] and 87.4% [9], which
tested in RIM-ONE [8] and ORIGA [9] dataset. In the previous studies, CNN model was
designed using more than three hidden layers and used softmax activation. In this research,
CNN model design will be done using three hidden layers and sigmoid activation, which is more
effective for classifying two conditions such as normal and glaucoma.

2. Method
This study designed a detection system to classify glaucoma and normal condition based on
image processing. The system used the CNN method with three hidden layers, which used
3 × 3 filter size of each layers and 16, 32, 64 output channels respectively. Furthermore, fully
connected layers and sigmoid activation used to classify normal and glaucoma conditions. In
general, the CNN model proposed in this study is shown in Figure 1.

Figure 1. The proposed system model of CNN for glaucoma detection

2.1. Dataset
This study uses RIM-ONE R2 dataset. The dataset has 455 fundus images, consist of 255 fundus
images in normal condition and 200 fundus images in glaucoma condition. The distribution of
training data and validation data in this study are 75% and 25%, respectively. Thus, the training
data used are 341 fundus images, while the validation data are 114 fundus images. The data is
then resized with a size of 64 × 64 to be processed at the next stage.
2.2. Convolutional Neural Network
Convolutional Neural Network (CNN) is one of the Deep Neural Networks, which implemented
for image recognition [11]. CNN is a development of the Multilayer Perceptron (MLP). MLP
accepts one-dimensional input data and propagates the data on the network to produce output.
The CNN data propagated on the network are two-dimensional [12]. Thus, it can only be used
on data that has a two-dimensional structure such as image data. In general, CNN consists of
a Feature Extraction Layer and Classification Layer, which shown by the CNN architecture in
Figure 2.

Figure 2. The architecture of CNN

The feature extraction layer consists of convolutional layers and pooling layers. Convolutional
layers convert images with convolution processes to produce feature maps that show the original
image’s unique characteristics. Convolutional layers operate differently from other neural
network layers that use connection weights. The convolutional layers use convolutional filters to
produce a feature map [11].
Furthermore, the convolutional layer’s activation process uses Rectified Linear Units (ReLU)
activation to increase the training stage on the neural network so it can minimize errors and
saturation. In this study, the ReLU activation function is used at each hidden layer of the neural
network. The ReLU activation function is shown in (1)[13].

x, x > 0
f (x) = (1)
0, x ≤ 0
The ReLu activation function changes the negative pixel value on the image to 0 on the feature
map. The pooling layer on the feature extraction layer serves to reduce the size of the layer.
There are two types of pooling methods, namely maximum pooling, which sees the maximum
value and mean pooling that looks for the average value. An illustration of the pooling process
can be seen in Figure 3[11].

Figure 3. Illustrate of Pooling Layer. (a) The 4 × 4 pixel input image ; (b) Mean Pooling ; (c)
Maximum Pooling

Based on Figure 4, it can be seen that the Pooling Layer reduces the size of the image, the
image which was originally sized 4 × 4 to 2 × 2 without losing significant information. To avoid
overfitting and help generalize at the training stage, this study used a dropout at the last hidden
layer. The dropout consists of setting to zero the output of each hidden neuron according to the
probability value used. Hence when the neurons in CNN are dropped out, the neurons do not
contribute to the forward pass stage and do not participate in the back propagation stage [14].

2.3. Classification Layer

The classification stage consists of flattening stages to change feature maps, which are
multidimensional arrays into one-dimensional arrays [15]. The fully connected layers emit vectors
K, where K is the number of classes that can be predicted by the network. This study used two
classes, namely normal and glaucoma. In the last stage, the sigmoid activation function is used
following equation, which showed in (2) [16]. It can be seen that the sigmoid activation function
transform the input value x into the range of 0 − 1.
1
S(x) = (2)
1 + e−x
Table 1 shows the summary of the proposed CNN model as well as the output of each layer
that affects the image size.

Table 1. Details of CNN Model Proposed

Layer Type Output Shape Parameter Layer Type Output Shape Parameter
Input Image 64,64,3 0 Convolution 16,16,64 18496
Convolution 64,64,16 448 ReLU 16,16,64 0
ReLU 64,64,16 0 Max Pooling 8,8,64 0
Max Pooling 32,32,16 0 Dropout 8,8,64 0
Convolution 32,32,32 4640 Flatten 4096 0
ReLU 32,32,32 0 Dense 2 8194
Max Pooling 16,16,32 0 Sigmoid 2 0

3. System Performance
This study uses accuracy, recall, precision, and f1 scores to measure the performance. The
calculation is shown in (3)(4)(5) and (6) [9]. True Positive (TP) shows the exact glaucoma data
detected as glaucoma, True Negative (TN) shows normal detected as normal data, False Positive
(FP) shows glaucoma detected normal data while False Negative (FN) shows normal detected
glaucoma data.
TP + TN
Accuracy = (3)
TP + FP + TN + FN
TP
Recall = (4)
TP + FN
TP
P recision = (5)
TP + FP
recall × precision
F 1score = 2 × (6)
recall + precision

4. Result and Discussion

The RIM-ONE R2 dataset used in this study consist of 455 fundus images for glaucoma and
normal conditions. The number of training data used is 341 fundus images. On the other
hand, the amount of test data used is 114 fundus images for 67 normal condition images
and 47 glaucoma conditions. For glaucoma and normal conditions, fundus images are trained
using the CNN model with Adam optimizer learning rate 0.001, and loss binary cross-entropy.
Performance parameters measured in this study are accuracy, recall, precision, F1 score, and
loss. The model of accuracy and the loss of the proposed model is shown in figure 4.

Figure 4. (a) Accuracy of the proposed model; (b) Loss of the proposed model

Based on the experiment conducted in this study, the accuracy increases for each iteration
(epoch), and the difference in accuracy between the training data and the validation data
accuracy is not much different, as shown in figure 4 (a). Based on these results, it can be
concluded that there is no overfitting of the system designed using the proposed model. Figure
4 (b) shows a decrease in the value of the loss at each iteration (epoch) with a value that is
not much difference between the value of loss training and loss validation. It can be concluded
that learning errors that occur both for training data and test data achieved are minimum so
that the model can recognize normal conditions and glaucoma conditions with the best accuracy
performance of 91.22%, and loss of 0.1758.

Figure 5. Confusion matrix of validation data

Table 2. Performance of CNN Model Proposed

Class Precision Recall F1-Score No of Images
Glaucoma 0.91 0.94 0.92 65
Normal 0.91 0.88 0.9 49
Total 0.91 0.91 0.91 114

Based on the Confusion Matrix shown in figure 5 , it can be seen that from 114 validation
data used, 104 data are successfully classified according to their class. Other parameters used
to evaluate system performance are precision, recall, and F1-score, which have a range of values
from 0 to 1 (a value of 1 indicates no error). Based on the data shown in Table 2, the value
of system performance parameters is closed to 1. This condition shows that the CNN model
can classify glaucoma and normal conditions with high accuracy and minimal missclassification.
Compared to some previous studies, which used RIM ONE data set and Convolution Neural
Network with softmax activation as the main method to classify normal and glaucoma conditions,
the proposed CNN method with sigmoid activation outperforms the previous studies. The
sigmoid activation that used in the classification layer has proven to be more precise in classifying
two conditions.

5. Conclusion
In this research, a computer-aided diagnose system for the early detection of glaucoma based
on digital image processing is designed. The CNN model used in this study consists of three
hidden layers. Each of them is using 3 × 3 filter sizes with 16, 32, and 64 channel outputs, a
fully connected layer, and sigmoid activation. The experiment showed that the proposed CNN
model is able to classify raw fundus image datasets directly into glaucoma conditions and normal
conditions with an accuracy of 91.22%, loss of 0.1758, and the value of precision, recall, an f1-
score average of each amounted to 0.91. In further research, a system for the classification of
glaucoma conditions can be developed based on its severity.

References
[1] Ahmad H, Yamin A, Shakeel A, Gillani S O and Ansari U 2014 Detection of glaucoma using retinal fundus
images 2014 Int. Conf. Robot. Emerg. Allied Technol. Eng. iCREATE 2014 - Proc. 321–4
[2] Carrillo J, Bautista L, Villamizar J, Rueda J, Sanchez M and Rueda D 2019 Glaucoma Detection Using
Fundus Images of the Eye 2019 22nd Symp. Image, Signal Process. Artif. Vision, STSIVA 2019 - Conf.
Proc. 1–4
[3] Nucci C, Osbornec N N, Bagetta G and Cerulli L 2008 Glaucoma: an open-window to neurodegeneration and
neuroprotection. Preface. Prog. Brain Res. 173 173
[4] Kim M, Han J C, Hyun S H, Janssens O, Van Hoecke S, Kee C and De Neve W 2019 Medinoid: Computer-
aided diagnosis and localization of glaucoma using deep learning Appl. Sci. 9 2357–62
[5] Septiarini A, Khairina D M, Kridalaksana A H and Hamdani H 2018 Automatic glaucoma detection method
applying a statistical approach to fundus images Healthc. Inform. Res. 24 53–60
[6] Geetha Ramani R and Dhanapackiam C 2014 Automatic Localization and Segmentation of Optic Disc in
Retinal Fundus Images through Image Processing Techniques International Conference on Recent Trends
in Information Technology Automatic
[7] Maheshwari S, Pachori R B and Acharya U R 2017 Automated Diagnosis of Glaucoma Using Empirical Wavelet
Transform and Correntropy Features Extracted from Fundus Images IEEE J. Biomed. Heal. Informatics 21
803–13
[8] Memon I, Ursani A A, Bohyo M A and Chandio R 2019 Automated Diagnosis of Glaucoma using Deep
Learning Architecture Eng. Sci. Technol. Int. Res. J. 3
[9] Bajwa M N, Malik M I, Siddiqui S A, Dengel A, Shafait F, Neumeier W and Ahmed S 2019 Correction to:
Two-stage framework for optic disc localization and glaucoma classification in retinal fundus images using
deep learning (BMC Medical Informatics and Decision Making (2019) 19 (136) DOI: 10.1186/s12911-019-
0842-8) BMC Med. Inform. Decis. Mak. 19 1–16
[10] Fumero F, Alayon S, Sanchez J L, Sigut J and Gonzalez-Hernandez M 2011 RIM-ONE: An open retinal
image database for optic nerve evaluation 2011 24th International Symposium on Computer-Based Medical
Systems (CBMS) (IEEE) pp 1–6
[11] Kim P 2017 MATLAB Deep Learning: With Machine Learning, Neural Networks and Artificial Intelligence
[12] Noeman A and Handayani D 2020 Detection of Mad Lazim Harfi Musyba Images Uses Convolutional Neural
Network IOP Conf. Ser. Mater. Sci. Eng. 771
[13] Agarap A F 2018 Deep Learning using Rectified Linear Units (ReLU) 2–8
[14] Wang L, Wen M and Wu Y 2019 Glaucoma Detection based on Deep Convolutional Neural Network Acta
Microsc. 28 331–8
[15] Yamashita R, Nishio M, Do R K G and Togashi K 2018 Convolutional neural networks: an overview and
application in radiology Insights Imaging 9 611–29
[16] Feng J and Lu S 2011 Performance Analysis of Various Activation Functions in Artificial Neural Networks
Int. J. Artif. Intell. Expert Syst. 1