Deep Learning Approach for Automatic Microaneurysms Detection

Mateen, Muhammad; Malik, Tauqeer Safdar; Hayat, Shaukat; Hameed, Musab; Sun, Song; Wen, Junhao

doi:10.3390/s22020542

Cited by 18 publications

(7 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, it was observed that our model's classification performance improved when the extracted features were subjected to a frequency and spatial channel attention approach. Mateen et al, 25 and Quellec et al, 26 along with the idea of channel attention, the proposed framework was further optimized using the back propagation algorithm by updating weights and bias. Feed‐forward, backward, combining all the values, and calculating the updated weight value are the steps in backpropagation.…”

Section: Resultsmentioning

confidence: 99%

Automatic detection of microaneurysms using a novel segmentation algorithm based on deep learning techniques

Birlin

Divya

Livingston³

2023

Computational Intelligence

View full text Add to dashboard Cite

Microaneurysms is the first stage of diabetic retinopathy (DR) and it plays a vital role in the computerized diagnosis. However, it is difficult to automatically detect microaneurysms in fundus images due to the complicated background and various illumination reasons. The motivation behind this, is the number of increases in diabetic patients is very large when compared with the number of ophthalmologists. The FSCA‐UNET (Frequency Spatial Channel Attention UNET) segmentation model, is proposed and it is an improvement over UNET. We first use the frequency channel attention mechanism to analyze the features that were extracted from the first stage of the convolution layer, and we obtain good results. Then, we included a spatial attention map with frequency attention, also known as FSCA‐UNET, which makes use of inter‐spatial connections between features. Our deep neural model with an encoder‐decoder structure termed FSCA‐UNET produced more accurate results. Our novel algorithm outdated the performance measures of the existing segmentation algorithms. The proposed segmentation algorithm was trained and tested on Indian Diabetic Retinopathy Image Dataset (IDRiD), and E‐ophtha Dataset and we got promising results in terms of sensitivity, specificity, dice coefficient, precision, F1 score, and accuracy.

show abstract

Section: Resultsmentioning

confidence: 99%

Automatic detection of microaneurysms using a novel segmentation algorithm based on deep learning techniques

Birlin

Divya

Livingston³

2023

Computational Intelligence

View full text Add to dashboard Cite

show abstract

“…Therefore, it provides high f-measure and low specificity. Mateen et al (2022) have presented deep learning approach for automatic MA detection. The suggested method uses VGG-19 and Inception-v3 pre-trained CNN models in a hybrid feature embedding approach to perform early identification of MAs.…”

Section: Related Workmentioning

confidence: 99%

Micro aneurysm detection using optimizedresidual‐basedtemporal attentionConvolutional Neural NetworkwithInception‐V3transfer learning

Alotaibi

2024

Microscopy Res & Technique

View full text Add to dashboard Cite

In this manuscript micro aneurysm detection using residual‐based temporal attention Convolutional Neural Network (CNN) with Inception‐V3 transfer learning optimized with equilibrium optimization algorithm (MA‐RTCNN‐Inception V3‐EOA) is proposed. The proposed research work contains four phases: (1) pre‐processing, (2) segmentation, (3) post‐processing, and (4) classification. At first, guided box filtering for contrast enhancement and background exclusion of input image. The proposed MA‐RTCNN‐Inception V3‐EOA based classification framework is implemented in MATLAB using several performances evaluating metrics like precision, sensitivity, f‐measure, specificity, accuracy, classification error rate, and Matthews's correlation coefficient and RoC analysis. The experimental outcome demonstrates that the proposed method provides 23.56%, 14.99%, and 21.37% higher accuracy and 31.26%, 57.69%, and 21.14% minimum classification error rate compared to existing methods, such as diabetic retinopathy identification utilizing prognosis of micro aneurysm and early diagnosis for non‐proliferative diabetic retinopathy depending on deep learning approaches (DRD‐CNN‐NPDR), a magnified adaptive feature pyramid network for automatic micro aneurysms identification (MAFPN‐AMD‐MAFP‐Net) respectively.Research Highlights Micro aneurysm detection using residual‐based temporal attention Convolutional Neural Network (CNN) is proposed. To get rid of the retina background, guided box filtering is applied. COAT is used for segmenting the images into smaller parts RTCNN is used for accurate micro aneurysms disease classification. RT‐CNN algorithm successfully identifies the micro aneurysms using EOA.

show abstract

“…Computer vision approaches for image processing help automation specialists create computer-assisted diagnostic systems (CADSs) [50] that improve diagnosis accuracy in contemporary ophthalmology [51][52] [53]. Standard processes must therefore be developed in order to ease into conventional ways because computer algorithms and methods are used in imaging analysis.…”

Section: B Related Workmentioning

confidence: 99%

Classification of Diabetic Retinopathy Images through Deep Learning Models - Color Channel Approach: A Review

Salih,

Abdulazeez

2024

ijcs

View full text Add to dashboard Cite

On a global scale diabetic retinopathy, or DR, is the most common cause of vision loss. Blindness can be prevented with prompt treatment and early identification with retinal screening. Automated analysis of fundus imagery is growing prominently as a means of increasing screening efficiency, thanks to the development of deep learning. This work focuses on deep learning methods for automatic DR severity grading using color channel information. First, we give some basic information on the etiology and color features of DR lesions. Next, a novel support for deep learning technique that use unprocessed color photos as input for comprehensive feature learning. A review is mentioned on color space encodings, data augmentation methods. A summary of the evaluation parameters and public databases that were utilized to benchmark DR techniques are provided. The objective of how color channel information in retinal pictures can be efficiently utilized by deep learning models for automated DR screening has been discussed with statistical support.

show abstract

Deep Learning Approach for Automatic Microaneurysms Detection

Cited by 18 publications

References 40 publications

Automatic detection of microaneurysms using a novel segmentation algorithm based on deep learning techniques

Automatic detection of microaneurysms using a novel segmentation algorithm based on deep learning techniques

Micro aneurysm detection using optimizedresidual‐basedtemporal attentionConvolutional Neural NetworkwithInception‐V3transfer learning

Classification of Diabetic Retinopathy Images through Deep Learning Models - Color Channel Approach: A Review

Contact Info

Product

Resources

About