Influence of background preprocessing on the performance of deep learning retinal vessel detection.

Vascular diseases cause tortuosity in the retinal blood vessel's structure which can be identified using fundus images.The computerized retinal vessel tortuosity diagnosis is suffered from high time consuming and less accuracy due to the less effective retinal vessel segmentation algorithms.The success of retinal vessel tortuosity diagnosis depends on the performance quality of the vessel segmentation algorithm.Hence, this research proposes a new retinal vessel segmentation method, namely 'Retinal Vessel Segmentation using CLAHE and VJSTFD (RVSCV)'.The proposed segmentation method separates the vessel region from the background, tissue region, and Optic Disc (OD).It is constituted by two algorithms, viz.(i) vessel enhancement based on Contrast Limited Adaptive Histogram Equalization (CLAHE) and (ii) Vessel Joining algorithm based on vessel Skeletonization and Tetra vessel Flow Directions (VJSTFD).The CLAHE method, which is an existing method, enhances the vessel structure.The main contribution of this paper is the VJSTFD algorithm, which is a novel method to segment the vessel region from the enhanced fundus image.The VJSTFD algorithm makes a binary segmentation and joints the broken vessels in a novel approach through vessel skeletonization and tetra vessel flow directions-based vessel extension.The specialty of this approach is that minute vessels are also captured for segmentation even at the low contrast level because of the proposed preeminence vessel joining approach that is powered by tetra direction-based vessel searching.The CHASE_DB, RITE_DB, and STARE_DB databases have been involved in the testing process.These databases have retinal fundus images which clearly exposes the retinal vessel areas.Experimental results reveal that the proposed RVSCV method accomplishes the overall average F-score of 0.9580, along with the average segmentation accuracy of 96.73% in retinal vessel segmentation.The proposed method can be activated as a tool for ophthalmologists to extract the vessel course essentially.

The structure and appearance of the blood vessel network in retinal fundus images is an essential part of diagnosing various problems associated with the eyes, such as diabetes and hypertension. In this paper, an automatic retinal vessel segmentation method utilizing matched filter techniques coupled with an AdaBoost classifier is proposed. The fundus image is enhanced using morphological operations, the contrast is increased using contrast limited adaptive histogram equalization (CLAHE) method and the inhomogeneity is corrected using Retinex approach. Then, the blood vessels are enhanced using a combination of B-COSFIRE and Frangi matched filters. From this preprocessed image, different statistical features are computed on a pixel-wise basis and used in an AdaBoost classifier to extract the blood vessel network inside the image. Finally, the segmented images are postprocessed to remove the misclassified pixels and regions. The proposed method was validated using publicly accessible Digital Retinal Images for Vessel Extraction (DRIVE), Structured Analysis of the Retina (STARE) and Child Heart and Health Study in England (CHASE_DB1) datasets commonly used for determining the accuracy of retinal vessel segmentation methods. The accuracy of the proposed segmentation method was comparable to other state of the art methods while being very close to the manual segmentation provided by the second human observer with an average accuracy of 0.972, 0.951 and 0.948 in DRIVE, STARE and CHASE_DB1 datasets, respectively.

Diabetic Retinopathy (DR) has become nowadays a considerable world-wide threat due to increased growth of blind people at early ages. From the engineering viewpoint, the detection of DR pathologies (microaneurysms, hemorrhages and exudates) through computer vision techniques is of prime importance in medical assistance. Such methodologies outperform traditional screening of retinal color fundus images. Moreover, the identification of landmark features as the optic disk (OD), fovea and retinal vessels is a key pre-processing step to detect the aforementioned potential pathologies. In the same vein, this paper works with the well-known Convexity Shape Prior algorithm to segment the main anatomical structure of the retina, the OD. At first, some pre-processing techniques such as the Contrast Limited Adaptive Histogram Equalization (CLAHE) and Brightness Preserving Dynamic Fuzzy Histogram Equalization (BPDFHE) are applied to enhance the image contrast and eliminate the artifacts. Subsequently, several morphological operations are performed to improve the post-segmentation of the OD. Finally, blood vessels are extracted through a novel fusion of the average, median, Gaussian and Gabor wavelet filters.

In the recent era, the increasing diabetic ratio effects in producing complex retinal diseases. The early diagnosis will avoid the consequences of diabetic-related complications. So, there is an urge to develop a computerized model for identifying the retinopathy features caused due to diabetes while evading the false positives. The aim of this paper is to identify and segment the retinal features like blood vessels, optic disc in diabetic retinopathy (DR) fundus images in order to remove the non-diabetic retinopathy features which makes the detection of DR features (lesions) easier. And this is done by using a hybrid segmentation algorithm called BINI which combines the model of both binary and Niblack’s thresholding. Initially, the input image is standardized by resizing the image using bi-cubic interpolation area method. Then, the fundus image quality is enhanced using preprocessing techniques like green channel extraction, intensity channel extraction, median filtering, contrast limited adaptive histogram equalization and morphological operations. The preprocessed images are segmented to produce retinal features using the BINI algorithm. It is a novel method which hybrids both binary thresholding and Niblack’s thresholding. The performance of the segmentation methods is evaluated using the validation measures like Rand index (Ri) and Jaccard index (Ji), Precision (Pr), Recall (Rc) and F-Measure (Fm). The proposed method for segmenting the retinal features using BINI thresholding has given an accuracy of about 96.48%; it leads to an accuracy of 100% clear segmentation of the lesions of diabetic retinopathy images.

Breast cancer remains one of the most common cancers amongst women globally. Early detection is crucial for improving survival rates. While mammography is widely used and an effective imaging technique, it can sometimes yield false positive or false negatives. Mammogram interpretation is highly operator-dependent, introducing variability and the potential for diagnostic errors. Additionally, mammographic images have limitations, such as low contrast in breast tissue and overlapping structures that can obscure lesions or mimic abnormalities. These limitations can lead to unnecessary biopsies or delayed diagnosis. These challenges highlight the needs for advanced and data driven diagnostic tools to support and enhance current screening practices.</p> <p>In recent years, machine learning has emerged as a transformative force in the health care industry offering innovative solution to complex medical challenges. One of the most promising applications of this technology is in the early and accurate detection of breast cancer particularly through the mammography image analysis. Machine learning algorithms, especially deep learning models, have shown significant potential in identifying patterns in mammography that may indicate abnormal cases. This study utilizes the power of deep learning particularly Convolutional Neural Networks (CNNs) to enhance the classification of breast ultrasound images.</p> <p>This research proposes a novel two-level classification approach. At the first level classification, the system distinguishes normal from abnormal cases. Next, the second level classification, it further classifies abnormal cases into benign or malignant tumors. To address challenges such as limited and imbalanced datasets, the study incorporates comprehensive preprocessing techniques including contrast enhancement via Contrast Limited Adaptive Histogram Equalization (CLAHE) and region of interest (ROI). The preprocessing step were evaluated separately and in combination to assess their individual contributions. Additionally, data augmentation is utilized to improve model generalization. To further enhance performance and robustness, multiple CNN architectures are combined using an ensemble mechanism. Soft voting and hard voting leverage the strengths of diverse CNNs reducing individual model biases and improving overall accuracy.</p> <p>The main contributions of this research are: (1) the design of a two-level classification system (2) the utilization of multiple preprocessing techniques to improve image quality, (3) the application of voting ensemble learning to improve predictive stability and accuracy, and (4) an extensive evaluation on the BUSI dataset. This breaks down the complex diagnostic task into smaller, manageable steps to allow for more accurate comparisons, analysis, and improvements at each stage.</p> <p>The first preprocessing step involved the application of CLAHE. In the first level classification (normal vs abnormal), performance of various models has been evaluated. The models achieved an 82.05% accuracy for AlexNet, 84.62% for DenseNet, 85.26% for Inception, 82.69% for MobileNet, 82.05% for ResNet and 82.69% for VGG16. The accuracy of soft voting is 96.20% and accuracy of hard voting 92.09%. In the second level classification (benign vs malignant), the models achieved an 75.00% accuracy for AlexNet, 80.47% for DenseNet, 35.94% for Inception, 41.41% for MobileNet, 35.94% for ResNet and 82.81% for VGG16. The accuracy of soft voting is 90.84% and hard voting is 85.76%.</p> <p>The second preprocessing step involved the ROI was implemented. This step helped isolate the regions containing potential abnormalities, such as tumors, while filtering out irrelevant areas. After the application of ROI (normal vs abnormal), the system evaluated the performance of various models. In the first level classification (normal vs abnormal), the models achieved an 82.69% accuracy for AlexNet, 82.69% for DenseNet, 83.97% for Inception, 87.18% for MobileNet, 82.69% for ResNet and 82.05% for VGG16. The accuracy of soft voting is 100.00% and hard voting is 98.08%. In the second level classification (benign vs malignant), the system evaluated the performance of various models. The models achieved an 75.00% accuracy for AlexNet, 80.47% for DenseNet, 35.94% for Inception, 41.41% for MobileNet, 35.94% for ResNet and 82.81% for VGG16. Soft voting achieved 89.15% and hard voting achieved 80.95%.</p> <p>The final step in the preprocessing stage combines both CLAHE and ROI. The combination led to an improvement in classification performance. In the first level classification (normal vs abnormal), the models achieved an 91.77% accuracy for AlexNet, 92.72% for DenseNet, 95.57% for Inception, 96.52% for MobileNet, 76.90% for ResNet and 91.14% for VGG16. The accuracy of soft voting is 90.88% and hard voting is 93.99%. In the second level classification (benign vs malignant), the system evaluated the performance of various models. The models achieved an 87.69% accuracy for AlexNet, 99.23% for DenseNet, 96.92% for Inception, 93.08% for MobileNet, 94.62% for ResNet and 97.69% for VGG16. Soft voting achieved 96.92% and hard voting achieved 97.69%.</p> <p>These results indicate that every preprocessing step contributed to an improvement in classification accuracy across all models. This study demonstrates that a two-level, ensemble-based deep learning system can significantly improve the diagnostic accuracy of breast ultrasound images, offering a promising tool to assist radiologists and support early detection efforts, particularly in resource-constrained healthcare environments.

A thresholding based technique to extract retinal blood vessels from fundus images

Analyzing retinal fundus image is important for early detection of diseases related to the eye. However, in fundus images the contrast between retinal blood vessels and the background is very low. Hence, analyzing or visualizing tiny blood vessels is difficult. Fluorescein angiogram overcomes this imaging problem but it is an invasive procedure that leads to other physiological problems. In this work, we enhance the low contrast of retinal vasculature in the ocular fundus image by determining the retinal pigments makeup, namely haemoglobin, melanin and macular pigment using independent component analysis. Independent component image due to haemoglobin pigment obtained exhibits higher contrast retinal vasculature. Even though the fundus fluorescein angiograph method produces higher contrast images, results show that this approach outperforms other noninvasive enhancement method, such as contrast limited adaptive histogram equalization and can be beneficial for vessel segmentation. Contrast enhancement factor of 3.2 for digital retinal images is achieved. This improvement in contrast reduces the need of applying contrasting agent on patients.

Automatic Detection and Classification of Diabetic Retinopathy from Retinal Fundus Images by Abdullah Biran, Master of Applied Science, lectrical and computer engineering Department, Ryerson University, 2017. Diabetic Retinopathy (DR) is an eye disease that leads to blindness when it progresses to proliferative level. The earliest signs of DR are the appearance of red and yellow lesions on the retina called hemorrhages and exudates. Early diagnosis of DR prevents from blindness. In this thesis, an automatic algorithm for detecting diabetic retinopathy is presented. The algorithm is based on combination of several image processing techniques including Circular Hough Transform (CHT), Contrast Limited Adaptive Histogram Equalization (CLAHE), Gabor filter and thresholding. In addition, Support Vector Machine (SVM) classifier is used to classify retinal images into normal or abnormal cases of DR including non-proliferative (NPDR) or proliferative diabetic retinopathy (PDR). The proposed method has been tested on fundus images from Standard Diabetic Retinopathy Database (DIARETDB). The implementation of the presented methodology was done in MATLAB. The methodology is tested for sensitivity and accuracy.

The retinal blood vessels segmentation algorithm is a powerful tool for the early detection of ophthalmic and cardiovascular diseases and biometrics of the automatic tracking system. Accurate segmentation of blood vessels from a retinal image plays a significant role in the prudent examination of the vessels. Therefore, a combined algorithm of a supervised generalized linear model and an unsupervised contrast limited adaptive histogram equalization is proposed in this paper. Using a generalized linear model integrated with multi-scale information by Gabor wavelet transform, the proposed supervised process can extract more prominent features of retinal blood vessels. Besides, the contrast limited adaptive histogram equalization uses the local histogram equalization, which can handle the illumination variation and adjust the enlargement of details. The method is evaluated on a publicly available DRIVE dataset, as it contains ground truth images precisely marked by experts. The segmentation results show that the proposed method can segment the blood vessels accurately.

The most common causes of blindness in adults worldwide, 2.6% of them are caused by diabetic retinopathy, which is a progressive disease caused by complications of diabetes mellitus. Early detection and prompt treatment help save eyesight. The artificial intelligence technology can provide objective and accurate screening results. Deep learning, especially the Convolutional Neural Network (CNN), which is part of artificial intelligence, has proven successful in solving image problems and is very well used in medical image analysis. CNN works well on large datasets, but it will affect network performance to overfitting in fewer datasets. So to solve the problem of small data, data augmentation techniques can be used. There are various kinds of data augmentation techniques. This study used the CNN method to classify diabetic retinopathy disease, compared several suitable data augmentation techniques for retinal fundus images, and used Contrast Limited Adaptive Histogram Equalization (CLAHE) for image enhancement. This study found that the augmented random zoom technique, together with CLAHE, provided the best accuracy of 98% with 96% sensitivity and 100% specificity.

Fréchet PDF based Matched Filter Approach for Retinal Blood Vessels Segmentation

Automated blood vessel segmentation of retinal images offers huge potential benefits for medical diagnosis of different ocular diseases. In this paper, 2D Matched Filters (MF) are applied to fundus retinal images to detect vessels which are enhanced by Contrast Limited Adaptive Histogram Equalization (CLAHE) method. Due to the Gaussian nature of blood vessel profile, the MF with Gaussian kernel often misclassifies non-vascular structures (e.g., step, ramp or other transients) as vessels. To avoid such false detection, this paper introduces Laplacian of Gaussian (LoG) filters in the vessel segmentation process. The inherent zero-crossing property of LoG filter is used in the algorithm, along with the MF, in order to extract vessels reliably from the retinal images. The proposed method is validated against three publicly available databases, STARE, DRIVE and HRF. Simulation results show that the proposed method is able to segment vessels accurately from the three database images with an average accuracy that is competitive to the existing methodologies.

Diabetic retinopathy is a kind of disorder which occurs due to high blood sugar level. This disorder affects retina in many ways. Blood vessels in the retina gets altered . Exudates are secreted, hemorrhages occur, swellings appear in the retina. Diabetic Retinopathy (DR) is the major cause of blindness. Automatic Recognition of DR lesions like Exudates, in digital fundus images can contribute to the diagnosis and screening of this disease. In this approach, an automatic and efficient method to detect the exudates is proposed. The real time retinal images are obtained from a nearby hospital. The retinal images are pre-processed via. Contrast Limited Adaptive Histogram Equalization (CLAHE). The preprocessed colour retinal images are segmented using K-Means Clustering technique. The segmented images establish a dataset of regions. To classify these segmented regions into Exudates and Non-Exudates, a set of features based on colour and texture are extracted. Classification is done using support Vector Machine This method appears promising as it can detect the very small areas of exudates. Keywords - Diabetic Retinopathy, Exudates, fundus image, k means clustering, SVM I. INTRODUCTION The retina is the inner and most important layer of the eye. It is composed of several important anatomical structures which can indicate various diseases. Cardiovascular disease such as stroke and myocardial infarcation can be identified from retinal blood vessels. Diabetic Retinopathy is the common retinal complication associated with diabetes. It is a major cause of blindness in both middle and advanced age groups. The International Diabetes Federation reports that over 50 million people in India have this disease and it is growing rapidly (IDF 2009a) (2). The estimated prevalence of diabetes for all age groups worldwide was 2.8% in 2000 and 4.4% in 2030 meaning that the total number of diabetes patients is forecasted to rise from 171 million in 2000 to 366 million in 2030 (3). Therefore regular screening is the most efficient way of reducing the vision loss. Diabetic Retinopathy is mainly caused by the changes in the blood vessels of the retina due to

Fundus images are consistently used for the analysis of numerous pathological syndromes. This paper presents a three step thresholding based method to extract the blood vessels from the retinal images. In the first step a unique combination of principal component analysis (PCA) and contrast limited adaptive histogram equalization (CLAHE) is used to enhance the retinal images. In the second step the blood vessels are extracted with the help of global Otsu thresholding method. Finally in the last step morphological cleaning is applied to remove unwanted frills. The proposed method enables implementation to be easier and takes less computational time. The two publicly available DRIVE and STARE databases are taken for evaluation of the performance of the presented method. It achieves an average accuracy, sensitivity and specificity about 0.956, 0.723 and 0.984 for DRIVE database while 0.954, 0.737 and 0.982 for STARE database respectively.

Diabetic Retinopathy (DR) is a vision loss impairment due to complications arising from the diabetic condition and affects the retina and resulting pathologies can be monitored by analysing the colour fundus image. However, in retinal fundus images, the contrast between the retinal vasculature and the background is very low and varies within the image making visualisation and analysis of small retinal vasculatures difficult. Therefore, enhancement of the fundus image is important to provide the best visualization of the retinal blood vessels. Fluorescein angiogram overcomes this imaging problem but it is invasive and leads to other physiological problems. In this research work, a non-invasive digital image enhancement technique called RETICA has been developed that overcomes the problem of varied and low contrast in fundus images. RETICA first normalises the varied contrast using a Retinex based method that separates the illumination from the reflectance part of the image followed by ICA that forms the original retinal pigment makeup namely the macular, haemoglobin and melanin retinal pigment. The haemoglobin image exhibits the highest contrast for retinal vessels. Results based on a dataset of 13 fundus images show that RETICA successfully normalises the low and varied contrast and enhances the retinal vessels. It achieved a better average contrast improvement factor of up to 5.56 compared to the invasive FFA with 5.34. This improvement in contrast reduces the need for fluorescein angiogram in DR assessment.