Modelling of retinal vasculature based on genetically tuned parametric L-system

Structures of retinal blood vessels are of great importance in diagnosis and treatment of diseases that affect the eyes. Parametric Lindenmayer system (L-system) is one of the powerful rule-based methods that has a great capability for generating tree-like structures using simple rewriting rules. In this study, a novel framework, which can be used to model the retinal vasculature based on available images, has been proposed. This framework presents a solution to special instance of a general open problem, the L-system inverse problem, in which L-system rules should be obtained based on images representing a particular tree-like structure. In this study, genetic algorithm with a novel objective function based on feature matching and an L-system grammar comparison has been used along with nonlinear regression to solve the parametric L-system inverse problem. The resulting L-system growth rules have been employed to predict inaccessible vascular branches. Graphical and quantitative comparison between model results and target structures of different case studies reveals that the proposed framework can be used to generate the structure of retinal blood vessels accurately. Even in the cases lacking sufficient image data, it can provide acceptable predictions.

Download Full-text

DAVS-NET: Dense Aggregation Vessel Segmentation Network for retinal vasculature detection in fundus images

PLoS ONE ◽

10.1371/journal.pone.0261698 ◽

2021 ◽

Vol 16 (12) ◽

pp. e0261698

Author(s):

Mohsin Raza ◽

Khuram Naveed ◽

Awais Akram ◽

Nema Salem ◽

Amir Afaq ◽

...

Keyword(s):

Deep Learning ◽

Blood Vessels ◽

Medical Image Analysis ◽

Retinal Disease ◽

Disease Diagnosis ◽

Vessel Segmentation ◽

Diabetic Patients ◽

Retinal Vasculature ◽

Retinal Blood Vessels ◽

Dense Aggregation

In this era, deep learning-based medical image analysis has become a reliable source in assisting medical practitioners for various retinal disease diagnosis like hypertension, diabetic retinopathy (DR), arteriosclerosis glaucoma, and macular edema etc. Among these retinal diseases, DR can lead to vision detachment in diabetic patients which cause swelling of these retinal blood vessels or even can create new vessels. This creation or the new vessels and swelling can be analyzed as biomarker for screening and analysis of DR. Deep learning-based semantic segmentation of these vessels can be an effective tool to detect changes in retinal vasculature for diagnostic purposes. This segmentation task becomes challenging because of the low-quality retinal images with different image acquisition conditions, and intensity variations. Existing retinal blood vessels segmentation methods require a large number of trainable parameters for training of their networks. This paper introduces a novel Dense Aggregation Vessel Segmentation Network (DAVS-Net), which can achieve high segmentation performance with only a few trainable parameters. For faster convergence, this network uses an encoder-decoder framework in which edge information is transferred from the first layers of the encoder to the last layer of the decoder. Performance of the proposed network is evaluated on publicly available retinal blood vessels datasets of DRIVE, CHASE_DB1, and STARE. Proposed method achieved state-of-the-art segmentation accuracy using a few number of trainable parameters.

Download Full-text

Automatic Retinal Vessel Segmentation Based on an Improved U-Net Approach

Scientific Programming ◽

10.1155/2021/5520407 ◽

2021 ◽

Vol 2021 ◽

pp. 1-15

Author(s):

Zihe Huang ◽

Ying Fang ◽

He Huang ◽

Xiaomei Xu ◽

Jiwei Wang ◽

...

Keyword(s):

Blood Vessels ◽

Structural Changes ◽

Area Under The Curve ◽

Image Data ◽

Retinal Vessel ◽

Vessel Segmentation ◽

Circulation System ◽

Retinal Blood Vessels ◽

Image Patches ◽

Retinal Vessel Segmentation

Retinal blood vessels are the only deep microvessels in the blood circulation system that can be observed directly and noninvasively, providing us with a means of observing vascular pathologies. Cardiovascular and cerebrovascular diseases, such as glaucoma and diabetes, can cause structural changes in the retinal microvascular network. Therefore, the study of effective retinal vessel segmentation methods is of great significance for the early diagnosis of cardiovascular diseases and the vascular network’s quantitative results. This paper proposes an automatic retinal vessel segmentation method based on an improved U-Net network. Firstly, the image patches are rotated to amplify the image data, and then, the RGB fundus image is preprocessed by normalization. Secondly, after the improved U-Net model is constructed with 23 convolutional layers, 4 pooling layers, 4 upsampling layers, 2 dropout layers, and Squeeze and Excitation (SE) block, the extracted image patches are utilized for training the model. Finally, the fundus images are segmented through the trained model to achieve precise extraction of retinal blood vessels. According to experimental results, the accuracy of 0.9701, 0.9683, and 0.9698, sensitivity of 0.8011, 0.6329, and 0.7478, specificity of 0.9849, 0.9967, and 0.9895, F1-Score of 0.8099, 0.8049, and 0.8013, and area under the curve (AUC) of 0.8895, 0.8845, and 0.8686 were achieved on DRIVE, STARE, and HRF databases, respectively, which is better than most classical algorithms.

Download Full-text

Inception Capsule Network for Retinal Blood Vessel Segmentation and Centerline Extraction

10.1101/815555 ◽

2019 ◽

Author(s):

C. Kromm ◽

K. Rohr

Keyword(s):

Blood Vessels ◽

Data Augmentation ◽

State Of The Art ◽

Automatic Segmentation ◽

Image Data ◽

Vessel Segmentation ◽

Deep Convolutional Neural Networks ◽

Centerline Extraction ◽

Retinal Blood Vessels ◽

Using Data

ABSTRACTAutomatic segmentation and centerline extraction of retinal blood vessels from fundus image data is crucial for early detection of retinal diseases. We have developed a novel deep learning method for segmentation and centerline extraction of retinal blood vessels which is based on the Capsule network in combination with the Inception architecture. Compared to state-of-the-art deep convolutional neural networks, our method has much fewer parameters due to its shallow architecture and generalizes well without using data augmentation. We performed a quantitative evaluation using the DRIVE dataset for both vessel segmentation and centerline extraction. Our method achieved state-of-the-art performance for vessel segmentation and outperformed existing methods for centerline extraction.

Download Full-text

3D Structural Model and Visualization of Blood Vessels Based on L-System

Trends in Sciences ◽

10.48048/tis.2021.1407 ◽

2021 ◽

Vol 18 (24) ◽

pp. 1407

Author(s):

Siriprapa Ritraksa ◽

Khamron Mekchay

Keyword(s):

Blood Vessels ◽

Structural Model ◽

Future Research ◽

Total Blood ◽

Murray's Law ◽

Murray’S Law ◽

L System ◽

Angle Condition ◽

System L ◽

Additional Assumptions

The insight in structures of the blood vessels is a basis for study of blood flows to help understanding the abnormalities of blood vessels that can cause vascular diseases. Basic concept used for constructing structures of blood vessels in organs is arterial branching, which is usually characterized by fractal similarity in the bifurcation pattern. In this work, the concept of Lindenmayer system (L-system) is modified for three-dimensional (3D) tree-like structures to model structures of blood vessels in organs, and then, applied to construct and visualize structural blood vessels via our software created based on openGL and Lazarus program. The structure of blood vessels is constructed based on the physiological law of arterial branching proposed Murray (Murray’s law) under additional assumptions and constraints such as the spreading of blood vessels to cover all directions, the angle condition and the non-overlapping vessels condition. The concept is applied to simulate structures of blood vessels in 3 study cases, including symmetric arterial branching, non-symmetric arterial branching and structure of blood vessel on different domains. The results of structures of blood vessels generated from all cases are measured based on the number of segments, the total blood volume and the fractal dimension. The results of modeling and simulation in this work are illustrated by comparing with other results appeared literature. Moreover, the constructed structures of the blood vessels based on this 3D L-system could be useful for future research such as blood flow, pressure and other properties involving in structures of blood vessels in different organs of human and animals. HIGHLIGHTS A new 3D L-system is developed based on directional vectors for construction of 3D tree-like structures such as structures of blood vessels The model of structures of blood vessels is constructed based on the physiological laws of arterial branching (Murray’s law) with additional assumptions on the spreading of blood vessels, the angle condition, and the non-overlapping of blood vessels Algorithm and software are developed based on L-system to simulate and visualize 3D structures of blood vessels GRAPHICAL ABSTRACT

Download Full-text