Diabetic retinopathy is a common blinding complication in diabetic patients. Compared with conventional fundus color photography, fundus fluorescein angiography can dynamically display retinal vessel permeability changes, offering unique advantages in detecting early small lesions such as microaneurysms. However, existing intelligent diagnostic research on diabetic retinopathy images primarily focuses on fundus color photography, with relatively insufficient research on complex lesion recognition in fluorescein angiography images. This study proposed an adaptive multi-label classification model (D-LAM) to improve the recognition accuracy of small lesions by constructing a category-adaptive mapping module, a label-specific decoding module, and an innovative loss function. Experimental results on a self-built dataset demonstrated that the model achieved a mean average precision of 96.27%, a category F1-score of 91.21%, and an overall F1-score of 94.58%, with particularly outstanding performance in recognizing small lesions such as microaneurysms (AP = 1.00), significantly outperforming existing methods. The research provides reliable technical support for clinical diagnosis of diabetic retinopathy based on fluorescein angiography.
Diabetic Retinopathy/diagnostic imaging* ; Humans ; Fluorescein Angiography ; Microaneurysm/diagnostic imaging* ; Retinal Vessels ; Algorithms

The existing retinal vessels segmentation algorithms have various problems that the end of main vessels are easy to break, and the central macula and the optic disc boundary are likely to be mistakenly segmented. To solve the above problems, a novel retinal vessels segmentation algorithm is proposed in this paper. The algorithm merged together vessels contour information and conditional generative adversarial nets. Firstly, non-uniform light removal and principal component analysis were used to process the fundus images. Therefore, it enhanced the contrast between the blood vessels and the background, and obtained the single-scale gray images with rich feature information. Secondly, the dense blocks integrated with the deep separable convolution with offset and squeeze-and-exception (SE) block were applied to the encoder and decoder to alleviate the gradient disappearance or explosion. Simultaneously, the network focused on the feature information of the learning target. Thirdly, the contour loss function was added to improve the identification ability of the blood vessels information and contour information of the network. Finally, experiments were carried out on the DRIVE and STARE datasets respectively. The value of area under the receiver operating characteristic reached 0.982 5 and 0.987 4, respectively, and the accuracy reached 0.967 7 and 0.975 6, respectively. Experimental results show that the algorithm can accurately distinguish contours and blood vessels, and reduce blood vessel rupture. The algorithm has certain application value in the diagnosis of clinical ophthalmic diseases.
Algorithms ; Fundus Oculi ; Optic Disk ; ROC Curve ; Retinal Vessels/diagnostic imaging*

In robot-assisted eye surgery, such as retinal vascular bypass surgery, precise positioning of operating points is required. In this study, a binocular vision-based 3D reconstruction method is proposed to locate the incision points on retinal vessels. Vessels in the image were extracted by CLAHE algorithm to remove the influence of background, then stereo matching was performed. Finally, the retinal vessel image was reconstructed by using the principle of parallax in binocular vision. Experimental results show that this method can accurately locate the incision points on retinal vessels and meet the requirements of ophthalmic surgery.
Algorithms ; Humans ; Imaging, Three-Dimensional ; Ophthalmologic Surgical Procedures ; Retinal Vessels/diagnostic imaging* ; Robotic Surgical Procedures ; Vision, Binocular

Retinal vascular function is complex, morphological structure varies from person to person, and is susceptible to vascular diseases and systemic vascular diseases. Its accurate segmentation is of great significance for disease diagnosis and identification. In this paper, a multi-scale matching filtering algorithm is proposed for the uneven size of retinal blood vessels. On the basis of the traditional singlescale Gaussian matching filter, multiscale Gaussian matched filters with two sizes are used to enhance grayscale images. Enhancement is performed, and the superimposed image is binarized using a twodimensional maximum entropy threshold segmentation algorithm. The algorithm is tested in the DRIVE database with sensitivity, specificity and accuracy of 0.803, 0.959, 0.981, respectively. Comparing with the traditional algorithm, the algorithm has high sensitivity, fast running speed and rich details of segmentation results.
Algorithms ; Entropy ; Humans ; Image Processing, Computer-Assisted ; Retinal Vessels/diagnostic imaging*

OBJECTIVEThe aim of this study was to determine the repeatability and reproducibility of optical coherence tomography angiography (OCTA) based on optical microangiography (OMAG) measurements of macular vessels in normal eyes.

METHODSIn this prospective cohort study, 40 eyes of 40 healthy volunteers underwent repeated OCTA (Cirrus HD-OCT 5000 angiography system, Carl Zeiss Meditec, Inc.) scans on two separate visit days. On each visit day, the eyes were scanned three times. The following parameters were used to quantitatively describe the OCTA images of the superficial vascular network: vessel area density (VAD), vessel skeleton density (VSD), vessel diameter index (VDI), vessel perimeter index (VPI), vessel complexity index (VCI), flux, and foveal avascular zone (FAZ). Coefficient of variation (CV) and intraclass correlation coefficient (ICC) were calculated for evaluating intravisit and intervisit repeatability, as well as interobserver reproducibility.

RESULTSThe measurements showed high repeatability [CVs ⪕ 4.2% (intravisit) and ⪕ 4.6% (intervisit)] and interobserver reproducibility (ICCs ⪖ 0.923) for all parameters.

CONCLUSIONThis study demonstrated good repeatability and reproducibility of OCTA based on OMAG for the measurement of superficial vessel parameters in normal eyes.

Adult ; Cohort Studies ; Evaluation Studies as Topic ; Female ; Fluorescein Angiography ; standards ; Healthy Volunteers ; Humans ; Image Processing, Computer-Assisted ; Male ; Microvessels ; diagnostic imaging ; Middle Aged ; Prospective Studies ; Reproducibility of Results ; Retina ; diagnostic imaging ; Retinal Vessels ; diagnostic imaging ; Tomography, Optical Coherence ; standards ; Young Adult

In view of the evaluation of fundus image segmentation, a new evaluation method was proposed to make up insufficiency of the traditional evaluation method which only considers the overlap of pixels and neglects topology structure of the retinal vessel. Mathematical morphology and thinning algorithm were used to obtain the retinal vascular topology structure. Then three features of retinal vessel, including mutual information, correlation coefficient and ratio of nodes, were calculated. The features of the thinned images taken as topology structure of blood vessel were used to evaluate retinal image segmentation. The manually-labeled images and their eroded ones of STARE database were used in the experiment. The result showed that these features, including mutual information, correlation coefficient and ratio of nodes, could be used to evaluate the segmentation quality of retinal vessel on fundus image through topology structure, and the algorithm was simple. The method is of significance to the supplement of traditional segmentation evaluation of retinal vessel on fundus image.
Algorithms ; Databases, Factual ; Diagnostic Imaging ; methods ; Diagnostic Techniques, Ophthalmological ; Fundus Oculi ; Humans ; Image Processing, Computer-Assisted ; Retina ; Retinal Vessels ; anatomy & histology

According to fundus fluorescein angiography images characteristics, this paper proposes a feature based image mosaic vessel edge method. First, detect edge of blood vessels by carrying on the pretreatment to the fundus fluorescein angiography image in the foundation, wavelet edge detection algorithm. Then, the matching method based on chain code feature is described. Finally, a local area based on gray level information of the image fusion method is applied to angiographic image series. Data processing results show that the method can generate an ideal mosaic effect.
Fluorescein Angiography ; methods ; Humans ; Image Processing, Computer-Assisted ; methods ; Radiography ; Retinal Vessels ; diagnostic imaging