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

Accurate segmentation of retinal blood vessels is of great significance for diagnosing, preventing and detecting eye diseases. In recent years, the U-Net network and its various variants have reached advanced level in the field of medical image segmentation. Most of these networks choose to use simple max pooling to down-sample the intermediate feature layer of the image, which is easy to lose part of the information, so this study proposes a simple and effective new down-sampling method Pixel Fusion-pooling (PF-pooling), which can well fuse the adjacent pixel information of the image. The down-sampling method proposed in this study is a lightweight general module that can be effectively integrated into various network architectures based on convolutional operations. The experimental results on the DRIVE and STARE datasets show that the F₁-score index of the U-Net model using PF-pooling on the STARE dataset improved by 1.98%. The accuracy rate is increased by 0.2%, and the sensitivity is increased by 3.88%. And the generalization of the proposed module is verified by replacing different algorithm models. The results show that PF-pooling has achieved performance improvement in both Dense-UNet and Res-UNet models, and has good universality.
Algorithms ; Retinal Vessels ; Image Processing, Computer-Assisted

OBJECTIVE: We wanted to investigate the radial peripapillary capillary (RPC) network in patients with Bietti crystalline dystrophy (BCD). METHODS: We compared RPC densities in the disk and different peripapillary regions, obtained using optical coherence tomography angiography in 22 patients with BCD (37 eyes) and 22 healthy subjects (37 eyes). The BCD group was then divided into Stage 2 and Stage 3 subgroups based on Yuzawa staging, comparing the RPC densities of the two. RESULTS: The disk area RPC density was 38.8% ± 6.3% in the BCD group and 49.2% ± 6.1% in the control group ( P < 0.001), and peripapillary region RPC density was significantly lower in the BCD group than in the control group (49.1% ± 4.7% and 54.1% ± 3.0%, respectively, P < 0.001). There were no significant RPC density differences between the tempo quadrant and inside disk of Stages 2 and 3 subgroups; the other areas showed a significantly lower RPC density in Stage 3 than in Stage 2 BCD. CONCLUSION The BCD group RPC density was significantly lower than the control group. The reduction of RPC density in the tempo quadrant occurred mainly in the Stage 1 BCD. In contrast, the reduction of RPC density in superior, inferior, and nasal quadrants occurred mainly in Stage 2.
Adult ; Aged ; Angiography ; Corneal Dystrophies, Hereditary/physiopathology* ; Female ; Humans ; Male ; Microvascular Density ; Microvessels/physiopathology* ; Middle Aged ; Retinal Diseases/physiopathology* ; Retinal Vessels/physiopathology* ; Tomography, Optical Coherence

Objective: To report the clinical profile and treatment outcomes of neovascular glaucoma (NVG). Methods: A retrospective cohort study was done in a single tertiary hospital. Medical records of patients diagnosed with NVG from January 2000 to August 2018 were reviewed and pertinent data were collected. Study outcomes included visual acuity (VA) and intraocular pressure (IOP) and were analyzed for eyes that received any of the following: intravitreal bevacizumab (IVBe), pan-retinal photocoagulation (PRP), trabeculectomy with mitomycin (trab-MMC), or diode laser cyclophotocoagulation (DLCP) with at least 1 month of follow-up. Results: There were 162 patients (181 eyes) diagnosed with NVG. Mean age at the time of diagnosis was 55.6 ± 14 years. Diabetic retinopathy (DR) was observed in 81 (45%) eyes and central retinal vein occlusion (CRVO) in 48 (27%) eyes. Baseline VA was hand movement in 67 (37%) eyes and no light perception (NLP) in 49 (27%) eyes. Only 60 (33%) eyes had ≥1 month of follow up (mean of 73 ± 119.1 weeks) after procedures were done. IVBe was done in 22 (37%) eyes, trab-MMC in 20 (33%), PRP in 22 (37%), and DLCP in 24 (40%) eyes. IOP decreased from 45 to 20 mmHg (p<0.001) but VA decreased from LogMAR 1.7 to 2.1 (p<0.01). There was significant VA decrease in eyes that underwent a single procedure (p<0.02) but none in eyes that underwent 2 or 3 procedures. Nonetheless, there was significant IOP decrease (p<0.05) when one procedure was done. Of the 49 eyes that had baseline sight, 19 (41%) converted to NLP (p<0.01). Conclusion Most patients presenting with advanced NVG had DR and CRVO. Procedures led to better IOP but not VA and some lost vision. Aggressive screening for NVG among high-risk groups is warranted to institute treatment early.
Bevacizumab ; Retinal Vein Occlusion ; Diabetic Retinopathy ; Glaucoma, Neovascular ; Trabeculectomy

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*

artery (ICA) aneurysms.MATERIALS AND METHODS: Between April 2016 and 2018, 50 patients (mean age, 56.5 years, 45 females [90%]) with 54 unruptured distal ICA aneurysms (average diameter: 5.6 ± 1.7 mm) were enrolled. The primary endpoint for effectiveness was successful coil embolization with the Alpha stent, and subsequent complete or near-complete occlusion at the 6-month magnetic resonance angiography assessment. The primary safety endpoint was the absence of serious adverse events (SAEs) up to 6 months from the procedure.RESULTS: The primary effectiveness endpoint was observed in 94.4% (51/54) aneurysms. In one patient with technical failure, the stent could not be deployed because of parent artery tortuosity; therefore, a different type of stent was used. Of the 53 aneurysms treated with the Alpha stent, complete occlusion was achieved in 64.1% (34/53) cases, and near-complete occlusion was achieved in 32.0% (17/53) cases by the 6-month follow-up. Two cases (3.7%) required retreatment because of major recurrence. In 4% (2/50) patients, SAEs, i.e., retinal artery thromboembolism and corona radiata lacunar infarction, were reported after the procedure.CONCLUSION: For endovascular treatment of unruptured, wide-necked, distal ICA aneurysms, coil embolization using the newly developed Alpha stent showed excellent procedural and mid-term clinical follow-up results in terms of effectiveness and safety.]]>
Adult ; Aneurysm ; Arteries ; Carotid Artery, Internal ; Embolization, Therapeutic ; Female ; Follow-Up Studies ; Humans ; Magnetic Resonance Angiography ; Parents ; Prospective Studies ; Recurrence ; Retinal Artery ; Retreatment ; Stents ; Stroke, Lacunar ; Thromboembolism

artery sparing central retinal artery occlusion (CRAO).METHODS: A retrospective study was conducted on 90 eyes diagnosed with complete CRAO, including 16 cases of cilioretinal artery sparing CRAO. Clinical features, visual outcome, papillomacular bundle involvement, and remnant visual field were analyzed according to cilioretinal artery sparing.RESULTS: Among eyes with complete CRAO, the proportion of cilioretinal artery sparing CRAO was 17.8% (16 / 90). Mean initial best-corrected visual acuities (BCVAs) (2.04 ± 0.69 vs. 2.34 ± 0.47, p = 0.039) and final BCVAs (1.65 ± 0.87 vs. 2.22 ± 0.84, p = 0.001) were significantly better in eyes of the cilioretinal artery sparing group than the non-sparing group. The proportion with poor visual outcome (final BCVA <20 / 200) was 81.3% in the cilioretinal artery sparing group and 97.3% in the non-sparing group (p = 0.01). In sub-group analysis within cilioretinal artery sparing CRAO eyes, ischemic involvement of the papillomacular bundle at disease onset was significantly more frequent in the poor vision group (BCVA <20 / 200, 12 / 13 [92.3%]) than in the good vision group (BCVA ≥20 / 200, 1 / 3 [33.3%], p = 0.016) and it was associated with preserved central visual field.CONCLUSIONS: Although cilioretinal artery sparing is common in CRAO and has a better prognosis than complete CRAO, the visual outcome is generally poor and only a small proportion of eyes has preserved small central visual field. Ischemic injury of the papillomacular bundle at the acute stage of CRAO correlates with poor visual outcome and could be a prognostic sign.]]>
Arteries ; Ciliary Arteries ; Prognosis ; Retinal Artery Occlusion ; Retinal Artery ; Retrospective Studies ; Tomography, Optical Coherence ; Visual Acuity ; Visual Fields

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*

PURPOSE: To report a case of retinal toxicity after an intravitreal ganciclovir injection to treat acute retinal necrosis in an eye filled with silicone oil.CASE SUMMARY: A 56-year-old male presented with ocular pain and visual loss in his right eye. His best-corrected visual acuity was 20/25, inflammatory cells in the anterior chamber, multiple retinitis lesions and retinal vessel occlusions in the peripheral retina and vitreous opacity were showed. Acute retinal necrosis was suspected, anterior chamber polymerase chain reaction (PCR) test was done. Aciclovir 2,400 mg/day intravenously and ganciclovir 2.0 mg were administered by intravitreal injection. After 4 days, retinitis was worsened and PCR test was positive for varicella zoster virus. Ganciclovir intravitreal injections were increased twice a week. After 16 days, retinal detachment occurred, so scleral encircling, vitrectomy, laser photocoagulation, and silicone oil tamponade were conducted. Ganciclovir 1.0 mg was injected at the end of surgery. The patient's visual acuity decreased to hand motion, and multiple crystal deposits with multiple retinal hemorrhages were observed in the right eye the next day. Visual acuity did not recover and optical coherent tomography showed that the macula was thinned.CONCLUSIONS: Visual loss seemed to be related with the retinal toxicity of ganciclovir. The increased local concentration due to the silicone oil tamponade is thought to have caused the toxicity.
Acyclovir ; Anterior Chamber ; Ganciclovir ; Hand ; Herpesvirus 3, Human ; Humans ; Intravitreal Injections ; Light Coagulation ; Male ; Middle Aged ; Polymerase Chain Reaction ; Retina ; Retinal Detachment ; Retinal Hemorrhage ; Retinal Necrosis Syndrome, Acute ; Retinal Vessels ; Retinaldehyde ; Retinitis ; Silicon ; Silicones ; Visual Acuity ; Vitrectomy