Search

Background/objectives We aim to develop an objective fully automated Artificial intelligence (AI) algorithm for MNV lesion size and leakage area segmentation on fluorescein angiography (FA) in patients with neovascular age-related macular degeneration (nAMD). Subjects/methods Two FA image datasets collected form large prospective multicentre trials consisting of 4710 images from 513 patients and 4558 images from 514 patients were used to develop and evaluate a deep learning-based algorithm to detect CNV lesion size and leakage area automatically. Manual segmentation of was performed by certified FA graders of the Vienna Reading Center. Precision, Recall and F1 score between AI predictions and manual annotations were computed. In addition, two masked retina experts conducted a clinical-applicability evaluation, comparing the quality of AI based and manual segmentations. Results For CNV lesion size and leakage area segmentation, we obtained F1 scores of 0.73 and 0.65, respectively. Expert review resulted in a slight preference for the automated segmentations in both datasets. The quality of automated segmentations was slightly more often judged as good compared to manual annotations. Conclusions CNV lesion size and leakage area can be segmented by our automated model at human-level performance, its output being well-accepted during clinical applicability testing. The results provide proof-of-concept that an automated deep learning approach can improve efficacy of objective biomarker analysis in FA images and will be well-suited for clinical application.

Hassan Khaled,1,* Bilal Haj Najeeb2,* 1Magrabi Eye & Ear Hospital, Makkah, Saudi Arabia; 2Department of Ophthalmology, Medical University of Vienna, Vienna, Austria*These authors contributed equally to this workCorrespondence: Hassan Khaled, Magrabi Eye & Ear Hospital, The Holy Mosque Street, Al Murasalat, Makkah, 24247, Saudi Arabia, Email hassan.adel@magrabi.com.saAbstract: Leakage of fluid through the side port during aspiration of the cortex leads to instability of the anterior chamber. In addition, eye movement may cause an unintended pulling of the irrigation probe out of the corneal wound resulting in collapsing of the anterior chamber. Both situations could pose a challenge to the surgeon and increase the risk of serious intraoperative complications. Therefore, we describe a simple effective maneuver to avoid these conditions during bimanual cortex removal and viscoelastic washout. In Khaled technique, rotating the irrigation probe outwards causes complete occlusion of the lumen of the side port and a simultaneous stabilization of the anterior chamber. Twisting the incision by the irrigation probe also offers better fixation of the eye at the edge of the side port and a subsequent reduction of eye movement.Keywords: irrigation probe, irrigation/aspiration, cataract surgery, cortex removal, phacoemulsification

Purpose To report on the morphological characteristics and regional distribution of multifocal macular neovascularization type 3 (mMNV3). Methods Twenty-two consecutive eyes of 21 patients with mMNV3 were included using multimodal imaging. The count and stage of lesions of all MNV types and the existence of exudate and hemorrhage were determined. Also, we addressed the regional distribution of MNV3 lesions between the superior-inferior and the nasal-temporal halves of the macula, and the range of the distance of the lesions from the central fovea. Furthermore, we explored the number of feeding vessels including the cilioretinal artery. Results We found 51 lesions in 22 eyes of 21 patients. They were bifocal in 16 (73%) eyes, trifocal in 5 (23%), and quadrifocal in one (4%). No lesion of MNV1 or 2 was found. Fifteen (68%), 2 (9%), and 16 (73%) eyes were associated with retinal hard exudate, subretinal pigment epithelium exudate, and intraretinal hemorrhage, respectively. Thirty (59%) lesions were located in the temporal half of the macula, whereas 21 (41%) were located nasally (p = 0.07). One (2%) lesion was closer than 500 µm, 49 (96%) between 500 and 1500 µm, and one (2%) between 1500 and 3000 µm. The lesions were supplied by one arteriole in one (4%) eye, two arterioles in 16 (73%) eyes, and 3 arterioles in 5 (23%) eyes. The CRA contributed as a feeding vessel in 5 (23%) eyes. Conclusion The multifocal variant of MNV3 has specific morphological and topographical characteristics. Multimodal imaging allows the understanding of the pathomorphological condition in more detail.

Given the wide spectrum of unique characteristics of macular neovascularization type 3 (MNV3) compared with types 1 and 2, we suggest regrading the colour photography assessment of the AREDS study to verify the impact of AREDS supplements on eyes with MNV3.

PurposeTo report on patients with macular neovascularisation type III (MNV3) arising from cilioretinal arteries (CRAs) (cilioretinal macular neovascularisation type III (cMNV3)).MethodsWe reviewed baseline examinations of patients with neovascular age-related macular degeneration using multimodal imaging. We determined the type and distribution of MNV lesions in each cMNV3 case, the range of distances from the fovea, existence of exudative maculopathy, intraretinal haemorrhage and other morphological characteristics. 50 consecutive eyes with usual MNV3 without CRA were included as a control group.Results102 eyes of 102 patients were identified with MNV3 lesions. Among these, we found 12 eyes (12%) with cMNV3, 84 eyes (82%) with usual MNV3 without CRA and 6 eyes (6%) with usual MNV3 with CRA. Ten cases of cMNV3 had one lesion, and two cases had two lesions. The lesions were distributed equally between the superior and inferior halves of the macula, whereas in the nasal and temporal halves, there were 8 (57%) and 6 (43%) lesions, respectively. All cMNV3 lesions were located between 500 and 1500 µm from the central fovea except one, which was located between 1500 and 3000 µm. None of the lesions had macular neovascularisation type I (MNV1) or macular neovascularisation type II (MNV2) elsewhere in both groups. Exudative maculopathy and intraretinal haemorrhage were found in seven (88%) and five (63%) of the eight pure cMNV3 cases, respectively.ConclusioncMNV3 can be solitary or multiple, isolated or accompanied with usual MNV3 lesions, but not with concurrent MNV1 or MNV2. It is frequently associated with extensive exudative maculopathy, intraretinal haemorrhage and subretinal fluid.

To explore the condition of fellow eyes of patients with macular neovascularization Type 3 (MNV3) and to verify whether the retinal-choroidal anastomosis (RCA) develops equally in all MNV types.The contralateral eyes of 94 patients with MNV3, 96 patients with MNV1, and 96 patients with MNV2 were included. Multimodal imaging was performed. The MNV3 stage including the development of fibrosis and RCA over 24 months was determined.In the contralateral eyes of patients of the solitary (one lesion) MNV3 group, 32 eyes (42.1%) showed early/intermediate age-related macular degeneration, 25 eyes (33%) showed MNV3, and 11 eyes (14.5%) experienced fibrosis, of which 4 eyes (5.2%) had a RCA, 7 eyes (9.2%) had atrophy after resolved MNV3, and 1 eye (1.3%) developed MNV1. In the multifocal (more than one lesion) MNV3 group, 2 eyes (11.1%) showed early/intermediate age-related macular degeneration, 9 eyes (50%) showed 15 MNV3 lesions, and 4 eyes (22.2%) showed fibrosis, of which 2 eyes (11.1%) manifested with a RCA and 3 eyes (16.7%) showed atrophy after resolved MNV3. The number of eyes with a RCA accounted for 40% of all eyes with fibrosis. The count of simultaneous bilateral multifocal MNV3 was 5 (55.6%). In the MNV1 and MNV2 groups, no eye developed a RCA. The incidence of RCAs in the scarred eyes in MNV3 was significantly higher (P0.0001).Retinal-choroidal anastomosis is an exclusive clinical feature of MNV3. The development of the multifocal MNV3 is usually bilateral and simultaneous. The occurrence of fibrosis in MNV3 has decreased dramatically after the introduction of the antiangiogenic therapy.

BACKGROUND/AIMS To explore whether the existence and pattern of distribution of macular haemorrhage or exudate can be valuable diagnostic markers for macular neovascularization type 3 (MNV3) in patients with neovascular age-related macular degeneration. METHODS Eighty-three eyes of 83 consecutive treatment naive patients with stage 3 MNV3 were enrolled. The diagnosis was based on fluorescein angiography (FA) and optical coherence tomography (OCT). Subretinal and intraretinal haemorrhage and dense exudates were evaluated on colour fundus photography. Fluorescein angiography (FA) images and OCT scans were used to identify the axial location of the haemorrhage. 83 patients with MNV1 and 83 with MNV2 were included as two control groups. RESULTS In the MNV3 group, 62 (75%) eyes had intraretinal haemorrhage and 52 (63%) had dense exudates. 73 (88%) eyes had intraretinal haemorrhage and/or dense exudates. 41 (49%) had both pathologies. The intraretinal haemorrhage was flame shaped over the lesion and punctate or semi-punctate further away from it and directed to the fovea. No subretinal haemorrhage was noticed. In the MNV1 and MNV2 groups, 11 (13%) and 24 (29%) eyes had subretinal haemorrhage or dense exudates, respectively. No intraretinal haemorrhage was seen in the two control groups. The prevalence of exudates and haemorrhage (irrespective of its location) was greater in MNV3 than in MNV1 or 2 (p

AimTo objectively assess disease activity and treatment response in patients with retinal vein occlusion (RVO), neovascular age-related macular degeneration (nAMD) and centre-involved diabetic macular oedema (DME), using artificial intelligence–based fluid quantification.MethodsPosthoc analysis of 2311 patients (11 151 spectral-domain optical coherence tomography volumes) from five clinical, multicentre trials, who received a flexible antivascular endothelial growth factor (anti-VEGF) therapy over a 12-month period. Fluid volumes were measured with a deep learning algorithm at baseline/months 1, 2, 3 and 12, for three concentric circles with diameters of 1, 3 and 6 mm (fovea, paracentral ring and pericentral ring), as well as four sectors surrounding the fovea (superior, nasal, inferior and temporal).ResultsIn each disease, at every timepoint, most intraretinal fluid (IRF) per square millimetre was present at the fovea, followed by the paracentral ring and pericentral ring (pConclusionFully automated localisation and quantification of IRF/SRF over time shed light on the fluid dynamics in each disease. There is a specific anatomical response of IRF/SRF to anti-VEGF therapy in all diseases studied.

Purpose To explore the regional distribution of macular neovascularization type 3 (MNV3). Methods Seventy-eight eyes of 78 patients were reviewed. We defined the location of each lesion after applying a modified ETDRS grid and the incidence of simultaneous MNV1 or 2. Also, we investigated the distribution of MNV3 at the outline of the foveal avascular zone and when the diameter of foveal avascular zone was less than 325 µm. Results The distribution of MNV3 was 4 lesions (5%) from the center to 500 µm, 72 (92%) from 500 µm to 1500 µm, and 2 (3%) from 1,500 µm to 3000 µm. The distribution in respect of the ETDRS fields was 7 (9%) nasal, 16 (20%) superior, 32 (40%) temporal, and 23 (31%) inferior. No additional MNV1 or 2 were found elsewhere. Most lesions tended to distribute along straight bands radiating from the perifoveal area, mainly in the temporal half (72%). None of the cases had MNV3 at the boundary of the foveal avascular zone. Only five cases had foveal avascular zone diameter of less than 325 µm, the closest lesion was 425 µm away from the center. Conclusion MNV3 lesions are most likely neither symmetrical nor uniformly distributed. They have a higher affinity to distribute radially in the temporal perifoveal area.

Background/aimsTo characterise neuroretinal atrophy in retinal vein occlusion (RVO).MethodsWe included patients with central/branch RVO (CRVO=196, BRVO=107) who received ranibizumab according to a standardised protocol for 6 months. Retinal atrophy was defined as the presence of an area of retinal thickness (RT) Results23 patients with CRVO and 11 patients with BRVO demonstrated retinal atrophy, presenting as sharply demarcated retinal thinning confined to a macular quadrant. The mean RT in the atrophic quadrant at month 6 was 249±26 µm (CRVO) and 244±29 µm (BRVO). Individual layer analysis revealed pronounced thinning in the IPL to ONL compartment. Change in BCVA at 6 months was similar between the groups (BRVO, +15 vs +18 letters; CRVO, +14 vs +18 letters).ConclusionsIn this exploratory analysis, we describe the characteristics of neuroretinal atrophy in RVO eyes with resolved macular oedema after ranibizumab therapy. Our analysis shows significant, predominantly retinal thinning in the IPL to ONL compartment in focal macular areas in 11% of patients with RVO. Eyes with retinal atrophy did not show poorer BCVA outcomes.

Purpose To determine the distribution of leakage on fluorescein angiography (FA) and explore the clinically protective role of astrocytes against damage to the inner blood retinal barrier (iBRB) in diabetic macular edema (DME). Methods A consecutive case series of 87 eyes of 87 patients with DME was included. We measured the leakage area in each field of the Early Treatment Diabetic Retinopathy Study (ETDRS) grid on late-phase FA images. The normative thickness of the nerve fiber layer (NFL), in which the astrocytes are confined, was derived from a previous work using spectral-domain optical coherence tomography. We explored the difference in leakage areas in every two fields. Moreover, we investigated the correlation between the mean of the leakage area and the mean of thickness of the NFL in each ETDRS field. Results The leakage areas in the nasal, inferior, superior, and temporal fields were 2.34 mm2, 2.84 mm2, 3.03 mm2, and 3.96 mm2. The difference in leakage area between each two fields was significant in all cases (P < 0.05) except between the inferior and superior fields (P = 0.65). The temporal field was the only field that showed leakage in all 87 cases. The correlation between the leakage area and the thickness of the NFL in the ETDRS fields was negative and highly significant: r = -0.96 (95% confidence interval -0.99 to -0.02). Conclusion The distribution of leakage correlates inversely and statistically significantly with the thickness of the NFL, suggesting astrocytes in the NFL play a pivotal role in preventing damage to the iBRB and subsequent evolution of microaneurysms in DME. Moreover, fluid extravasation due to damage to the iBRB is expressed earlier in the temporal than in the other three fields.