Geometric Correspondence-Based Multimodal Learning for Ophthalmic Image Analysis.

Color fundus photography (CFP) and Optical coherence tomography (OCT) images are two of the most widely used modalities in the clinical diagnosis and management of retinal diseases. Despite the widespread use of multimodal imaging in clinical practice, few methods for automated diagnosis of eye diseases utilize correlated and complementary information from multiple modalities effectively. This paper explores how to leverage the information from CFP and OCT images to improve the automated diagnosis of retinal diseases. We propose a novel multimodal learning method, named geometric correspondence-based multimodal learning network (GeCoM-Net), to achieve the fusion of CFP and OCT images. Specifically, inspired by clinical observations, we consider the geometric correspondence between the OCT slice and the CFP region to learn the correlated features of the two modalities for robust fusion. Furthermore, we design a new feature selection strategy to extract discriminative OCT representations by automatically selecting the important feature maps from OCT slices. Unlike the existing multimodal learning methods, GeCoM-Net is the first method that formulates the geometric relationships between the OCT slice and the corresponding region of the CFP image explicitly for CFP and OCT fusion. Experiments have been conducted on a large-scale private dataset and a publicly available dataset to evaluate the effectiveness of GeCoM-Net for diagnosing diabetic macular edema (DME), impaired visual acuity (VA) and glaucoma. The empirical results show that our method outperforms the current state-of-the-art multimodal learning methods by improving the AUROC score 0.4%, 1.9% and 2.9% for DME, VA and glaucoma detection, respectively.

Home OCT Imaging for Newly Diagnosed Neovascular Age-Related Macular Degeneration: A Feasibility Study.

OBJECTIVE: To assess the feasibility of daily Home OCT imaging among patients with neovascular age-related macular degeneration (nAMD). DESIGN: Prospective observational study. PARTICIPANTS: Participants with ≥ 1 eye with previously untreated nAMD and visual acuity 20/20 to 20/320. METHODS: Participants meeting the ocular eligibility criteria were considered for enrollment; those who provided consent received a Notal Vision Home OCT device. Participants were instructed to scan both eyes daily. Retina specialists managed treatment according to their standard practice, without access to the Home OCT data. The presence of fluid detected by a reading center (RC) from in-office OCT scans was compared with fluid volumes measured by the Notal OCT Analyzer (NOA) on Home OCT images. MAIN OUTCOME MEASURES: Proportion of participants meeting ocular eligibility criteria who participated in daily scanning, frequency and duration of scanning, proportion of scans eligible for fluid quantification, participant experience with the device, agreement between the RC and NOA fluid determinations, and characteristics of fluid dynamics. RESULTS: Among 40 participants meeting ocular eligibility criteria, 14 (35%) initiated self-scanning. Planned travel (n = 7, 17.5%) and patient-reported inadequate cell reception for the upload of images (n = 5, 12.5%) were the most frequent reasons for not participating. Considering scans of the study eye only, the mean (standard deviation) was 6.3 (0.6) for weekly scanning frequency and 47 (17) seconds for scan duration per eye. Among 2304 scans, 86.5% were eligible for fluid quantification. All participants agreed that scanning became easier over time, and only 1 did not want to continue daily scanning. For 35 scan pairs judged as having fluid by in-office OCT, the NOA detected fluid on 31 scans (89%). For 14 scan pairs judged as having no fluid on in-office OCT, the NOA did not detect fluid on 10 scans (71%). Daily fluid patterns after treatment initiation varied considerably between patients. CONCLUSIONS: For patients with nAMD who initiated home scanning, frequency and quality of scanning and accuracy of fluid detection were sufficient to assess the monitoring of fluid at home. Accommodations for travel and Wi-Fi connectivity could improve uptake of the Home OCT device. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Optical Coherence Tomography Angiography Quality Across Three Multicenter Clinical Studies of Diabetic Retinopathy.

Purpose: To explore optical coherence tomography angiography (OCTA) quality and associated factors in multicenter clinical studies. Methods: OCTA scans were obtained from participants with diabetic retinopathy from three DRCR Retina Network clinical studies using the Optovue AngioVue and ZEISS AngioPlex. Macular (3 × 3 mm and 6 × 6 mm) and optic nerve scans were captured. Quality was assessed by the Casey Reading Center. Scans were considered "poor" if the signal strength index (SSI) was less than 55 (AngioVue) or 7 (AngioPlex) or if excess motion, media opacities, beam defocus, incorrect axial position, or other artifacts were present. Results: Included were 7539 scans from 787 eyes (461 participants). Sixty-one percent of scans were considered "good" (n = 4630). Of the 3 × 3-mm (n = 2294), 6 × 6-mm (n = 2705), and optic nerve scans (n = 2540), 62%, 63%, and 59%, respectively, were good. Differences in percentage of good scans by machine were not identified (61% of 6216 for the AngioVue and 63% of 1323 for the AngioPlex). The primary reason for poor scans was low SSI for the AngioVue (67%) and excess motion for the AngioPlex (47%). Good scans were associated with younger age (60 ± 12 years vs. 65 ± 11 years; P < 0.001), male gender (64% of males had good scans vs. 57% female; P = 0.007), and better visual acuity (ETDRS letter score 86.5 ± 6.4 [approximate Snellen equivalent 20/20] vs. 81.6 ± 9.7 [approximate Snellen equivalent 20/25]; P < 0.001). Conclusions: Scan quality or analysis must be improved for OCTA metrics to be used as outcomes in future research. Translational Relevance: Clinicians and researchers should be aware that poor SSI and artifacts are common issues for OCTA images.

Pneumatic Vitreolysis with Perfluoropropane for Vitreomacular Traction with and without Macular Hole: DRCR Retina Network Protocols AG and AH.

PURPOSE: To evaluate pneumatic vitreolysis (PVL) in eyes with vitreomacular traction (VMT) with and without full-thickness macular hole (FTMH). DESIGN: Two multicenter (28 sites) studies: a randomized clinical trial comparing PVL with observation (sham injection) for VMT without FTMH (Protocol AG) and a single-arm study assessing PVL for FTMH (Protocol AH). PARTICIPANTS: Participants were adults with central VMT (vitreomacular adhesion was ≤3000 µm). In Protocol AG, visual acuity (VA) was 20/32 to 20/400. In Protocol AH, eyes had a FTMH (≤250 µm at the narrowest point) and VA of 20/25 to 20/400. METHODS: Pneumatic vitreolysis using perfluoropropane (C3F8) gas. MAIN OUTCOME MEASURES: Central VMT release at 24 weeks (Protocol AG) and FTMH closure at 8 weeks (Protocol AH). RESULTS: From October 2018 through February 2020, 46 participants were enrolled in Protocol AG, and 35 were enrolled in Protocol AH. Higher than expected rates of retinal detachment and tear resulted in early termination of both protocols. Combining studies, 7 of 59 eyes (12% [95% CI, 6%-23%]; 2 eyes in Protocol AG, 5 eyes in Protocol AH) that received PVL developed rhegmatogenous retinal detachment (n = 6) or retinal tear (n = 1). At 24 weeks in Protocol AG, 18 of 23 eyes in the PVL group (78%) versus 2 of 22 eyes in the sham group (9%) achieved central VMT release without rescue vitrectomy (adjusted risk difference, 66% [95% CI, 44%-88%]; P< 0.001). The mean change in VA from baseline at 24 weeks was 6.7 letters in the PVL group and 6.1 letters in the sham group (adjusted difference, -0.8 [95% CI, -6.1 to 4.5]; P = 0.77). In Protocol AH, 10 of 35 eyes (29% [95% CI, 16%-45%]) achieved FTMH closure without rescue vitrectomy at 8 weeks. The mean change in VA from baseline at 8 weeks was -1.5 letters (95% CI, -10.3 to 7.3 letters). CONCLUSIONS: In most eyes with VMT, PVL induced hyaloid release. In eyes with FTMH, PVL resulted in hole closure in approximately one third of eyes. These studies were terminated early because of safety concerns related to retinal detachments and retinal tears.