Elastic Deformation of Optical Coherence Tomography Images of Diabetic Macular Edema for Deep-Learning Models Training: How Far to Go?

- Objective: To explore the clinical validity of elastic deformation of optical coherence tomography (OCT) images for data augmentation in the development of deep-learning model for detection of diabetic macular edema (DME). METHODS: Prospective evaluation of OCT images of DME (n = 320) subject to elastic transformation, with the deformation intensity represented by ([Formula: see text]). Three sets of images, each comprising 100 pairs of scans (100 original & 100 modified), were grouped according to the range of ([Formula: see text]), including low-, medium- and high-degree of augmentation; ([Formula: see text] = 1-6), ([Formula: see text] = 7-12), and ([Formula: see text] = 13-18), respectively. Three retina specialists evaluated all datasets in a blinded manner and designated each image as 'original' versus 'modified'. The rate of assignment of 'original' value to modified images (false-negative) was determined for each grader in each dataset. RESULTS: The false-negative rates ranged between 71-77% for the low-, 63-76% for the medium-, and 50-75% for the high-augmentation categories. The corresponding rates of correct identification of original images ranged between 75-85% ([Formula: see text]0.05) in the low-, 73-85% ([Formula: see text]0.05 for graders 1 & 2, p = 0.01 for grader 3) in the medium-, and 81-91% ([Formula: see text]) in the high-augmentation categories. In the subcategory ([Formula: see text] = 7-9) the false-negative rates were 93-83%, whereas the rates of correctly identifying original images ranged between 89-99% ([Formula: see text]0.05 for all graders). CONCLUSIONS: Deformation of low-medium intensity ([Formula: see text] = 1-9) may be applied without compromising OCT image representativeness in DME. Clinical and Translational Impact Statement-Elastic deformation may efficiently augment the size, robustness, and diversity of training datasets without altering their clinical value, enhancing the development of high-accuracy algorithms for automated interpretation of OCT images.

Cystoid macular edema secondary to ibrutinib.

PURPOSE: To describe a unique case of cystoid macular edema associated with Ibrutinib treatment for Chronic Lymphocytic Leukemia (CLL). OBSERVATIONS: A 73-year-old male patient presented to the ophthalmology clinic complaining of decreased vision in his seeing-eye ('only eye', left). Further clinal examination and imaging revealed the presence of a cystoid macular edema (CME). With no apparent cause to this condition, topical treatment with NSAIDS and steroids continued over two years with only partial response and persistent macular edema, resulting in decreased vision. Cessation of Ibrutinib treatment resulted in resolution of the macular edema and improvement in visual acuity over 6 months. CONCLUSIONS AND IMPORTANCE: Several novel oncologic therapies have been associated with CME in recent years. This case demonstrates an association between Ibrutinib an oral, irreversible inhibitor of Bruton's Tyrosine Kinase (BTK), and the development of CME. CME was resistant to topical treatment but resolved after treatment cessation. Along with two previous cases reported, this case suggests that CME is a rare adverse event of Ibrutinib therapy. Screening for CME in Ibrutinib treated patients who report visual symptoms should be considered.

Disorganization of the Retinal Inner Layers as a Predictor of Visual Acuity in Eyes With Macular Edema Secondary to Vein Occlusion.

PURPOSE: To determine whether spectral-domain optical coherence tomography (SD-OCT) disorganization of the retinal inner layers (DRIL) is predictive of best-corrected visual acuity (BCVA) in retinal vein occlusion macular edema (RVO-ME). DESIGN: Retrospective cohort study. METHODS: Setting: Two tertiary centers. PATIENTS: Treatment-naïve RVO-ME receiving monthly intravitreal bevacizumab. OBSERVATION PROCEDURES: Spectral-domain optical coherence tomography (SD-OCT) images from baseline, 4-month, and 8-month visits. The DRIL extent and additional parameters were evaluated in the 1-mm-wide foveal centered area. MAIN OUTCOME MEASURES: BCVA at each visit. RESULTS: A total of 136 eyes were included in this study. Greater DRIL extent at baseline correlated with worse baseline BCVA (point estimate, 0.04; 95% CI, 0.01-0.07 per 100 µm, P = .003). In multivariate analysis extent of DRIL (P = .03) and ellipsoid zone disruption (P < .001) correlated with baseline BCVA. Four-month DRIL reduction was associated with 8-month BCVA improvement (point estimate, 0.02; 95% CI, 0.01-0.04, P = .01), with only DRIL remaining significant in multivariate analysis (P = .02). Baseline DRIL predicted 8-month improvement in BCVA (point estimate, 0.03; 95% CI, 0.01-0.05, P = .04), with only DRIL remaining significant in multivariate analysis (P = .006). Eight-month DRIL change was associated with 8-month BCVA change (point estimate, 0.03, 95% CI 0.01-0.05, P = .001), with only DRIL remaining significant in multivariate analysis (P = .002). CONCLUSIONS: The change of the DRIL extent following the first 3 monthly injections identifies eyes with a high likelihood of subsequent BCVA improvement or decline. Therefore, the extent of DRIL before and after treatment is an additional SD-OCT parameter that may serve as a biomarker for patients with RVO-ME.

Haptoglobin phenotype in age-related macular degeneration patients.

PURPOSE: To investigate a possible role of the haptoglobin phenotype in the development of exudative age-related macular degeneration (AMD) in human subjects. DESIGN: Prospective, observational, comparative population study. METHODS: The study was carried out in an institutional setting. All patients referred because of exudative AMD in one eye during an 18-month period were included in the study group. A group of patients treated for other ocular diseases and not having AMD in either eye served as control. Haptoglobin phenotype was determined from a blood sample drawn from each patient in both the study and control groups. The main outcome measure was the distribution of the haptoglobin phenotype in the study and control group. RESULTS: One hundred eighty-five participants were included in the study. Ninety-eight had exudative AMD, and 87 were AMD-free. The difference between the study and control groups in distribution of the haptoglobin phenotype was found to be statistically insignificant. CONCLUSIONS: Our results suggest that the haptoglobin phenotype has no effect on the prevalence of exudative AMD.

Anatomic and functional results of vitrectomy and long-term intraocular tamponade for stage 2 macular holes.

PURPOSE: To investigate the results of pars plana vitrectomy combined with long-term intraocular tamponade in patients with a stage 2 macular hole. METHODS: In a retrospective study, 50 consecutive eyes operated on for stage 2 macular hole were reviewed. The surgical technique included pars plana vitrectomy, separation of the posterior hyaloid, and intraocular tamponade with either 12.5% perfluropropane (C(3)F(8)) gas or silicone oil. Patients treated with C(3)F(8) were instructed to keep face-down positioning for 3 or 4 weeks. Removal of silicone oil took place 4 or 5 weeks after the vitrectomy. RESULTS: Closure of the macular hole was achieved in all 50 eyes after one operation. The mean postoperative best-corrected distance visual acuity was 20/32 (range, 20/63-20/20), with 49 eyes (98%) having a postoperative visual acuity of 20/50 or better. Visual acuity improved in all eyes postoperatively, with a mean gain of 4.84 +/- 1.95 ETDRS lines. CONCLUSION: For stage 2 macular holes, vitrectomy combined with long-term intraocular tamponade can result in a very favorable anatomic and functional outcome that is as good as or better than the results described for other modalities. This approach simplifies the surgical technique, and both adjuvant-related complications and potential retinal damage related to internal limiting membrane peeling or use of indocyanine green are avoided.