Automatic evaluation of graft orientation during Descemet membrane endothelial keratoplasty using intraoperative OCT.

Correct Descemet Membrane Endothelial Keratoplasty (DMEK) graft orientation is imperative for success of DMEK surgery, but intraoperative evaluation can be challenging. We present a method for automatic evaluation of the graft orientation in intraoperative optical coherence tomography (iOCT), exploiting the natural rolling behavior of the graft. The method encompasses a deep learning model for graft segmentation, post-processing to obtain a smooth line representation, and curvature calculations to determine graft orientation. For an independent test set of 100 iOCT-frames, the automatic method correctly identified graft orientation in 78 frames and obtained an area under the receiver operating characteristic curve (AUC) of 0.84. When we replaced the automatic segmentation with the manual masks, the AUC increased to 0.92, corresponding to an accuracy of 86%. In comparison, two corneal specialists correctly identified graft orientation in 90% and 91% of the iOCT-frames.

Diagnostic pars plana vitrectomy and aqueous analyses in patients with uveitis of unknown cause.

PURPOSE: To compare the yield of diagnostic pars plana vitrectomy (PPV) with the yield of aqueous analyses in patients with uveitis of unknown cause. METHODS: Seventy-five consecutive patients (84 eyes) with uveitis involving posterior eye segment who undergo a diagnostic PPV from 2005 through 2009 were retrospectively reviewed. Vitreous specimens were simultaneously analyzed by microbiological culture, flow cytometry, and cytology as well as by polymerase chain reaction and for intraocular antibody production by Goldmann-Witmer coefficient. In 53 eyes, both aqueous and vitreous samples were assessed. The primary outcome measure was the comparison between vitreous and aqueous analyses. RESULTS: Vitreous analysis was positive in 18 of 84 eyes (21%). Positive results indicated infectious uveitis in 12 of 18 cases (67%) and lymphoma in 6 of 18 (33%) cases. Of the 53 eyes with both aqueous and vitreous samples available, aqueous analysis revealed the diagnosis in 6 of 53 eyes and vitreous in 9 of 53 eyes. Unilateral uveitis (P = 0.022), panuveitis and uveitis posterior (P ≤ 0.001), preoperative immunosuppressive therapy (P = 0.004), and increasing age (P = 0.018) were associated with an increased diagnostic yield of PPV. Overall, 1 year after PPV, median visual acuity improved from 20/200 to 20/80 (Snellen, P ≤ 0.001). Of 18 patients who were on immunosuppressive treatment before PPV, 8 (44%) were able to stop immunosuppressive therapy during 1-year follow-up. The complications of PPV consisted predominantly of cataract development (33/65, 51%). CONCLUSION: Diagnostic PPV with the analysis of vitreous fluid by multiple laboratories for infectious and malignant disorders was useful in diagnosing uveitis of unknown cause. Previous aqueous analysis was especially valuable for the diagnosis of intraocular infections and may therefore decrease the number of patients who would otherwise undergo an invasive diagnostic PPV. Furthermore, PPV was associated with improved visual acuity and decreased use of immunosuppressive therapy.