Search

PurposeWe recently showed how a refined sequential detection pattern and image processing pipeline for multi-offset adaptive optics scanning light ophthalmoscopy (AOSLO) can increase the contrast of weakly scattering inner retinal structures, including microglia. However, sequential detection was still time-consuming, preventing dynamics from being monitored over short intervals (< 3 mins). Here we show that simultaneous fiber-bundle (FB) detection can overcomes this limitation to reveal the structure and dynamic activity of microglia and macrophage-like cells in healthy and diseased retinae.MethodsWe designed and implemented a custom 7-fiber optical FB with one central confocal fiber and six larger fibers for multi-offset detection in AOSLO at a single focal plane. We imaged the ganglion cell layer at several locations at multiple timepoints (from minutes to weeks) in 8 healthy participants and in 4 patients with ocular infections or inflammation, including ocular syphilis and posterior uveitis. Microglia and immune cells were manually segmented to quantify cell morphometry and motility.ResultsFiber-bundle detection reduced single acquisition time to 20-30 seconds, enabling imaging over larger areas and monitoring of dynamics over shorter intervals. Presumed microglia in healthy retinas had an average diameter of 12.8 μm and with a spectrum of morphologies including circular cells and elongated cells with visible processes. We also detected the somas of putative macroglia, potentially astrocytes, near the optic nerve head. Microglia moved slowly in normal eyes (0.02μm/sec, on average) but speed increased in patients with active infections or inflammation (up to 2.37μm/sec). Microglia activity was absent in a patient with chronic uveitis that was quiescent but apparent over short intervals in an active uveitis retina. In a patient with ocular syphilis imaged at multiple timepoints during treatment, macrophage-like cells containing granular internal structures were seen. Decreases in the quantity and motility of these immune cells were correlated with improvements to vision and other structural and systemic biomarkers.ConclusionsFB-AOSLO enable simplified optical setup with easy alignment and implementation. We imaged the fine-scale structure and dynamics of microglia and macrophage-like cells during active infection and inflammation in the living eye for the first time. In healthy eyes we also detected putative glia cell near the optic nerve head. FB-AOSLO offers promise as a powerful tool for detecting and monitoring retinal inflammation and infection in the living eye over short response to treatment.

Suppression or mitigation of edge-localized mode (ELM) crashes is necessary for ITER. The strategy to suppress all the ELM crashes by the resonant magnetic perturbation (RMP) should be applied as soon as the first low-to-high confinement (L–H) transition occurs. A control algorithm based on real-time machine learning (ML) enables such an approach: it classifies the H-mode transition and the ELMy phase in real-time and automatically applies the preemptive RMP. This paper reports the algorithm design, which is now implemented in the KSTAR plasma-control system, and the corresponding experimental demonstration of typical high-δ KSTAR H-mode plasmas. As a result, all initial ELM crashes are suppressed with an acceptable safety factor at the edge (q 95) and with RMP field adjustment. Moreover, the ML-driven ELM crash suppression discharges remain stable without further degradation due to the regularization of the plasma pedestal.