Obstructive sleep apnea syndrome (OSA) and restless legs syndrome (RLS) with periodic limb movement during sleep (PLMS) are two sleep disorders characterized by repetitive respiratory or movement events associated with cortical arousals. We compared the cerebral hemodynamic changes linked to periodic apneas/hypopneas with arousals (AHA) in four OSA-patients with periodic limb movements (PLMA) with arousals in four patients with RLS-PLMS using near-infrared spectroscopy (NIRS). AHA induced homogenous pattern of periodic fluctuations in oxygenated (HbO2) and deoxygenated (HHb) hemoglobin, i.e., the decrease of HbO2 was accompanied by an increase of HHb during the respiratory event and resolved to reverse pattern when cortical arousal started. Blood volume (BV) showed the same pattern as HHb but with relative smaller amplitude in most of the AHA events.These changing patterns were significant as Wilcoxon signed-rank tests gave p < 0.001 when comparing the area under the curve of these hemodynamic parameters to zero. By contrast, in PLMA limb movements induced periodic increments in HbO2 and BV (Wilcoxon signed-rank tests, p < 0.001), but HHb changed more heterogeneously even during the events coming from the same patient. Heart rate (HR) also showed different patterns between AHA and PLMA. It significantly decreased during the respiratory event (Wilcoxon signed-rank test, p < 0.001) and then increased after the occurrence of cortical arousal (Wilcoxon signed-rank test, p < 0.001); while in PLMA HR first increased preceding the occurrence of cortical arousal (Wilcoxon signed-rank test, p < 0.001) and then decreased. The results of this preliminary study show that both AHA and PLMA induce changes in cerebral hemodynamics. The occurrence of cortical arousal is accompanied by increased HR in both events, but by different BV changes (i.e., decreased/increased BV in AHA/PLMA, respectively). HR changes may partially account for the increased cerebral hemodynamics during PLMA; whereas in AHA probable vasodilatation mediated by hypoxia/hypercapnia is more crucial for the post-arousal hemodynamics. The differences between changes of cerebral hemodynamics and HR may indicate different pathological mechanisms behind these two sleep disorder events.

In obstructive sleep apnea syndrome (OSA) the periodic reduction or cessation of breathing due to narrowing or occlusion of the upper airway during sleep leads to an impaired cerebral vascular autoregulation that is associated with an increased cardiovascular risk, including stroke. Continuous positive airways pressure (CPAP) therapy at night is the most e®ective treatment for OSA and has been shown to reduce the cardiovascular risk in OSA patients. However, there is no suitable bedside monitoring method evaluating the recovery of cerebral hemodynamics during CPAP therapy. Near-infrared spectroscopy (NIRS) is ideally suited for non-invasive monitoring the cerebral hemodynamics during sleep due to its properties of local measurement, totally safe application and good tolerance to motion. In this pilot study, we monitored cerebral hemodynamics during standard CPAP therapy at night in three patients with severe OSA using NIRS. We found periodic oscillations in HbO 2 , HHb, tissue oxygenation index (TOI) and blood volume (BV) associated with periodic apnea events without CPAP in all OSA patients. These oscillations were eliminated under the optimal CPAP pressures in all patients. These results suggested that the recovery of cerebral hemodynamics impaired by apnea events can be evaluated by bedside NIRS measurements in real time during all night CPAP therapy. NIRS is a useful bedside monitoring tool to evaluate the treatment e±cacy of CPAP therapy in patients with OSA.

With leaps in machine learning techniques and their application on Earth observation challenges has unlocked unprecedented performance across the domain. While the further development of these methods was previously limited by the availability and volume of sensor data and computing resources, the lack of adequate reference data is now constituting new bottlenecks. Since creating such ground-truth information is an expensive and error-prone task, new ways must be devised to source reliable, high-quality reference data on large scales. As an example, we showcase EUROCROPS, a reference dataset for crop type classification that aggregates and harmonizes administrative data surveyed in different countries with the goal of transnational interoperability.

No abstract