BACKGROUND:Ultrafine particles (≤100 nm diameter) may have a higher toxicity than larger particles but are still not regulated nor part of routine air pollution monitoring. So far, health effects of long-term exposure to ambient ultrafine particles are not well understood, owing to a lack of exposure data and epidemiological studies. METHODS:We conducted a systematic review and meta-analysis on the health effects of long-term exposure to ultrafine particles, including studies published until December 2024. A meta-analysis was conducted for outcomes with at least four available effect estimates. Confidence in the body of evidence was evaluated using the Office of Health Assessment and Translation method. RESULTS:We identified 85 studies investigating various mortality, morbidity and subclinical outcomes. In meta-analyses of single-pollutant models, we found positive associations with natural mortality (hazard ratio 1.06, 95% CI 1.04-1.08) and C-reactive protein (10.14% increase (95% CI -0.51-21.99%) per 10 000 pt·cm-3 increase in long-term exposure to ultrafine particles, with low and inadequate levels of evidence, respectively. The remaining studies revealed overall limited evidence for adverse effects on a wide range of outcomes. Less than half of the studies adjusted for co-pollutants. CONCLUSION:The evidence base on long-term health effects of ultrafine particles has increased substantially in the past decade, while the overall evidence for independent effects of long-term ultrafine particle exposure remains inadequate to low. More studies are needed to draw firm conclusions about the independent adverse effects of long-term ultrafine particles on various health end-points, with a special focus on the influence of co-pollutant adjustment.
Im Februar 1947 wurde Werner Catel, der zuvor in einer zentralen Funktion an der „NS-Kindereuthanasie“ mitgewirkt hatte, als ärztlicher Direktor der Landeskinderheilstätte Mammolshöhe angestellt. Unter seiner Leitung wurden dort kurz darauf Tuberkulose-Heilversuche durchgeführt. Hierbei kam es aufgrund der toxischen Wirkung des getesteten Präparats zu Todesfällen, dennoch wurden die Versuche fortgesetzt. Eine daraufhin eingeleitete Untersuchung konnte jedoch kein Fehlverhalten des Versuchsleiters erkennen, obwohl dieser gegen standesethische Grundsätze verstoßen hatte.
We characterize monic cubic CNS polynomials with only real roots in terms of relations between the other coefficients.
This paper presents a novel approach for data-driven self-learning control of highly flexible, modular manufacturing systems. Specifically, we employ a novel framework for model-based reinforcement learning which introduces approximate inverse process models within the training of reinforcement policies. This approach disentangles the learning of actuation dynamics and the dynamics in state space, resulting in RL-based training solely within the task space. We propose a lightweight feedforward architecture for approximate inverse models and integrate them within the policy network of standard RL algorithms. We apply the approach to a laboratory modular production testbed with heterogeneous production modules. The results underline the efficiency improvements for modular manufacturing units in terms of both performance and training speed, particularly for off-policy algorithms.
The adoption of Directive (EU) 2024/2881 on ambient air quality and cleaner air for Europe marks a paradigm shift in regulatory expectations: tighter pollutant limits—including new obligations for ultrafine particles (UFPs), elemental and organic carbon (EC/OC), and ammonia (NH3)—expanded monitoring requirements for hotspots and supersites, and the formal recognition of indicative and supplementary measurement methods alongside fixed reference stations. Against this backdrop, drones (often also termed uncrewed/unmanned aerial vehicles (UAVs), uncrewed/unmanned aerial systems (UASs), remotely piloted aircraft systems (RPASs) or multicopters) equipped with low-cost sensor (LCS) payloads have emerged as a technically promising but methodologically contested instrument for bridging spatial gaps on small scales in conventional monitoring networks. For readability, the term “drone” is used throughout this review interchangeably with uncrewed/unmanned aerial vehicle (UAV), uncrewed/unmanned aerial system (UAS), remotely piloted aircraft system (RPAS), and multicopter; no distinction in platform type, regulatory classification, or degree of autonomy is implied by the choice of term. This review critically synthesizes the peer-reviewed literature on drone-based air quality monitoring (AQM), with emphasis on its potential to support—rather than replace—stationary monitoring within the regulatory architecture established by the European Air Quality Directive. In line with the screening protocol adopted here, the evidence synthesized is centred on particulate matter and aerosol measurements; gas-phase pollutants are addressed only where co-reported by the included studies and are discussed as a secondary, contextual finding. A structured literature search of Web of Science (23 July 2026) using a three-component Boolean keyword string identified 1657 records (2016–2026). A four-criterion sequential screening protocol—requiring, in turn, (i) a thematic focus on particulate/aerosol air quality, (ii) evidence of an airborne sensor or in-flight concentration data, (iii) thematic relevance (excluding acronym and keyword coincidences), and (iv) an unambiguous, platform-attributable particulate/aerosol measurement—narrowed this to a corpus of 312 records. Full-text retrieval and assessment excluded a further 124 records found on full reading to be off-scope (e.g., records without any onboard particulate/aerosol measurement, gas-phase-only measurements, meteorology-only platforms, non-drone studies that had passed the keyword screen, or reviews without original data), leaving 188 records for methodological quality appraisal against a six-domain rubric (calibration transparency, co-location validation, uncertainty reporting, meteorological controls, repeatability/reproducibility, regulatory relevance; each scored 0–2, maximum 12) grounded in the philosophy of data quality objectives (DQOs) of CEN/TS 17660 (parts 1 and 2). A quality-gate threshold of ≥4 points, applied here as an inclusion criterion, yielded a final corpus of 164 studies for qualitative synthesis. These were classified into three analytical groups (platform and sensor performance, Group I, n = 44, 26.8%; spatial and vertical measurement applications, Group II, n = 129, 78.7%; regulatory and methodological integration, Group III, n = 35, 21.3%); 44 studies (26.8%) received multiple-group assignment. The synthesis reveals that while drone–LCS systems offer robust gradient detection and hotspot characterization capabilities, persistent challenges in absolute calibration accuracy, rotor-induced measurement artefacts, temporal representativeness, and the absence of standardized protocols limit their current applicability as indicative measurements under the DQO framework of CEN/TS 17660 (parts 1 and 2). Four development priorities are derived—in-flight DQO-validated calibration, source-type-specific flight and sampling protocols, temporal representativeness frameworks, and integration into EU assessment workflows—and translated into five standardization domains and operational recommendations for the regulatory use of drone-based aerosol measurements.
