Automated Classification of Airborne Pollen using Neural Networks.

Pollen allergies are considered as a global epidemic nowadays, as they influence more than a quarter of the worldwide population, with this percentage expected to rapidly increase because of ongoing climate change. To date, alerts on high-risk allergenic pollen exposure have been provided only via forecasting models and conventional monitoring methods that are laborious. The aim of this study is to develop and evaluate our own pollen classification model based on deep neural networks. Airborne allergenic pollen have been monitored in Augsburg, Bavaria, Germany, since 2015, using a novel automatic Bio-Aerosol Analyzer (BAA 500, Hund GmbH). The automatic classification system is compared and evaluated against our own, newly developed algorithm. Our model achieves an unweighted average precision of 83.0 % and an unweighted average recall of 77.1 % across 15 classes of pollen taxa. Automatic, real-time information on concentrations of airborne allergenic pollen will significantly contribute to the implementation of timely, personalized management of allergies in the future. It is already clear that new methods and sophisticated models have to be developed so as to successfully switch to novel operational pollen monitoring techniques serving the above need.

Human exposure to airborne pollen and relationships with symptoms and immune responses: Indoors versus outdoors, circadian patterns and meteorological effects in alpine and urban environments.

Pollen exposure is a major cause of respiratory allergies worldwide. However, it is unclear how everyday exposure is related to symptoms and how allergic patients may be affected spatially and temporally. Hence, we investigated the relationship of pollen, symptoms and immune responses under a controlled regime of 'high-low-moderate' pollen exposure in urban versus alpine environment. The research was conducted in 2016 in two locations in Germany: urban Augsburg (494 m) and Schneefernerhaus (UFS) on Zugspitze mountain (2656 m). Monitoring of airborne pollen took place using Hirst-type volumetric traps. On UFS, both indoor and outdoor samples were taken. Grass pollen allergic human volunteers were monitored daily during the peak of the grass pollen season, in Augsburg, on UFS, then again in Augsburg. Nasal biosamples were obtained throughout the study to investigate immune responses. All symptoms decreased significantly during the stay on UFS and remained low even after the return to Augsburg. The same was observed for nasal total IgE and IgM levels and for nasal type 2 cytokines and chemokines. Augsburg showed higher pollen concentrations than those on UFS. At all sites, pollen were present throughout each day, but were more abundant in Augsburg during morning. On UFS, outdoor pollen levels were up to 6-fold higher than those indoors. Nasal, ocular and pulmonary symptoms correlated with current and previous days' pollen concentrations and relative humidity. Stays in low-exposure environments during the peak pollen season can be an efficient means of reducing allergic symptoms and immune responses. However, in alpine environments, even occasional pollen exposure during short intervals may still trigger symptoms because of the additional environmental stress posed onto allergics. This highlights the need for the consideration of additional environmental factors, apart from symptom diaries and immune responses, so as to efficiently predict high-risk allergy periods.

Pollen-Associated Microbiome Correlates with Pollution Parameters and the Allergenicity of Pollen.

Pollen allergies have been rapidly increasing over the last decades. Many allergenic proteins and non-allergenic adjuvant compounds of pollen are involved in the plant defense against environmental or microbial stress. The first aim of this study was to analyze and compare the colonizing microbes on allergenic pollen. The second aim was to investigate detectable correlations between pollen microbiota and parameters of air pollution or pollen allergenicity. To reach these aims, bacterial and fungal DNA was isolated from pollen samples of timothy grass (Phleum pratense, n = 20) and birch trees (Betula pendula, n = 55). With this isolated DNA, a terminal restriction fragment length polymorphism analysis was performed. One result was that the microbial diversity on birch tree and timothy grass pollen samples (Shannon/Simpson diversity indices) was partly significantly correlated to allergenicity parameters (Bet v 1/Phl p 5, pollen-associated lipid mediators). Furthermore, the microbial diversity on birch pollen samples was correlated to on-site air pollution (nitrogen dioxide (NO2), ammonia (NH3), and ozone (O3)). What is more, a significant negative correlation was observed between the microbial diversity on birch pollen and the measured NO2 concentrations on the corresponding trees. Our results showed that the microbial composition of pollen was correlated to environmental exposure parameters alongside with a differential expression of allergen and pollen-associated lipid mediators. This might translate into altered allergenicity of pollen due to environmental and microbial stress.