A Laboratory Evaluation of the New Automated Pollen Sensor Beenose: Pollen Discrimination Using Machine Learning Techniques

Abstract

The monitoring of airborne pollen has received much attention over the last decade, as the prevalence of pollen-induced allergies is constantly increasing. Today, the most common technique to identify airborne pollen species and to monitor their concentrations is based on manual analysis. Here, we present a new, low-cost, real-time optical pollen sensor, called Beenose, that automatically counts and identifies pollen grains by performing measurements at multiple scattering angles. We describe the data pre-processing steps and discuss the various statistical and machine learning methods that have been implemented to distinguish different pollen species. The analysis is based on a set of 12 pollen species, several of which were selected for their allergic potency. Our results show that Beenose can provide a consistent clustering of the pollen species based on their size properties, and that pollen particles can be separated from non-pollen ones. More importantly, 9 out of 12 pollen species were correctly identified with a prediction score exceeding 78%. Classification errors occur for species with similar optical behaviour, suggesting that other parameters should be considered to provide even more robust pollen identification.

Current state and prospects of artificial intelligence in allergy

“…Traditionally, pollen monitoring has relied on manual labeling methods, that, while accurate, can be labor-intensive and timeconsuming. 151 In contrast, public weather websites now incorporate real-time pollen counts sourced from research institutes that provide AI-based pollen counts. 152 Recently, the BAA500 monitoring system was introduced, which automates the counting using microscopy and a supervised image recognition algorithm, 153 reaching a 90% detection rate within a 3-h sampling time.…”

“…Another example is the beenose software, which uses a classification model based on optical signatures of pollen. 151 Besides detection, ML has also been used to forecast future pollen concentrations. 154,155 Cordero et al 155 combined various models to predict the Olea pollen concentrations in the Community of Madrid.…”

“…AI has also attracted attention for the monitoring of airborne pollen and allergens. Traditionally, pollen monitoring has relied on manual labeling methods, that, while accurate, can be labor‐intensive and time‐consuming 151 . In contrast, public weather websites now incorporate real‐time pollen counts sourced from research institutes that provide AI‐based pollen counts 152 .…”

“…Recently, the BAA500 monitoring system was introduced, which automates the counting using microscopy and a supervised image recognition algorithm, 153 reaching a 90% detection rate within a 3‐h sampling time. Another example is the beenose software, which uses a classification model based on optical signatures of pollen 151 . Besides detection, ML has also been used to forecast future pollen concentrations 154,155 .…”

Spatial Variation of Airborne Pollen Concentrations Locally around Brussels City, Belgium, during a Field Campaign in 2022–2023, Using the Automatic Sensor Beenose

“…The same principle has also been used to develop optical counters with typology determination capabilities, to distinguish between anthropogenic pollution particles, natural mineral particles, and pollens [ 31 , 32 ]. Beenose has already been tested in laboratory conditions, demonstrating its capacity to discriminate pollen particles from other particles and to characterize some pollen taxa [ 33 ].…”