Pollen clustering strategies using a newly developed single-particle fluorescence spectrometer

Swanson, Benjamin E.; Huffman, J. A.

doi:10.1080/02786826.2019.1711357

Cited by 12 publications

(6 citation statements)

References 84 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The resulting measurements from bioaerosol monitors are further analysed for the purpose of discriminating between bacteria, fungal spores and pollen and different AI based strategies are tested (Crawford et al, 2015;Pan, Huang, and Chang, 2012 ;Robinson et al, 2013;Ruske et al, 2017;Ruske et al, 2018;Swanson and Huffman, 2020). It was shown that for the task of airborne particle classification, clustering in general performs slightly worse than the supervised learning methods (Ruske et al 2017).…”

Section: Related Workmentioning

confidence: 99%

RealForAll: real-time system for automatic detection of airborne pollen

Tešendić

Krstićev

Matavulj

et al. 2020

Enterprise Information Systems

View full text Add to dashboard Cite

The aim of this paper is to describe a solution suitable for the automation of standard pollen information service (EN 16868:2019). We are describing the RealForAll integrated information system developed for automatic airborne pollen detection and real-time data delivery to end-users. This solution is based on the measurements from the Rapid-E airborne particle monitor. The system incorporates an AI-enabled subsystem based on a convolutional neural network that continuously retrieves raw data from Rapid-E and performs the classification of airborne pollen. The main advantages of this system reflect in real-time data delivery and independence of aerobiology experts during the pollen season.

show abstract

Section: Related Workmentioning

confidence: 99%

RealForAll: real-time system for automatic detection of airborne pollen

Tešendić

Krstićev

Matavulj

et al. 2020

Enterprise Information Systems

View full text Add to dashboard Cite

show abstract

“…Therefore, prior to the detection of pollen grains, an enrichment step of pollen in the air is generally required [13,14]. Traditional methods for the enrichment and analysis of air pollen grains are usually performed off-line, with the enrichment mainly relying on the adhesion of pollen particles to sticky surfaces, while both the identification and the counting of pollen particles are performed manually [15,16]. Early pollen enrichment systems such as the Durham pollen sampler [17] (Fig.…”

Section: Introductionmentioning

confidence: 99%

An integrated system for automated measurement of airborne pollen based on electrostatic enrichment and image analysis with machine vision

Yang

Klinkenberg

Pan

et al. 2022

Talanta

View full text Add to dashboard Cite

“…Clustering also becomes a convenient tool for analysing the possibilities of automatic monitoring and assessing aerobiological station arrangement [ 12 , 22 ]. Clustering particles of biological origin to analyse spectral data of single particles from a small test set of pollen types has a practical aspect, especially when seeking to improve the characteristics of fluorescence spectrometer sensors [ 23 ].…”

Section: Introductionmentioning

confidence: 99%

Clustering approach for the analysis of the fluorescent bioaerosol collected by an automatic detector

et al. 2021

View full text Add to dashboard Cite

Automatically operating particle detection devices generate valuable data, but their use in routine aerobiology needs to be harmonized. The growing network of researchers using automatic pollen detectors has the challenge to develop new data processing systems, best suited for identification of pollen or spore from bioaerosol data obtained near-real-time. It is challenging to recognise all the particles in the atmospheric bioaerosol due to their diversity. In this study, we aimed to find the natural groupings of pollen data by using cluster analysis, with the intent to use these groupings for further interpretation of real-time bioaerosol measurements. The scattering and fluorescence data belonging to 29 types of pollen and spores were first acquired in the laboratory using Rapid-E automatic particle detector. Neural networks were used for primary data processing, and the resulting feature vectors were clustered for scattering and fluorescence modality. Scattering clusters results showed that pollen of the same plant taxa associates with the different clusters corresponding to particle shape and size properties. According to fluorescence clusters, pollen grouping highlighted the possibility to differentiate Dactylis and Secale genera in the Poaceae family. Fluorescent clusters played a more important role than scattering for separating unidentified fluorescent particles from tested pollen. The proposed clustering method aids in reducing the number of false-positive errors.

show abstract

Pollen clustering strategies using a newly developed single-particle fluorescence spectrometer

Cited by 12 publications

References 84 publications

RealForAll: real-time system for automatic detection of airborne pollen

RealForAll: real-time system for automatic detection of airborne pollen

An integrated system for automated measurement of airborne pollen based on electrostatic enrichment and image analysis with machine vision

Clustering approach for the analysis of the fluorescent bioaerosol collected by an automatic detector

Contact Info

Product

Resources

About