Evaluation of machine learning algorithms for classification of primary biological aerosol using a new UV-LIF spectrometer

Abstract

Abstract. Characterisation of bioaerosols has important implications within environment and public health sectors. Recent developments in ultraviolet light-induced fluorescence (UV-LIF) detectors such as the Wideband Integrated Bioaerosol Spectrometer (WIBS) and the newly introduced Multiparameter Bioaerosol Spectrometer (MBS) have allowed for the real-time collection of fluorescence, size and morphology measurements for the purpose of discriminating between bacteria, fungal spores and pollen.This new generation of instruments has enabled ever larger data sets to be compiled with the aim of studying more complex environments. In real world data sets, particularly those from an urban environment, the population may be dominated by non-biological fluorescent interferents, bringing into question the accuracy of measurements of quantities such as concentrations. It is therefore imperative that we validate the performance of different algorithms which can be used for the task of classification.For unsupervised learning we tested hierarchical agglomerative clustering with various different linkages. For supervised learning, 11 methods were tested, including decision trees, ensemble methods (random forests, gradient boosting and AdaBoost), two implementations for support vector machines (libsvm and liblinear) and Gaussian methods (Gaussian naïve Bayesian, quadratic and linear discriminant analysis, the k-nearest neighbours algorithm and artificial neural networks).The methods were applied to two different data sets produced using the new MBS, which provides multichannel UV-LIF fluorescence signatures for single airborne biological particles. The first data set contained mixed PSLs and the second contained a variety of laboratory-generated aerosol.Clustering in general performs slightly worse than the supervised learning methods, correctly classifying, at best, only 67.6 and 91.1 % for the two data sets respectively. For supervised learning the gradient boosting algorithm was found to be the most effective, on average correctly classifying 82.8 and 98.27 % of the testing data, respectively, across the two data sets.A possible alternative to gradient boosting is neural networks. We do however note that this method requires much more user input than the other methods, and we suggest that further research should be conducted using this method, especially using parallelised hardware such as the GPU, which would allow for larger networks to be trained, which could possibly yield better results.We also saw that some methods, such as clustering, failed to utilise the additional shape information provided by the instrument, whilst for others, such as the decision trees, ensemble methods and neural networks, improved performance could be attained with the inclusion of such information.

Systematic characterization and fluorescence threshold strategies for the wideband integrated bioaerosol sensor (WIBS) using size-resolved biological and interfering particles

“…Additionally, individuals utilizing WIBS instrumentation are cautioned to use the assignment of "biological aerosols" from UV-LIF measurements with great care and are rather encouraged to use "fluorescent aerosol" or some variation more liberally. Ultimately, further analysis methods, including clustering techniques (e.g., Crawford et al, 2015Crawford et al, , 2016Ruske et al, 2017), will likely need to employed to further improve discrimination between ambient particles and to reduce the relative rate of misidentification. It should also be noted, however, that a number of ambient studies have compared results of UV-LIF instruments with complementary techniques for bioaerosol detection and have reported favorable comparisons (Healy et al, 2014;Gosselin et al, 2016;.…”

Review: The Use of Real-Time Fluorescence Instrumentation to Monitor Ambient Primary Biological Aerosol Particles (PBAP)

“…Although being the first of the next generation of real-time techniques for airborne PBAP/FAP analysis to be developed, there is only one peer-reviewed publication that reports the use of an MBS [139]. The MBS is an evolution of the WIBS technology and provides data relating to the size (1-20 microns), shape, and autofluorescence of individual airborne particles.…”

“…The end-result is comparable to a traditional spectrum because individual fluorescing particles give rise to specific spectral distributions. When combined with the ability to determine FAP size and shape using an arrangement of two 512-pixel CMOS detector arrays to record high-resolution data, improved discrimination between biological and non-biological particles would appear to be possible with this instrument arrangement [139].…”

Identification of new particle formation events with deep learning

“…Ma et al (2015) used transfer learning in a SVM approach for classification of dust and clouds from satellite data. Other applications in atmospheric science include, for example, air quality predictions (Li et al, 2016;Ong et al, 2016), characterization of aerosol particles with an ultraviolet laser-induced fluorescence (UV-LIF) spectrometer (Ruske et al, 2017) and aerosol retrievals from ground-based observations (Di Noia et al, 2015, 2017. In addition, we recently applied different machine learning methods (e.g., neural network and SVM) for aerosol optical depth (AOD) retrieval from sun photometer data (Huttunen et al, 2016).…”