Abstract
Pollen studies play a critical role in various fields of science. In the last couple of decades, replacement of manual identification of pollen by image-based methods using pollen morphological features was a great leap forward, but challenges for pollen with similar morphology remain, and additional approaches are required. Spectroscopy approaches for identification of pollen, such as Raman spectroscopy has potential benefits over traditional methods, due to the investigation of the intrinsic molecular composition of a sample. However, current Raman-based characterization of pollen is complex and time-consuming, resulting in low throughput and limiting the statistical significance of the acquired data. Previously demonstrated high-throughput screening Raman spectroscopy (HTS-RS) eliminates the complexity as well as human interaction by incorporation full automation of the data acquisition process. Here, we present a customization of HTS-RS for pollen identification, enabling sampling of a large number of pollen in comparison to other state-of-the-art Raman pollen investigations. We show that using Raman spectra we are able to provide a preliminary estimation of pollen types based on growth habits using hierarchical cluster analysis (HCA) as well as good taxonomy of 37 different Pollen using principal component analysis-support vector machine (PCA-SVM) with good accuracy even for the pollen specimens sharing similar morphological features. Our results suggest that HTS-RS platform meets the demands for automated pollen detection making it an alternative method for research concerning pollen.
Highlights
IntroductionPollen are the male gametophyte of higher plants enabling fertilization of plants via pollination
Pollen are the male gametophyte of higher plants enabling fertilization of plants via pollination.The study of pollen can reveal the origin of flora and provide information about insect migration and climate change [1,2]
The unsupervised method hierarchical cluster analysis (HCA) estimated the pollen type based on the growth habit whereas the supervised method principal component analysis-support vector machine (PCA-support vector machine (SVM)) predicted pollen taxonomy
Summary
Pollen are the male gametophyte of higher plants enabling fertilization of plants via pollination. The study of pollen can reveal the origin of flora and provide information about insect migration and climate change [1,2]. Detection and analysis of pollen benefit the study of allergy-related respiratory health issues [3,4,5,6]. An important aspect of pollen research is to identify pollen specimens fast, reliably, and automatically [7,8]. The requirements demand consistent identification of a significant amount of pollen samples, while minimizing the intra and inter specimen bias. The bias is related to the deviations or variations that occur while identifying a particular pollen sample
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
