EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Biological contaminants analysis in microalgae culture by UV–vis spectroscopy and machine learning

Eduardo Maia Paiva, Eevi Hyttinen, Timo Dönsberg, Dorothee Barth

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy · 2025-04 · 330 · 125690

DOI: 10.1016/j.saa.2024.125690 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2025-04. Verfasst von Eduardo Maia Paiva und 3 weiteren Autor:innen. Beteiligte Einrichtung: VTT Technical Research Centre of Finland, Tekniikantie 1, 02150 Espoo, Finland. Electronic address: eduardo.maiapaiva@vtt.fi. Hauptthemen laut MEDLINE: Microalgae, Machine Learning. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European Commission, Business Finland.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study elucidates the utility and efficacy of UV-visible spectroscopy for the detection and characterization of biological contaminants within microalgae cultures, augmented by machine learning algorithms. Biological contamination, exemplified by flagellates and rotifers, poses a significant concern due to its potential to rapidly devastate entire cultures, thus jeopardizing commercial viability. Conventional analytical methods for monitoring contamination, such as microscopy and cytometry, are often labor-intensive, reliant on specialized expertise for microorganism identification, and may lack specificity in discerning the nature of contamination, impeding timely intervention. UV-visible spectroscopy offers a compelling solution by overcoming many of these challenges, affording specificity in analysis, real-time monitoring capabilities, and automation, owing to the intricate pigment chemistry inherent in the microalgae realm, which generates distinct UV-visible spectra. Through the measuring of contaminated and uncontaminated samples, coupled with machine learning analysis of their respective spectra, this study explores the underlying biochemical principles driving spectral data, thereby justifying the efficacy of the technique. The findings underscore the wealth of information encapsulated within UV-visible spectral data, which can be effectively harnessed through classification algorithms for early-stage identification of contamination in real-time applications.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MicroalgaeMachine LearningSpectrophotometry, Ultraviolet

Stichwörter der Autor:innen

Biological contaminantsFlagellateMachine learningMicroalgaeRotiferUV–visible spectroscopy

Autor:innen und Institutionen

  • VTT Technical Research Centre of Finland, Tekniikantie 1, 02150 Espoo, Finland. Electronic address: eduardo.maiapaiva@vtt.fi.
  • VTT Technical Research Centre of Finland Ltd, Tietotie 2, P.O. Box 1000, 02044 Espoo, Finland.
  • VTT Technical Research Centre of Finland, Tekniikantie 1, 02150 Espoo, Finland.

Förderung

  • European Commission
  • Business Finland

Zitieren

APA

Paiva, E. M., Hyttinen, E., Dönsberg, T., & Barth, D. (2025). Biological contaminants analysis in microalgae culture by UV–vis spectroscopy and machine learning. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 330, 125690. https://doi.org/10.1016/j.saa.2024.125690

Vancouver

Paiva EM, Hyttinen E, Dönsberg T, Barth D. Biological contaminants analysis in microalgae culture by UV–vis spectroscopy and machine learning. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy. 2025;330:125690. doi:10.1016/j.saa.2024.125690

BibTeX
@article{Paiva2025,
  author = {Paiva, Eduardo Maia and Hyttinen, Eevi and Dönsberg, Timo and Barth, Dorothee},
  title = {Biological contaminants analysis in microalgae culture by UV–vis spectroscopy and machine learning},
  journal = {Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy},
  year = {2025},
  volume = {330},
  pages = {125690},
  doi = {10.1016/j.saa.2024.125690}
}
Kostenlos starten →