Development of a Hyperspectral Optical System for Flow Cytometry
Filipchuk, Anna (2023)
2023
Master's Programme in Biomedical Sciences and Engineering
Lääketieteen ja terveysteknologian tiedekunta - Faculty of Medicine and Health Technology
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-06-27
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202306236899
https://urn.fi/URN:NBN:fi:tuni-202306236899
Tiivistelmä
Flow cytometry is a powerful method of biomedical imaging. It is being actively utilized in immunophenotyping, cell sorting, and cancer diagnostics among other applications. Flow cytometry market is rapidly developing and new approaches to signal collection are invented and taken into utilization. Among the latest trends is introduction of imaging into flow cytometrical devices to increase the information content of the studied sample.
However, implementation of an additional imaging unit in addition to existing optical subsystem increases the bulk of the instrument. A possible approach not only to decrease the size of the device but also to increase the information content of the recorded samples is the implementation of the hyperspectral imaging to flow cytometry.
In the course of this work, an optical setup for evaluating particles imaging with hyperspectral camera is established. A basic fluidics setup, not featuring hydrodynamic focusing, is also configured for testing purposes. Particles flowing through the flow cell are imaged with alternative cameras to compare on the quality and information content of the recorded data.
Imaging is performed with red-green-blue and hyperspectral cameras. The setup is configured in a way to allow acquisition of images with both imaging methods simultaneously for more direct comparison. In addition to mentioned methods, the studied sample is imaged with multimode plate reader to offer a comprehensive comparison of self-build imaging solutions to imaging method available in the market.
The results reveal the possibility to localise imaged particles and differentiate those from the surrounding fluid based on the spectra recorded with hyperspectral camera. It has also been demonstrated achievable to differentiate between different intensity levels of recorded particles for hyperspectral recordings, a capability not exhibited by alternative imaging methods in this research.
The peak of the imaged fluorophore emission, however, has either been shifted in the recorded images or not identifiable due to having lower intensity than the excitation laser light. Additional investigation is required to confirm the nature of the recoded spectrum shift.
Based on the obtained results, further research could focus on investigating techniques to distinguish the fluorophore signal from the background noise. Those could possibly involve filtering implementation coupled with refinement of the flow cell configuration. Potentially, multiple sample cells could be imaged simultaneously with this method, which would increase its flow-through.
However, implementation of an additional imaging unit in addition to existing optical subsystem increases the bulk of the instrument. A possible approach not only to decrease the size of the device but also to increase the information content of the recorded samples is the implementation of the hyperspectral imaging to flow cytometry.
In the course of this work, an optical setup for evaluating particles imaging with hyperspectral camera is established. A basic fluidics setup, not featuring hydrodynamic focusing, is also configured for testing purposes. Particles flowing through the flow cell are imaged with alternative cameras to compare on the quality and information content of the recorded data.
Imaging is performed with red-green-blue and hyperspectral cameras. The setup is configured in a way to allow acquisition of images with both imaging methods simultaneously for more direct comparison. In addition to mentioned methods, the studied sample is imaged with multimode plate reader to offer a comprehensive comparison of self-build imaging solutions to imaging method available in the market.
The results reveal the possibility to localise imaged particles and differentiate those from the surrounding fluid based on the spectra recorded with hyperspectral camera. It has also been demonstrated achievable to differentiate between different intensity levels of recorded particles for hyperspectral recordings, a capability not exhibited by alternative imaging methods in this research.
The peak of the imaged fluorophore emission, however, has either been shifted in the recorded images or not identifiable due to having lower intensity than the excitation laser light. Additional investigation is required to confirm the nature of the recoded spectrum shift.
Based on the obtained results, further research could focus on investigating techniques to distinguish the fluorophore signal from the background noise. Those could possibly involve filtering implementation coupled with refinement of the flow cell configuration. Potentially, multiple sample cells could be imaged simultaneously with this method, which would increase its flow-through.