Share Email Print
cover

Proceedings Paper • new

Deep learning-based label-free imaging flow cytometry for on-site analysis of water samples (Conference Presentation)

Paper Abstract

We introduce a field-portable and cost-effective holographic imaging flow cytometer, which provides phase contrast microscopic images of the contents of water samples at a throughput of 100 ml/h. This imaging cytometer uses a high power multi-colored LED and a custom designed circuit to illuminate continuously flowing water samples with short-pulses of red, green, and blue light that are simultaneously on, thereby eliminating motion blur and making the system vibration resistant. The recorded color holograms are segmented and reconstructed in real time and are phase recovered using a deep learning-based algorithm. Weighing 1kg with the dimensions of 15.5 cm × 15 cm × 12.5 cm, our label-free imaging flow-cytometer is controlled by a laptop computer equipped with a graphical processing unit. We tested the capabilities of our field-portable device by imaging micro- and nano-plankton inside ocean water samples collected at six beaches along the California coastline. We also determined Pseudo-Nitzschia algae concentration of these samples, providing a good agreement with the measurements made by the California Department of Public Health. Our device represents 1-2 orders of magnitude reduction in the cost and size of an imaging flow cytometer compared to state-of-the-art designs, while providing a similar or better performance in terms of volumetric throughput, detection limit and imaging resolution.

Paper Details

Date Published: 4 March 2019
PDF
Proc. SPIE 10869, Optics and Biophotonics in Low-Resource Settings V, 108690K (4 March 2019); doi: 10.1117/12.2508058
Show Author Affiliations
Zoltán S. Göröcs, Univ. of California, Los Angeles (United States)
Miu Tamamitsu, Univ. of California, Los Angeles (United States)
Vittorio Bianco, Univ. of California, Los Angeles (United States)
Patrick Wolf, Univ. of California, Los Angeles (United States)
Shounak Roy, Univ. of California, Los Angeles (United States)
Koyoshi Shindo, Univ. of California, Los Angeles (United States)
Kyrollos Yanny, Univ. of California, Los Angeles (United States)
Yichen Wu, Univ. of California, Los Angeles (United States)
Hatice Koydemir, Univ. of California, Los Angeles (United States)
Yair Rivenson, Univ. of California, Los Angeles (United States)
Aydogan Ozcan, Univ. of California, Los Angeles (United States)


Published in SPIE Proceedings Vol. 10869:
Optics and Biophotonics in Low-Resource Settings V
David Levitz; Aydogan Ozcan, Editor(s)

© SPIE. Terms of Use
Back to Top