We present a deep learning-based high-throughput cytometer to detect rare cells in whole blood using a cost-effective and light-weight design. This system uses magnetic-particles to label and enrich the target cells. Then, a periodically-alternating magnetic-field creates time-modulated diffraction patterns of the target cells that are recorded using a lensless microscope. Finally, a custom-designed convolutional network is used to detect and classify the target cells based on their modulated spatio-temporal patterns. This cytometer was tested with cancer cells spiked in whole blood to achieve a limit-of-detection of 10 cells/mL. This compact, cost-effective and high-throughput cytometer might serve diagnostics needs in resource-limited-settings.
We present a high-throughput and cost-effective computational cytometer for rare cell detection, where the target cells are specifically labeled with magnetic particles and exhibit an oscillatory motion under a periodically-changing magnetic field. The time-varying diffraction patterns of the oscillating cells are then captured with a holographic imaging system and are further classified by a customized pseudo-3D convolutional network. To evaluate the performance of our technique, we detected serially-diluted MCF7 cancer cells that were spiked in whole blood, achieving a limit of detection (LoD) of 10 cells per 1 mL of whole blood.
Research laboratories and the industry rely on yeast viability and concentration measurements to adjust fermentation parameters such as pH, temperature, and pressure. Beer-brewing processes as well as biofuel production can especially utilize a cost-effective and portable way of obtaining data on cell viability and concentration. However, current methods of analysis are relatively costly and tedious. Here, we demonstrate a rapid, portable, and cost-effective platform for imaging and measuring viability and concentration of yeast cells. Our platform features a lens-free microscope that weighs 70 g and has dimensions of 12 × 4 × 4 cm. A partially-coherent illumination source (a light-emitting-diode), a band-pass optical filter, and a multimode optical fiber are used to illuminate the sample. The yeast sample is directly placed on a complementary metal-oxide semiconductor (CMOS) image sensor chip, which captures an in-line hologram of the sample over a large field-of-view of >20 mm2. The hologram is transferred to a touch-screen interface, where a trained Support Vector Machine model classifies yeast cells stained with methylene blue as live or dead and measures cell viability as well as concentration. We tested the accuracy of our platform against manual counting of live and dead cells using fluorescent exclusion staining and a bench-top fluorescence microscope. Our regression analysis showed no significant difference between the two methods within a concentration range of 1.4 × 105 to 1.4 × 106 cells/mL. This compact and cost-effective yeast analysis platform will enable automatic quantification of yeast viability and concentration in field settings and resource-limited environments.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.