Sheathless inertial cell ordering for extreme throughput flow cytometry

TLDR

Cell type differentiation in heterogeneous solutions is critical for research and medicine, yet conventional flow cytometry, the gold standard for blood analysis, is limited by serial processing and insufficient throughput to detect rare cells. This study exploits inertial effects for label‑ and sheath‑free parallel flow cytometry with extreme throughput. A microfluidic device with 256 high‑aspect parallel channels delivers up to 1 million cells s⁻¹, focusing cells to a uniform z‑position and velocity to minimize overlap and blur for accurate imaging. Proof‑of‑concept tests on diluted whole blood achieved 86–97 % counting sensitivity and specificity for RBCs and leukocytes, demonstrating the system’s potential for cost‑effective, high‑throughput hematology and rare‑cell analysis.

Abstract

Rapid and accurate differentiation of cell types within a heterogeneous solution is a challenging but important task for various applications in biological research and medicine. Flow cytometry is the gold standard in cell analysis and is regularly used for blood analysis (i.e., complete blood counts). Flow cytometry, however, lacks sufficient throughput to analyze rare cells in blood or other dilute solutions in a reasonable time period because it is an inherently serial process. In this study, we exploit inertial effects for label- and sheath-free parallel flow cytometry with extreme throughput. We demonstrate a microfluidic device that consists of 256 high-aspect (W = 16 µm, H = 37 µm) parallel channels yielding a sample rate up to 1 million cells s−1, only limited by the field-of-view of our high-speed optical interrogation method. The particles or cells flowing through the channels are focused to one uniform z-position (SD = ±1.81 µm) with uniform downstream velocity (Uave = 0.208 ± 0.004 m s−1) to reduce the probability of overlap and out-of-focus blur and provide similar cell signature images for accurate detection and analysis. To demonstrate a proof-of-concept application of our system operating at these throughputs, we conducted automated RBC and leukocyte counts on diluted whole blood and achieved high counting sensitivity and specificity (86–97%) compared to visual inspection of raw images. As no additional external forces are required to create ordered streams of cells, this approach has the potential for future applications in cost-effective hematology or rare-cell analysis platforms with extreme throughput capabilities when integrated with suitable large field-of view imaging or interrogation methods.