Typing of acute leukemia by intelligent optical time-stretch imaging flow cytometry on a chip

Abstract

Acute leukemia (AL) is one of the top life-threatening diseases. Accurate typing of AL can significantly improve its prognosis. However, conventional methods for AL typing often require cell staining, which...

Real-time fluorescence imaging flow cytometry enabled by motion deblurring and deep learning algorithms

“…35 Recently, deep learning algorithms have been integrated into IFC to achieve high accuracy in identifying cell subpopulations of interest by leveraging their powerful image data analysis capabilities. [36][37][38] Specifically, these algorithms have been applied in the analysis of low-quality cell images, such as acquiring high-resolution images from low-resolution images, 39 removing blur from out-of-focus images, 40 and identifying healthy cells from degraded images. 41 However, directly utilizing deep learning algorithms to identify cell types from motion-blur images presents challenges as it requires annotated datasets for model training.…”

Is AI essential? Examining the need for deep learning in image-activated sorting of Saccharomyces cerevisiae

“…Furthermore, the analyses presented here were conducted using our previously reported IACS system 28 that operates using fluorescence microscopy. Flow cytometric systems incorporating other methods for cell imaging have been reported including bright-field, 56–59 2D light-scattering 59–61 and Raman, 15,25 as well as 3D imaging techniques such as tomographic phase microscopy 62 and light-sheet fluorescence microscopy. 63–65 Similar investigations using images generated by a wider range of imaging flow cytometric systems would allow us to again draw more universal deductions about the value of utilizing AI for image classification in IACS overall, as well as provide insights for developing AI-enabled microfluidic cell sorting systems based on imaging modalities that have yet to be extended to cell sorting.…”

Optical time‐stretch imaging flow cytometry in the compressed domain

“…In order to make the cells flow at a particular and constant speed in the image plane of the imaging system so that the cells can be stable and individually observed, we fabricated a hydrodynamic‐focusing microfluidic device using standard soft lithography methods [20]. We put a wafer on which the patterns of the channel have been developed with a SU‐8 photoresist into a petri dish which is then filled with polydimethylsiloxane (PDMS) base and curing agent (Sylgard184, Dow Corning, Midland, MI) solution with a mass ratio of 10:1.…”

“…As a result, the existing IFC products trade off image quality and sample flowing speed, which means the high flowing speed will reduce the image quality due to the shorter integration time at the image sensor. The optical time-stretch (OTS) IFC we previously reported [10][11][12][13][14][15][16][17][18][19] has achieved a high throughput of 1 000 000 events per second with a spatial resolution of 780 nm [18,20] in a label-free manner by combining OTS imaging and microfluidics. The OTS IFC has shown immense advantages in large-scale single-cell analysis, while the massive data resulting from high throughput causes tremendous pressure on data acquisition, transmission, and processing in real-time, hindering its practical application in clinical settings.…”