.png){kind=icon}
.png){kind=icon}
Return
Upgraded User-Friendly Image-Activated Microfluidic Cell Sorter Using an Optimized and Fast Deep Learning Algorithm
.png){kind=icon}
OA
AI
DOI：10.3390/mi13122105.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Image-based cell sorting is essential in biological and biomedical research. The sorted cells can be used for downstream analysis to expand our knowledge of cell-to-cell differences. We previously demonstrated a user-friendly image-activated microfluidic cell sorting technique using an optimized and fast deep learning algorithm. Real-time isolation of cells was carried out using this technique with an inverted microscope. In this study, we devised a recently upgraded sorting system. The cell sorting techniques shown on the microscope were implemented as a real system. Several new features were added to make it easier for the users to conduct the real-time sorting of cells or particles. The newly added features are as follows: (1) a high-resolution linear piezo-stage is used to obtain in-focus images of the fast-flowing cells; (2) an LED strobe light was incorporated to minimize the motion blur of fast-flowing cells; and (3) a vertical syringe pump setup was used to prevent the cell sedimentation. The sorting performance of the upgraded system was demonstrated through the real-time sorting of fluorescent polystyrene beads. The sorter achieved a 99.4% sorting purity for 15 mu m and 10 mu m beads with an average throughput of 22.1 events per second (eps).
Keywords:
Microfluidic Flow Cytometry
image-based cell sorting
deep learning
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.