An integrated microflow cytometry platform with artificial intelligence capabilities for point-of-care cellular phenotype analysis.
Journal
Biosensors & bioelectronics
Journal Volume
271
Start Page
Article number 117074
ISSN
1873-4235
Date Issued
2025-03-01
Author(s)
Kuo, Ju-Nan
Jian, Ming-Shen
Chiang, Chia-Huang
Kuo, Wen-Kai
Lin, I-En
Kuo, Yung-Ming
Ye, Yi-Ling
Abstract
The EZ DEVICE is an integrated fluorescence microflow cytometer designed for automated cell phenotyping and enumeration using artificial intelligence (AI). The platform consists of a laser diode, optical filter, objective lens, CMOS image sensor, and microfluidic chip, enabling automated sample pretreatment, labeling, and detection within a single compact unit. AI algorithms segment and identify objects in images captured by the CMOS sensor at 532 and 586 nm emission wavelengths. The device's counting performance, tested with rainbow and FITC-labeled beads, closely matched results from manual counting using an Olympus IX73 microscope. Antibody-staining efficiency was evaluated with antibody-particle beads and IgG k isotype-FITC, achieving a 99.06% staining efficiency. Practical feasibility was demonstrated through the phenotyping of immune cell lines (Jurkat T and THP-1) using specific fluorescent antibodies. The system successfully detected MHC I expression and distinguished CD3 expression patterns in Jurkat T-cells, demonstrating its capability to recognize distinct cell statuses. With its compact design, automated features, and efficient staining and detection, the EZ DEVICE shows promise for diverse applications, including clinical diagnostics, point-of-care testing, and research, enabling real-time immune monitoring and precise cell analysis. Future enhancements, such as optimized microfluidics, advanced imaging, and AI-driven algorithms, aim to improve single-cell throughput and expand its utility in personalized medicine and public health.
Subjects
Artificial intelligence
EZ DEVICE
Image processing
Microflow cytometer
Microfluidic
Point-of-care testing
SDGs
Type
journal article