Cell
detection is of great significance for biomedical research.
Surface enhanced Raman scattering (SERS) has been widely applied to
the detection of cells. However, there is still a lack of a general,
low-cost, rapid, and sensitive SERS method for cell detection. Herein,
a dynamic liquid SERS platform, which combines label-free SERS technique
with soft tubular microfluidics for cell detection, is proposed. Compared
with common static solid and static liquid measurement, the dynamic
liquid SERS platform can present dynamical mixing, precise control
of the mixing time, and continuous spectra collection. By characterizing
the model molecules, the proposed dynamic liquid SERS platform has
successfully demonstrated good stability and repeatability with 1.90%
and 4.98% relative standard deviation (RSD), respectively. Three cell
lines including one normal breast cell line (MCF-10A) and two breast
cancer cell lines (MCF-7 and MDA-MB-231) were investigated in this
platform. 270 cell spectra were selected as the training set for the
classification of the models based on the K-Nearest Neighbor (K-NN)
algorithm. In three independent experiments, three types of cells
were identified by a test set containing 180 cell spectra with sensitivities
above 83.3% and specificities above 91.6%. The accuracy was 94.1 ±
1.14% among three independent cell identifications. The dynamic liquid
SERS platform has shown higher signal intensity, better repeatability,
less pretreatment, and obtainment of more spectra with less time consumption.
It will be a powerful detection tool in the area of cell research,
clinical diagnosis, and food safety.