CC BY
4DOI 10.24412/CL-37135-2023-1-42-42
RAMAN-LIBS FOR TUMOR TISSUE IMAGING AND CELLS DETECTION
YOUYUAN CHEN, PENGKUN YIN, ZHENGYING PEENG, QINGYU LIN, YIXIANG DUAN
Research Center of Analytical Instrumentation, School of Mechanical Engineering, Sichuan University, Chengdu, People's Republic of China
qylin@scu.edu.cn
ABSTRACT
With cancer seriously hampering the increasing life expectancy of people, developing an instantly diagnostic method has become an urgent objective. we developed a label-free LIBS method for high-throughput recognition of tumor cells based on interpretable deep learning. Saliency maps thus obtained amplified the differences between the spectra of cell lines. The proposed method achieved high accuracy and is seen as an interpretable classification process for cancer cell lines. Moreover, we further focused on the analysis of tumor cells with Raman. We proposed a novel strategy based on signal amplification reaction (i.e. 3D DNA walker and CHA ) and SERS for ultrasensitive detection of tumor cells. The tumor cell can be recognized by the corresponding aptamer sequence and can be detected by this SERS method. Our group also proposed LIBS instrument and method for the diagnostic analysis of clinical lung cancer tissues based on label-free imaging. The heterogeneity of multi-elements and molecular fragments was obtained simultaneously for three types of clinical samples containing different proportions of cancerous tissue by laser-induced breakdown spectroscopy (LIBS) imaging.
REFERENCES
[1] Y. Chen, P. Yin, Z. Peng, Q. Lin*, Y. Duan, Q.. Fan, Z. Wei. High-throughput recognition of tumor cells using label-free elemental characteristics based on interpretable deep learning. Analytical Chemistry, 94, 3158-3164, 2022.
[2] Z. Wei, Q. Lin*, E. N. Lazareva, P. A. Dyachenko, J. Yang, Y. Duan, V. V. Tuchin*. Optical clearing of laser-induced tissue plasma. Laser Physics Letters, 18,085603, 2021.
