研究目的
To inspect and analyze human colorectal cancer cell lines using terahertz time-domain attenuated total reflection spectroscopy, study their THz absorption spectra characteristics at different concentrations, and develop an identification method for cancer cell lines.
研究成果
THz TD-ATR spectroscopy can characterize colorectal cancer cell lines with nonlinear absorption changes related to hydration states. Absorption coefficient and dielectric loss tangent are effective parameters for identifying cancer cells, achieving up to 88.9% accuracy. This provides a preliminary basis for liquid biopsy applications, with future work needed on mechanism exploration and model improvement.
研究不足
The study is limited to specific cell lines (DLD-1 and HT-29) and normal mouse cells; results may not generalize to other cancer types or human normal cells. The THz system has a valid frequency range reduced in aqueous samples, and sample preparation may introduce variability. The classification accuracy is not perfect, and more samples are needed for improved model feasibility.
1：Experimental Design and Method Selection:
Used terahertz time-domain attenuated total reflection spectroscopy (THz TD-ATR) to study absorption spectra of cell lines. Applied machine learning (KNN) for classification.
2：Sample Selection and Data Sources:
Human colon cancer cell lines HT-29 and DLD-1, and mouse fibroblast epithelial cells NIH/3T3 as normal cell line, purchased from Cell Bank of Shanghai Institutes for Biological Sciences of Chinese Academy of Sciences. Cultured in RPMI-1640 medium.
3：List of Experimental Equipment and Materials:
THz TD-ATR system (TAS7500 by Advantest Corporation), silicon ATR prism, optical fiber femtosecond laser, incubator, microscope, blood cell count plate, pipette tips, distilled water, RPMI-1640 medium.
4：Experimental Procedures and Operational Workflow:
Cells cultured to logarithmic phase, counted, concentrations adjusted to 5×10^6 cells/ml and diluted to lower concentrations. Samples dropped (400 μl each), spectra measured at room temperature and low humidity. Reference spectra of medium and distilled water taken.
5：Data Analysis Methods:
Complex refractive index, absorption coefficient, complex permittivity, dielectric loss tangent calculated. Principal component analysis for feature extraction, KNN for classification with five-fold cross-validation.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
