RESUMO
About 75% of epithelial ovarian cancer (EOC) patients suffer from relapsing and develop drug resistance after primary chemotherapy. The commonly used clinical examinations and biological tumor tissue models for chemotherapeutic sensitivity are time-consuming and expensive. Research studies showed that the cell morphology-based method is promising to be a new route for chemotherapeutic sensitivity evaluation. Here, we offer how the drug resistance of EOC cells can be assessed through a label-free and high-throughput microfluidic flow cytometer equipped with a digital holographic microscope reinforced by machine learning. It is the first time that such type of assessment is performed to the best of our knowledge. Several morphologic and texture features at a single-cell level have been extracted from the quantitative phase images. In addition, we compared four common machine learning algorithms, including naive Bayes, decision tree, K-nearest neighbors, support vector machine (SVM), and fully connected network. The result shows that the SVM classifier achieves the optimal performance with an accuracy of 92.2% and an area under the curve of 0.96. This study demonstrates that the proposed method achieves high-accuracy, high-throughput, and label-free assessment of the drug resistance of EOC cells. Furthermore, it reflects strong potentialities to develop data-driven individualized chemotherapy treatments in the future.
RESUMO
We report a fast autofocusing and accurate 3D tracking scheme for a digital hologram (DH) that intrinsically exploits a polarization microscope setup with two off-axis illumination beams having different polarization. This configuration forms twin-object images that are recorded in a digital hologram by angular and polarization multiplexing technique. We show that the separation of the two images on the recording plane follows a linear relationship with the defocus distance and indicates the defocus direction. Thus, in the entire field of view (FOV), the best focus distance of each object can be directly retrieved by identifying the respective separation distance with a cross-correlation algorithm, at the same time, 3D tracking can be performed by calculating the transverse coordinates of the two images. Moreover, we estimate this linear relationship by utilizing the numerical propagation calculation based on a single hologram, in which the focus distance of one of the objects in the FOV is known. We proved the proposed approach in accurate 3D tracking through multiple completely different experimental cases, i.e., recovering the swimming path of a marine alga (tetraselmis) in water and fast refocusing of ovarian cancer cells under micro-vibration stimulation. The reported experimental results validate the proposed strategy's effectiveness in dynamic measurement and 3D tracking without multiple diffraction calculations and any precise knowledge about the setup. We claim that it is the first time that a holographic polarization multiplexing setup is exploited intrinsically for 3D tracking and/or fast and accurate refocusing. This means that almost any polarization DH setup, thanks to our results, can guarantee accurate focusing along the optical axis in addition to polarization analysis of the sample, thus overcoming the limitation of the poor axial resolution.