|
I will review our latest results in the field of quantitative imaging flow cytometry using off-axis holography. Flow cytometry has a great diagnosis potential, due to its ability to analyze a large number of cells during flow for samples obtained from body fluids. Since cellular morphology analysis plays an important role in various clinical diagnoses, such as screening cancer and various chronic diseases, flow cytometry is much anticipated to incorporate imaging capabilities, providing a more comprehensive analysis by presenting a detailed morphological structure image of individual cells. In addition, some erroneous analysis results, yielded in conventional flow cytometry, can be eliminated by acquiring and analyzing such cell images by clearly distinguishing between cells, debris, and clusters of cells. While conventional flow cytometry measures the integral intensity of fluorescent emission, fluorescence microscopy is able to yield the exact morphology of the cell and its organelles. Recent advances in imaging technologies, as well as the exponentially evolving computational capacity, have enabled imaging flow cytometry (IFC) by integrating fluorescence microscopy and conventional flow cytometry. However, the current-generation IFC remains highly inaccessible technology, due to its cost, requirement for operator expertise, lack of accuracy, and lack of objectiveness of data produced. We developed new approach for IFC, which is based on stain-free interferometric phase microscopy, a digital holographic microscopy technique. Using an external interferometric module and cutting-edge deep-learning analysis methods, we generate virtually stained cell images of a volume of cells, with a clear morphological discrimination between various cellular organelles. The module acquires, in a single camera frame, the digital hologram of the cell during flow, and our rapid reconstruction algorithms retrieve the complex wavefront of the cell, from which the optical path delay (OPD) topological map is calculated. This map represents, on each spatial point, the product of the cell physical thickness and its refractive index, accounting for both the cell morphology and contents. Such map is subsequently used as the input to a deep convolutional neural network that is pre-trained to transform the cell OPD into a 2-D image with stained-like organelles. The same IFC setup is also used for automatic cell classification. We demonstrate using the system for cancer detection in liquid biopsies.
|
