Join us in welcoming Mark Gesley to give a talk entitled:
Spectral image microscopy for label-free blood and cancer cell recognition
A new approach to cancer cell recognition may improve detection of minimal residual disease from peripheral blood samples. Spectral images of human blood and cancer cells are produced by optical microscopy with filtered broadband visible light elastically scattered from throughout a cell. Cell spectra are assembled from a field of view series of monochromatic images. Exogenous tags, like clusters of differentiation (CD) markers, may introduce a bias and are not required. A prototype transmission optics at ~20x magnification yields images with 0.4 [μm] pixels at a Rayleigh resolution 0.44 ~ 0.54 [μm] over a wavelength range of λ = 440 – 620 [nm] using a xenon arc lamp, monochromator, CCD camera, and objective lens NA = 0.5. A training sample cell may be validated without detailed knowledge of intracellular spectra. Spectral object classification is scalable to any number of cell types. Small samples of erythrocytes, leukocytes, Jurkat cancer, and non-small cell adenocarcinoma cells are accurately classified and associated with unique cell spatial-spectral characteristics. A system architecture is described that makes fast, label-free recognition of aberrant blood cells feasible.
Read Dr. Gesley’s bio here.