Mueller matrix microscopy for label-free histopathology examinations (Conference Presentation)

Abstract

In clinical medicine, a pathologist often needs to examine cells or thin slices of tissues to identify abnormalities that are markers or precursors of diseases. Various chemical and immunohistochemical staining techniques have been developed to selectively label certain components to bring up the contrasts of specific microstructures. It is well known that a Mueller matrix contains rich information on the microstructure and optical properties of a sample. Using proper data analysis techniques, Mueller matrix images can also be transformed into new polarization parameters sensitive only to specific microstructural features. These new polarization parameters can selectively enhance the contrast of specific features in images of unstained pathological slide to help identify abnormalities. In recent studies, we set up a modulus design Mueller matrix microscope by adding polarization optics components into the optical path of a commercial transmission microscope. We take multiple measurements of the unstained pathological slide at different polar and azimuth angles, then derive an intrinsic Mueller matrix (IMM) which represents only the microstructural characters of the sample without the interference by the sample orientation. Such orientation-independent IMM images preserve to the maximum extent the pathological information of the tissue samples. Using Mueller matrix decomposition and transformation techniques, we demonstrate in preliminary tests that we are able to selectively enhance different characteristic features in different cancer tissues. With the fast advances in big-data analysis techniques, it is expected that label-free Mueller matrix microscopy is a potentially powerful tool for the histopathologists to identify characteristic features in complex tissue samples.