Tuberculosis histopathology on X ray CT

Abstract

Cutting-edge translational research on preclinical models of lung infectious diseases, such as Tuberculosis disease uses computed tomography (CT) images for assessing infection burden and drug efficacy over treatment. Biomarkers which characterize the distribution and extent of the disease-associated tissue are commonly based on the analysis of the intensity histogram as the involved tissues present abnormal densities in the organ being diagnosed. Often the cellular composition of the tissue represented by those grey-levels is ignored. Our hypothesis is that an accurate CT segmentation of the disease-associate tissue components could be based on the histopathological analysis of the sample. Drug development studies would then benefit of the efficacy assessment by lesion compartment response. We present here a protocol that allows to segment the healthy parenchyma, foamy macrophages and neutrophil foci in excised lung samples of healthy and tuberculous animal models.