Imaging the Neovasculature

Project title: Spectral Dark Field Microscopy for Micro Physiological Imaging of Cancer

Project Description: Angiogenesis is one of important hallmarks of cancer. Tumor-induced neovasculature is often characterized as leaky, tortuous and chaotic, unlike highly organized normal vasculature. Tumors cannot grow beyond 1-2 mm in diameter without angiogenesis. Capturing the early angiogenesis may aid early detection of pre-cancer. In this project, we are combining dark field microscopy and a vascular segmentation algorithm to quantify early neovascularization changes associated with the carcinogenesis process in vivo using a non-invasive, label-free, high resolution, reflected-light spectral dark field microscope.


Dark field microscopy reveals higher vascular tortuosity in precancerous and cancerous in the spontaneous hamster carcinogen model. Hamsters’ cheek pouches were treated with DMBA (7,12-Dimethylbenz(a)anthracene) to induce cancer. Non-invasive, label-free spectral dark field images were obtained from the treated and control cheek pouches at different time points during the cancer progression. Biopsies were obtained immediately after images were taken for pathology. Vascular features was segmented using Gabor filter. Tortuosity index was computed from the vessel masks. Tortuosity index increased during the cancer progression. Dark field microscopy may be an appropriate tools for pre-cancer and cancer diagnosis in resource limited settings. (SCC: squamous cell carcinoma, Tor: tortuosity)