ABSTRACT

Cancer is the second leading cause of death, after heart diseases [25]. It is well known that early detection of precancerous lesions can dramatically decrease morbidity and mortality [37]; thus, there is an urgent need for new diagnostic tools in order to aid in early cancer detection. More than 85% of all cancers begin as precancerous lesions that are conned to the supercial region of the epithelium [38]. However, of the currently used clinical imaging modalities, none have sufcient resolution and sensitivity to detect tumors less than a few cubic centimeters in volume [23]. The current gold standard for assessing suspicious cancerous lesions is histopathology of excised tissue biopsies. Generally the tissue biopsy is sectioned and stained before histopathological assessment. This is an invasive, labor intensive, and costly procedure. Furthermore, the accuracy of the pathological diagnosis depends on the subjective assessment of pathologists [58], which makes current tumor diagnostics ambiguous. Therefore, there is an urgent need for alternative techniques for early detection of cancer that can provide noninvasive, threedimensional (3D), depth-resolved imaging with microscopic resolution, high sensitivity and specicity.