AUTOMATICALLY DETECTING CELLS HAVING NUCLEAR AND CYTOPLASMIC DENSITOMETRIC FEATURES ASSOCIATED WITH DISEASE BY OPTICAL IMAGING

摘要

A system and method for rapidly detecting cells (1) of interest using multi-dimensional, highly quantitative, nuclear and cytoplasmic densitometric features (NDFs and CDFs) includes a flow optical tomography (FOT) instrument capable of generating various optical projection images (or shadowgrams) containing accurate density information from a cell, a computer (40) and software to analyze and reconstruct the projection images into a multi-dimensional data set, and automated feature collection and object classifiers. The system and method are particularly useful in the early detection of cancers such as lung cancer using a bronchial specimen from sputum or cheek scrapings and cervical/ovarian cancer using a cervical scraping, and the system can be used to detect rare cells in specimens including blood.