Development of a real time trimodal spectroscopy diagnostic tool for Barrett's esophagus

Abstract

Barrett's esophagus (BE) is a condition of the lower esophagus caused by gastroesophageal reflux disease. Patients with BE have an increased probability of developing dysplasia, an abnormal growth or development of cells. This dysplasia in BE is a precursor to cancer of the esophagus, but is currently difficult to detect and diagnose. If the dysplasia is allowed to progress to cancer, it is very difficult to treat successfully. Treatment for dysplasia itself, however, is very effective if done at an early stage. The goal of this thesis project will be to develop a real-time tool that uses spectroscopy to improve upon the methods of detecting dysplasia in BE. This will involve analyzing spectra acquired from patients with BE using models and extracting quantitative information on different aspects of tissue morphology and biochemistry. Using this information, diagnostic algorithms will be developed, optimized and displayed to the physician through a useful interface.