Designing a CT image processing technique to diagnose emphysema

Stanford University, BE110-Physiological Signals and Systems, Fall 2016

Chronic Obstructive Pulmonary Disease (COPD) is the third leading cause of death in the US. Emphysema, a subset of COPD, consists of gradual lung tissue damage and shortness of breath. The primary diagnostic tool used is spirometry as a part of pulmonary function tests, however this method is incapable of detecting early emphysema due to its low sensitivity. This project consisted of diagnosing emphysema via Computed Tomogrophy (CT) images. Using 39 patient samples, image processing was used to create a mask and threshold to identify the lungs, then take a histogram of the intensity. Based on the distribution of the intensity, we were able to determine the presence and severity of Emphysema (p<0.05). Computer aided detection of emphysema will bring higher speed and precision into medical care.