Sign In

Communications of the ACM

ACM TechNews

How the Power of Mathematics Can Help Assess Lung Function

View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
An illustration of lung structure based on computed tomography scan data.

Researchers at the University of Southampton in the U.K. have developed a novel mathematical method for analyzing x-rays and scans of human lungs.

Credit: University of Southampton (U.K.)

A novel mathematical method to analyze lung x-rays, developed by researchers at the University of Southampton in the U.K., could transform diagnosis and assessment of Chronic Obstructive Pulmonary Disease (COPD) and other respiratory maladies.

The team combined computed tomography (CT) scans, high-performance computing, and algorithms to compute the three-dimensional properties of the bronchial trees of 64 patients classified as either healthy non-smokers, healthy smokers, patients with moderate COPD, or those with mild COPD.

Topological data analysis of the bronchial trees' structure and size, their branches' length and direction, and comparative reconfigurations during deep inhalation and full exhalation revealed that a more-complex tree suggests improved lung function, while a smaller, distorted tree indicates poorer lung function.

Not only does the technique differentiate between specific patient cohorts, their lung function characteristics, and different disease stages, but it also can detect properties hidden from the naked eye.

From University of Southampton (U.K.)
View Full Article


Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


No entries found