MPI-INF Logo
Campus Event Calendar

Event Entry

What and Who

Statistical Learning on Metabolic Ion Mobility Spectrometry Profiles for Disease Identification

Anne-Christin Hauschild
Fachrichtung Informatik - Saarbrücken
PhD Application Talk
AG 1, AG 2, AG 3, AG 4, AG 5, SWS, RG1, MMCI  
Public Audience
English

Date, Time and Location

Monday, 7 May 2012
09:20
90 Minutes
E1 4
024
Saarbrücken

Abstract

Exhaled air carries information on the human health status. Ion mobility spectrometers combined with a multi-capillary column (MCC/IMS) is a well known technology for detecting volatile organic compounds (VOCs) within the breath. The technique is comparably cheap, robust and easy to use in every day practice. However, the  potential of the methodology depends on (1) the successful application of computational approaches for finding relevant VOCs and (2) classifying patients into  disease-specific profile groups based on the detected VOCs.  I will present a pilot study analyzing the ability of sophisticated statistical learning techniques for (1) supervised classification and (2) VOC-based feature selection.

We analyzed breath data from 84 volunteers, each of them suffering from chronic obstructive pulmonary disease (COPD), as well as 35 healthy volunteers, comprising  a control group (CG).

Our results suggests a strong potential for separating MCC/IMS chromatograms of healthy controls and COPD patients (best accuracy COPD vs. CG: 94%). Furthermore, we extracted a set of high-scoring VOCs that allow differentiating of COPD patients from healthy controls. Our findings demonstrate a generally high but improvable  accuracy of statistical learning methods when applied to well-structured, medical MCC/IMS data.

Contact

IMPRS Office Team
9325 1800
--email hidden
passcode not visible
logged in users only

Marc Schmitt, 05/04/2012 13:32 -- Created document.