Detecting Lung Diseases from Exhaled Aerosols: Non-Invasive Lung Diagnosis Using Fractal Analysis and SVM Classification.

Jinxiang Xi,Jiayao Eddie Yuan,Xiaowei Xu,Weizhong Zhao,Jongwon Kim,Xiuhua Si

doi:10.1371/journal.pone.0139511

Abstract

BackgroundEach lung structure exhales a unique pattern of aerosols, which can be used to detect and monitor lung diseases non-invasively. The challenges are accurately interpreting the exhaled aerosol fingerprints and quantitatively correlating them to the lung diseases.Objective and MethodsIn this study, we presented a paradigm of an exhaled aerosol test that addresses the above two challenges and is promising to detect the site and severity of lung diseases. This paradigm consists of two steps: image feature extraction using sub-regional fractal analysis and data classification using a support vector machine (SVM). Numerical experiments were conducted to evaluate the feasibility of the breath test in four asthmatic lung models. A high-fidelity image-CFD approach was employed to compute the exhaled aerosol patterns under different disease conditions.FindingsBy employing the 10-fold cross-validation method, we achieved 100% classification accuracy among four asthmatic models using an ideal 108-sample dataset and 99.1% accuracy using a more realistic 324-sample dataset. The fractal-SVM classifier has been shown to be robust, highly sensitive to structural variations, and inherently suitable for investigating aerosol-disease correlations.ConclusionFor the first time, this study quantitatively linked the exhaled aerosol patterns with their underlying diseases and set the stage for the development of a computer-aided diagnostic system for non-invasive detection of obstructive respiratory diseases.

Highlights

The ability to diagnose lung cancer at an early stage is crucial to patients’ survival
Even though diagnostic tools such as chest radiography, computed tomography, and biopsy are accurate in diagnosis, they have not been recommended for screening purposes
Analyzing exhaled breath from individuals who are at a high risk of lung cancer could be an inexpensive and non-invasive method of diagnosing the disease

Summary

Introduction

The ability to diagnose lung cancer at an early stage is crucial to patients’ survival. Examples include increased concentrations of antioxidants for chronic obstructive pulmonary disease (COPD)[3], nitric oxide for asthma[4], and isoprene for non-small cell lung cancer (NSCLC)[5]. These gas-fingerprint based devices, such as electronic noses[6], only measure the concentration of exhaled gaseous chemicals. Non-volatile molecules exhaled from the fluid that lines the lung are collected as condensates This method has been shown to be useful in studying inflammatory and oxidative processes on the surfaces of the respiratory tract [7]. The challenges are accurately interpreting the exhaled aerosol fingerprints and quantitatively correlating them to the lung diseases.

Methods

Results

Conclusion