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

Xi J, Zhao W, Yuan JE, Kim J, Si X, Xu X - PLoS ONE (2015)

Bottom Line: By 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.For 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.

View Article: PubMed Central - PubMed

Affiliation: School of Engineering and Technology, Central Michigan University, Mount Pleasant, Michigan, United States of America.

ABSTRACT

Background: Each 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 methods: In 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.

Findings: By 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.

Conclusion: For 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.

No MeSH data available.


Related in: MedlinePlus

Comparison of exhaled aerosol fingerprints (AFPs) among asthmatic models with varying constriction levels.(a) Particle distribution, (b) particle concentration distribution, and (c) fractal dimension distribution in a 6×6 matrix. The region of interest in (b) is outlined by the dashed contour and represents the most pronounced variations in exhaled aerosol patterns among the four models.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4589383&req=5

pone.0139511.g003: Comparison of exhaled aerosol fingerprints (AFPs) among asthmatic models with varying constriction levels.(a) Particle distribution, (b) particle concentration distribution, and (c) fractal dimension distribution in a 6×6 matrix. The region of interest in (b) is outlined by the dashed contour and represents the most pronounced variations in exhaled aerosol patterns among the four models.

Mentions: Exhaled particles collect into a unique pattern on the filter and can be seen as the “fingerprint” of the lung. The exhaled particle patterns of the four asthmatic models are illustrated in Fig 3A for 1 μm particles at 30 L/min. Both similarities and disparities in the particle distribution patterns were observed among models, the latter of which was presumably caused by the increasing level of airway constrictions. There were increasing particle attenuations in the upper image, which eventually grew into a crescent-shaped region depleted of particles. This observation was reasonable as there was a gradual airway loss in the four asthmatic models. The amount of particles exhaled from the constricted bronchioles also decreased gradually from D0 to D3. In the extremely constricted scenario (D3, 80% constriction), very few particles could be exhaled. Therefore, the particle-attenuation region is directly related to the disease site and can be selected as the region of interest (ROI) for later analysis. In light of the similarities, two vortices were apparent in the left lower and right lower areas. These two vortices were asymmetric along the central line of the circle, which might result from the right-left lung asymmetry. Considering the possibilities of particle overlapping that prevents an accurate visual interpretation of particle distributions, particle concentrations are also calculated, as shown in Fig 3B. Here blue represents zero concentration and red represents the maximum concentration. Two particle hot spots are apparent in Fig 3B, with one above the left vortex and the other at the right upper corner bordering the crescent-shaped particle-depletion region.


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

Xi J, Zhao W, Yuan JE, Kim J, Si X, Xu X - PLoS ONE (2015)

Comparison of exhaled aerosol fingerprints (AFPs) among asthmatic models with varying constriction levels.(a) Particle distribution, (b) particle concentration distribution, and (c) fractal dimension distribution in a 6×6 matrix. The region of interest in (b) is outlined by the dashed contour and represents the most pronounced variations in exhaled aerosol patterns among the four models.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4589383&req=5

pone.0139511.g003: Comparison of exhaled aerosol fingerprints (AFPs) among asthmatic models with varying constriction levels.(a) Particle distribution, (b) particle concentration distribution, and (c) fractal dimension distribution in a 6×6 matrix. The region of interest in (b) is outlined by the dashed contour and represents the most pronounced variations in exhaled aerosol patterns among the four models.
Mentions: Exhaled particles collect into a unique pattern on the filter and can be seen as the “fingerprint” of the lung. The exhaled particle patterns of the four asthmatic models are illustrated in Fig 3A for 1 μm particles at 30 L/min. Both similarities and disparities in the particle distribution patterns were observed among models, the latter of which was presumably caused by the increasing level of airway constrictions. There were increasing particle attenuations in the upper image, which eventually grew into a crescent-shaped region depleted of particles. This observation was reasonable as there was a gradual airway loss in the four asthmatic models. The amount of particles exhaled from the constricted bronchioles also decreased gradually from D0 to D3. In the extremely constricted scenario (D3, 80% constriction), very few particles could be exhaled. Therefore, the particle-attenuation region is directly related to the disease site and can be selected as the region of interest (ROI) for later analysis. In light of the similarities, two vortices were apparent in the left lower and right lower areas. These two vortices were asymmetric along the central line of the circle, which might result from the right-left lung asymmetry. Considering the possibilities of particle overlapping that prevents an accurate visual interpretation of particle distributions, particle concentrations are also calculated, as shown in Fig 3B. Here blue represents zero concentration and red represents the maximum concentration. Two particle hot spots are apparent in Fig 3B, with one above the left vortex and the other at the right upper corner bordering the crescent-shaped particle-depletion region.

Bottom Line: By 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.For 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.

View Article: PubMed Central - PubMed

Affiliation: School of Engineering and Technology, Central Michigan University, Mount Pleasant, Michigan, United States of America.

ABSTRACT

Background: Each 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 methods: In 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.

Findings: By 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.

Conclusion: For 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.

No MeSH data available.


Related in: MedlinePlus