Dataset Information

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

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.

SUBMITTER: Xi J

PROVIDER: S-EPMC4589383 | biostudies-literature | 2015

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Xi Jinxiang J Zhao Weizhong W Yuan Jiayao Eddie JE Kim JongWon J Si Xiuhua X Xu Xiaowei X

PloS one 20150930 9

<h4>Background</h4>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.<h4>Objective and methods</h4>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 con ...[more]

PMID: 26422016

Similar Datasets

Project description:Diffuse cystic lung diseases are characterized by cysts in more than one lung lobe, the cysts originating from various mechanisms, including the expansion of the distal airspaces due to airway obstruction, necrosis of the airway walls, and parenchymal destruction. The progression of these diseases is variable. One essential tool in the evaluation of these diseases is HRCT, because it improves the characterization of pulmonary cysts (including their distribution, size, and length) and the evaluation of the regularity of the cyst wall, as well as the identification of associated pulmonary and extrapulmonary lesions. When combined with clinical and laboratory findings, HRCT is often sufficient for the etiological definition of diffuse lung cysts, avoiding the need for lung biopsy. The differential diagnoses of diffuse cystic lung diseases are myriad, including neoplastic, inflammatory, and infectious etiologies. Pulmonary Langerhans cell histiocytosis, lymphangioleiomyomatosis, lymphocytic interstitial pneumonia, and follicular bronchiolitis are the most common diseases that produce this CT pattern. However, new diseases have been included as potential determinants of this pattern. RESUMO As doenças pulmonares císticas difusas se caracterizam pela presença de cistos envolvendo mais de um lobo pulmonar, que se originam por diversos mecanismos, incluindo dilatação dos espaços aéreos distais por obstrução, necrose das paredes das vias aéreas e destruição do parênquima. Essas doenças apresentam evolução variável. A TCAR é fundamental na avaliação dessas doenças uma vez que permite uma melhor caracterização dos cistos pulmonares, incluindo sua distribuição, tamanho, extensão e regularidade das paredes, assim como a determinação de outras lesões pulmonares e extrapulmonares associadas. Frequentemente a TCAR é suficiente para a definição etiológica dos cistos pulmonares difusos, associada a achados clínicos e laboratoriais, sem a necessidade de realização de biópsia pulmonar. O diagnóstico diferencial das doenças pulmonares císticas difusas é extenso, incluindo etiologias neoplásicas, inflamatórias e infecciosas, sendo as mais frequentes determinantes desse padrão tomográfico a histiocitose pulmonar de células de Langerhans, a linfangioleiomiomatose, a pneumonia intersticial linfocitária e a bronquiolite folicular. Novas etiologias foram incluídas como potenciais determinantes desse padrão.

Dataset Information

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

Background

Objective and methods

Findings

Conclusion

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets