Classification of Endomicroscopic Images of the Lung Based on Random Subwindows and Extra-Trees

• Published in: IEEE transactions on biomedical engineering Vol. 59; no. 9; pp. 2677 - 2683
• Main Authors: Desir, Chesner, Petitjean, Caroline, Heutte, Laurent, Salaun, Mathieu, Thiberville, Luc
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.09.2012; Institute of Electrical and Electronics Engineers
• Subjects: Algorithms; Applied sciences; Biological and medical sciences; Computer Science; Computerized, statistical medical data processing and models in biomedicine; Confocal fluorescence microscopy; Databases, Factual; Decision Trees; Diagnosis, Computer-Assisted - methods; Engineering Sciences; Exact sciences and technology; extra-trees; Feature extraction; Humans; image classification; Image processing; Information, signal and communications theory; Lungs; medical imaging; Medical management aid. Diagnosis aid; Medical sciences; Microscopy, Confocal - methods; Pathology; Pulmonary Alveoli - anatomy & histology; Pulmonary Alveoli - cytology; Pulmonary Alveoli - pathology; Pulmonary Alveoli - ultrastructure; reject rule; Reproducibility of Results; Signal and communications theory; Signal and Image Processing; Signal processing; Signal representation. Spectral analysis; Signal, noise; Smoking - pathology; Telecommunications and information theory; Thoracoscopy - methods; Training; Vectors; Vegetation; Confocal fluorescence microscopy; Image processing; image classification; Lung; extra-trees; Medical imagery; Confocal microscopy; reject rule; Respiratory system; Fluorescence microscopy; medical imaging; Biomedical engineering
• ISSN: 0018-9294; 1558-2531; 1558-2531
• DOI: 10.1109/TBME.2012.2204747

Recently, the in vivo imaging of pulmonary alveoli was made possible thanks to confocal microscopy. For these images, we wish to aid the clinician by developing a computer-aided diagnosis system, able to discriminate between healthy and pathological subjects. The lack of expertise currently available on these images has first led us to choose a generic approach, based on pixel-value description of randomly extracted subwindows and decision tree ensemble for classification (extra-trees). In order to deal with the great complexity of our images, we adapt this method by introducing a texture-based description of the subwindows, based on local binary patterns. We show through our experimental protocol that this adaptation is a promising way to classify fibered confocal fluorescence microscopy images. In addition, we introduce a rejection mechanism on the classifier output to prevent nondetection errors.