We are investigating the use of texture features derived from several different approaches involving gray-level statistics, topological or geometrical features in characterizing pathological patterns indicative of interstitial lung disease, as visualized on high-resolution CT. Our goal is to use such quantitative measures for automated classification of healthy and diseased lung tissue. While current work aims at a binary classification (healthy/diseased), we are also interested in expanding the use of such texture features for also distinguishing between different pathological patterns (as seen in the figure, leftmost sub-figure depicts healthy lung).
Our long term goals with respect to this project are to contribute to the development of a computer-aided diagnosis (CADx) tool to assist radiologists in characterization and quantification of pathological lung tissue on CT images. We believe that such work can also play a role in content-based image retrieval systems for purposes of identifying other patient cases with similar pathology.
Journal Publications
- M.B. Huber, K. Bunte, M.B. Nagarajan, M. Biehl, L.A. Ray, and A.Wismüller, "Texture feature ranking with relevance learning to classify interstitial lung disease patterns," Artificial Intelligence in Medicine 56(2):91-97, (2012). PMID: 23010586
- M.B. Huber, M.B. Nagarajan, G. Leinsinger, R. Eibel, L. Ray, and A. Wismüller, “Performance of topological texture features to classify fibrotic interstitial lung disease patterns,” Medical Physics, 38(4): 2035–2044 (2011). PMID: 21626936
Conference Publications
- M.B. Huber, M.B. Nagarajan, G. Leinsinger, L.A. Ray, and A. Wismüller, "Classification of interstitial lung disease patterns with topological texture features," Proceedings of SPIE Medical Imaging 7624:101-108, (2010).
- M.B. Huber, K. Bunte, M.B. Nagarajan, M. Biehl, L.A. Ray, and A. Wismüller, "Texture feature selection with relevance learning to classify interstitial lung disease patterns," Proceedings of SPIE Medical Imaging 7963:181-188, (2011).
- M.B. Huber, M.B. Nagarajan, G. Leinsinger, L.A. Ray, and A. Wismüller, "Estimating local scaling properties for the classification of interstitial lung disease patterns," Proceedings of SPIE Medical Imaging 7963:1Q1-1Q8, (2011).