RSNA 2014 

Abstract Archives of the RSNA, 2014


INE041-b

Computerized Detection and Classification of Pulmonary Pathologies from CT Images: Current Approaches, Challenges, and Future Trends

Education Exhibits

Presented on December 4, 2014
Presented as part of INS-THA: Informatics Thursday Poster Discussions

 Certificate of Merit

Participants

Awais Mansoor PhD, Presenter: Nothing to Disclose
Ulas Bagci PhD, MSc, Abstract Co-Author: Nothing to Disclose
Ziyue Xu PhD, Abstract Co-Author: Nothing to Disclose
Brent Foster, Abstract Co-Author: Nothing to Disclose
Georgios Z. Papadakis MD, Abstract Co-Author: Nothing to Disclose
Ken Olivier, Abstract Co-Author: Nothing to Disclose
Jason M. Elinoff, Abstract Co-Author: Nothing to Disclose
Anthony F. Suffredini MD, Abstract Co-Author: Nothing to Disclose
Jayaram K. Udupa PhD, Abstract Co-Author: Nothing to Disclose
Daniel Joseph Mollura MD, Abstract Co-Author: Nothing to Disclose

TEACHING POINTS

1. To identify the clinical importance of lung pathology detection, segmentation and classification. 2. To review the state-of-the-art image segmentation approaches for pathological lungs from CT scans. 3. Computerized techniques for disease detection and quantifications. 4. To discuss the future of lung segmentation methods and explain how engineering advancements in CT plays a valuable role.    

TABLE OF CONTENTS/OUTLINE

1. Introduction a. Lung disease and imaging patterns in pulmonary CT. b. Clinical importance of pathology segmentation and classification. 2. Segmenting Lung Pathologies from CT images a. Fleischner Society’s descriptions for imaging patterns of lung diseases. b. Computer aided analysis of of CT images of lungs. c. Determining lung and pathology volume as imaging markers in diagnostic tasks. 3. The Use of Advanced Imaging Features to Accurately Classify Imaging Patterns. a. Review of optimal textural feature selection for pulmonary pathology classification. b. Review of optimal machine learning classifiers for pulmonary pathology classification. 4. The Challenges of Segmentation and Classification of Lung Pathology a. Image quality. b. Sufficient discriminative information. c. Time needed for analysis. 5. Concluding Remarks and future trends.  

PDF UPLOAD

http://abstract.rsna.org/uploads/2014/14018415/14018415_99ac.pdf

Cite This Abstract

Mansoor, A, Bagci, U, Xu, Z, Foster, B, Papadakis, G, Olivier, K, Elinoff, J, Suffredini, A, Udupa, J, Mollura, D, Computerized Detection and Classification of Pulmonary Pathologies from CT Images: Current Approaches, Challenges, and Future Trends.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14018415.html