iXray: A Machine Learning-Based Digital Radiograph Pattern Recognition System for Lung Pathology Detection
A radiograph is a visualization aid that physicians use in identifying lung abnormalities. Although digitized x-ray images are available, diagnosis by a medical expert through pattern recognition is done manually. Thus, this paper presents a system that utilizes machine learning for pattern recognition and classification of six lung conditions. Classified into two categories, namely histogram-based (normal, pleural effusion, and pneumothorax) and statistics-based (cardiomegaly, hyperaeration, and possible lung nodules). Using preprocessing and feature extraction techniques, the designed system achieves an accuracy rate of 92.59% for the histogram-based lung conditions using Sequential Minimal Optimization (SMO) and 67.22% for the statistics-based lung conditions using logic operations.
KeywordsPattern recognition Sequential Minimal Optimization Image histogram
We would like to give gratitude to Dr. Lourd Loreto, Dr. Adrian Rabe of the Philippine General Hospital and Dr. Jun Parungao of De La Salle Health Sciences Institute, for imparting with us the basic medical knowledge needed.
- 1.Chapelle, O., P. Haffner, and V. Vapnik. 1999. SVMs for histogram-based image classification, IEEE Computational Intelligence Society. Red Bank, New Jersey: Speech and Image Processing Services Research Laboratory, AT&T Labs-Research.Google Scholar
- 2.Dick, E. 2000. Chest X-rays made easy. Student BMJ—The International Medical Journal for Students. 8:316–317.Google Scholar
- 3.Duda, R., and P.E. Hart. 1973. Pattern classification and scene analysis. California: Medical eBook, Standford Research Institute.Google Scholar
- 4.Gu, L., X. Peng, Y. Sun, L. Qian, S. Wang, Q. Weng, and J. Xu. 2009. Computer-aided diagnosis: A support vector machine-based approach of automatic Pulmonary nodule detection in chest radiographs. In IEEE Conference Paper – Proceedings of 2009 International Symposium on Bioelectronics & Bioinformatics. Melbourne, Australia.Google Scholar
- 5.Howley, T., and M.G. Madden. 2005. The genetic kernel support vector machine: Description and evaluation. 24 (3–4):379–395. https://doi.org/10.1007/s10462-005-9009-3.
- 6.Mousa, W.A.H., and M.A.U Khan. 2002. Lung nodule classification utilizing support vector machines. In IEEE Conference Paper—Proceedings of the International Conference on Image Processing. Dhahran, Saudi Arabia: Department of Electrical Engineering, King Fahd University of Petroleum and Minerals.Google Scholar
- 7.Noriyasu, H. 2009. Pattern recognition in medical image diagnosis. In Pattern recognition, ed. Peng-Yeng Yin, ISBN- 978-953-307-014-8. Available from: http://www.intechopen.com/books/patternrecognition/pattern-recognition-in-medical-image-diagnosis.
- 8.Platt, J.C. 1999. Training of support vector machines using sequential minimal optimization. In Advances in kernel methods, 185–208. Cambridge, MA.Google Scholar
- 9.Pedersen, S. 2007. Circular hough transform. In Aalborg University, vision, graphics, and interactive systems.Google Scholar
- 10.Shimada, T., et al. 2002. Proposal of a nodule density-enhancing filter for plain chest radiographs on the basis of the thoracic wall outline detected by Hough Transform. IEICE Transactions on Information and Systems—Special Issue on Measurements and Visualization Technology of Biological Information, E85-D (1): 88–95.Google Scholar
- 11.Tonpho, T., A. Leelasantihan, and S. Kiattisin. 2010. Investigation of chest X-ray images based on medical knowledge and balanced histograms. Bangkok, Thailand: International Symposium on Intelligent Signal Processing and Communication Systems.Google Scholar
- 12.Üstün, B., W.J. Melssen, and L.M.C. Buydens. 2007. Visualisation and interpretation of support vector regression models, vol. 595, pp 299–309.Google Scholar