Lung cancer detection using Local Energy-based Shape Histogram (LESH) feature extraction and cognitive machine learning techniques

Citation

Wajid S, Hussain A, Huang K & Boulila W (2017) Lung cancer detection using Local Energy-based Shape Histogram (LESH) feature extraction and cognitive machine learning techniques. In: 2016 IEEE 15th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC). 2016 IEEE 15th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC), Palo Alto, CA, USA, 22.08.2016-23.08.2016. Piscataway, NJ, USA: IEEE, pp. 359-366. https://doi.org/10.1109/ICCI-CC.2016.7862060

Abstract
The novel application of Local Energy-based Shape Histogram (LESH) feature extraction technique was recently proposed for breast cancer diagnosis using mammogram images [22]. This paper extends our original work to apply the LESH technique to detect lung cancer. The JSRT Digital Image Database of chest radiographs is selected for research experimentation. Prior to LESH feature extraction, we enhanced the radiograph images using a contrast limited adaptive histogram equalization (CLAHE) approach. Selected state-of-the-art cognitive machine learning classifiers, namely extreme learning machine (ELM), support vector machine (SVM) and echo state network (ESN) are then applied using the LESH extracted features for efficient diagnosis of correct medical state (existence of benign or malignant cancer) in the x-ray images. Comparative simulation results, evaluated using the classification accuracy performance measure, are further bench-marked against state-of-the-art wavelet based features, and authenticate the distinct capability of our proposed framework for enhancing the diagnosis outcome.

Keywords
Echo State Network (ESN); Clinical Decision Support Systems (CDSSs); Local Energy based Shape Histogram (LESH); Extreme Learning Machine (ELM); Support Vector Machine (SVM)