Small-Cell Lung Cancer Detection Using a Supervised Machine Learning Algorithm

Cancer-related medical expenses and labor loss cost annually $10,000 billion worldwide. Lung cancer-related deaths exceed 70,000 cases globally every year. Furthermore, 225,000 new cases were detected in the United States in 2016, and 4.3 million new cases in China in 2015. Statistically, most lung cancer related deaths were due to late stage detection. Like other types of cancer, early detection of lung cancer could be the best strategy to save lives. In this paper, we propose a novel neural-network based algorithm, which we refer to as entropy degradation method (EDM), to detect small cell lung cancer (SCLC) from computed tomography (CT) images. This research could facilitate early detection of lung cancers. The training data and testing data are high-resolution lung CT scans provided by the National Cancer Institute. We selected 12 lung CT scans from the library, 6 of which are for healthy lungs, and the remaining 6 are scans from patients with SCLC. We randomly take 5 scans from each group to train our model, and used the remaining two scans to test. Our algorithms achieves an accuracy of 77.8%.

[1]  Xiong Luo,et al.  A laguerre neural network-based ADP learning scheme with its application to tracking control in the Internet of Things , 2016, Personal and Ubiquitous Computing.

[2]  A. Jemal,et al.  Global cancer statistics, 2012 , 2015, CA: a cancer journal for clinicians.

[3]  D. Shin,et al.  Medical Costs and Healthcare Utilization among Cancer Decedents in the Last Year of Life in 2009 , 2015, Cancer research and treatment : official journal of Korean Cancer Association.

[4]  Jean-Baptiste Thibault,et al.  A three-dimensional statistical approach to improved image quality for multislice helical CT. , 2007, Medical physics.

[5]  Otto Zhou,et al.  Statistical iterative reconstruction to improve image quality for digital breast tomosynthesis. , 2015, Medical physics.

[6]  Xiong Luo,et al.  Towards enhancing stacked extreme learning machine with sparse autoencoder by correntropy , 2017, J. Frankl. Inst..

[7]  J. Ferlay,et al.  International trends in lung cancer incidence by histological subtype: adenocarcinoma stabilizing in men but still increasing in women. , 2014, Lung cancer.

[8]  Andrea Vedaldi,et al.  MatConvNet: Convolutional Neural Networks for MATLAB , 2014, ACM Multimedia.

[9]  Michael C. Lee,et al.  Computer-aided diagnosis of pulmonary nodules using a two-step approach for feature selection and classifier ensemble construction , 2010, Artif. Intell. Medicine.

[10]  Aly A. Farag,et al.  A Novel Approach for Lung Nodules Segmentation in Chest CT Using Level Sets , 2013, IEEE Transactions on Image Processing.

[11]  Berkman Sahiner,et al.  Lung nodule detection on thoracic computed tomography images: preliminary evaluation of a computer-aided diagnosis system. , 2002, Medical physics.

[12]  Xiong Luo,et al.  A Quantized Kernel Learning Algorithm Using a Minimum Kernel Risk-Sensitive Loss Criterion and Bilateral Gradient Technique , 2017, Entropy.

[13]  A. Jemal,et al.  Cancer statistics in China, 2015 , 2016, CA: a cancer journal for clinicians.

[14]  Jens Wiegert,et al.  Deep residual learning in CT physics: scatter correction for spectral CT , 2017, 2017 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC).

[15]  Xiong Luo,et al.  A quantized kernel least mean square scheme with entropy-guided learning for intelligent data analysis , 2016, China Communications.

[16]  A. Jemal,et al.  Cancer statistics, 2016 , 2016, CA: a cancer journal for clinicians.

[17]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.