DETECTION OF LUNG CANCER USING IMAGE SEGMENTATION

International Journal of Electrical Engineering & Technology (IJEET) Volume 11, Issue 2, March-April 2020, pp. 7-16, Article ID: IJEET_11_02_002 Available online at http://www.iaeme.com/IJEET/issues.asp?JType=IJEET&VType=11&IType=2 ISSN Print: 0976-6545 and ISSN Online: 0976-6553 Journal Impact Factor (2020): 10.1935 (Calculated by GISI) www.jifactor.com © IAEME Publication DETECTION OF LUNG CANCER USING IMAGE SEGMENTATION V. Kannan Professors, Department of ECE, GMR Institute of Technology, Rajam, Andhra Pradesh, India V. Jagan Naveen Professors, Department of ECE, GMR Institute of Technology, Rajam, Andhra Pradesh, India ABSTRACT In modern days, image processing methods are widely adopted in the medical field for enhancing the earlier detection of certain abnormalities, such as the breast cancer, lung cancer, brain cancer and so on. This paper mainly concentrates on the segmentation of lung cancer tumors from X-ray images, Computed Tomography (CT) images and MRI images. Image processing methods are adopted in segmenting the images. In the pre-processing stage mean and median filters are used. In the image segmentation stage, Otsu's thresholding and k-Means clustering segmentation approaches are used to segment the lung images and locate the tumors. To evaluate the performance of the methods used for segmentation, the performance evaluation parameters such as Signal to noise Ratio(SNR) ,Mean Square Error (MSE) and Peak Signal Noise to Ratio (PSNR)) are computed on the segmented images of the two different segmentation methods used for segmentation. Better results are obtained for the K-Means segmentation irrespective of the images. Keywords: Lung Cancer, Computed Tomography, MRI, Thresolding, K-Means Clustering, MSE, PSNR Cite this Article: V. Kannan and V. Jagan Naveen, Detection of Lung Cancer using Image Segmentation, International Journal of Electrical Engineering & Technology, 11(2), 2020, pp. 7-16. http://www.iaeme.com/IJEET/issues.asp?JType=IJEET&VType=11&IType=2 1. INTRODUCTION In the modern world, there is an increase in Lung cancer which is due to abnormal multiplication of certain cell multiplying and growing into tumor. Maximum cancers that begin within the lungs are called carcinomas. Two kinds of cancers are observed as small-cell lung carcinoma and nonsmall - cell lung carcinoma [1].Tobacco smoking for long years is the primary cause for 85% of lung cancers [2]. Approximately 10–15% of cases arise in humans who never smoked, but the illness is due to air pollution, 2nd hand smoking, asbestos, radon http://www.iaeme.com/IJEET/index.asp 7 editor@iaeme.com V. Kannan and V. Jagan Naveen fuel and so on. X-Ray imaging and Computer tomography (CT) are the traditional techniques to discover the affected lungs with cancer. The detection is done through biopsy that is commonly performed with the aid of bronchoscopy or CT scan. The cause of cancerassociated loss of life amongst men is specifically due to lung cancer. For this reason, it is crucial to determine a new strong method to diagnose the lung cancers at an earlier stage [3]. Lung cancer, also known as lung carcinoma, is a malignant tumor characterized with the aid of out of control growth of the tissue cells of the lungs. It is compulsory to treat this cancer to avoid spreading in to other parts of the body [5]. Cancer cells may be carried away from the lungs through blood, or lymph fluid that surrounds the tissues of lungs. Lymph fluid flows through lymphatic vessels and get drained into lymph nodes positioned inside the lungs and at the center of the chest. Lung cancers frequently spreads towards the center of the chest due to the fact that the herbal drift of lymph fluid flows out of the lungs is towards the center of the chest. Progression takes place when the cancerous cell leaves the site where it is originated and moves into a lymph node or to any other a part of the body via the blood circulation [4]. Cancerous tumor that got developed within the lung is referred to as primary lung cancer. CT and X-rays are the traditional techniques to detect the existence of lung cancers. The scanning is confirmed by using biopsy, that is generally finished by using bronchoscopy or CT experiment. The reason of most cancers-associated demise among men is specifically due to lung cancers. 1.1. Lungs and Lung Cancer Lungs consists of two parts as right lung and left lung within the chest. The right lung consists of 3 sections and they are called lobes. The left lung has 2 sections [6-8]. The right lung is bigger and the left lung is smaller in order to give space for the heart. Trachea is the wind pipe which takes air in to the lungs during breathing through nose or mouth. The trachea gets separates into tubes known as bronchi. The bronchi which go into the lungs is further separated into smaller bronchi. These smaller branches are called as bronchioles. At the edge of the bronchioles, minute air sacs are present called as alveoli. During breathe in the alveoli absorb oxygen into the blood and removes carbon dioxide from the blood while breathe out. Since the primary function of the lungs is to mix the oxygen in to the blood and removes carbon dioxide from it. Lung cancers usually start from the cells backing the bronchi and from the bronchioles or alveoli of the lungs. Figure 1 Anatomy of Human Lungs A protecting sheath surrounds the lungs known as the pleura which protects the lungs. This pleura assists the lungs to move backward and forward towards the chest wall during breathing. A slim, dome-shaped muscle present under the lungs, which separates the chest and http://www.iaeme.com/IJEET/index.asp 8 editor@iaeme.com Detection of Lung Cancer using Image Segmentation the abdomen is called diaphragm. During breathing, the air is forced in and out of the lungs by the up and down movement of the diaphragm. Most cancers begin, when there is an abnormal growth of cells in the body. A form of cancer that is found in the lungs are called Lung cancers. The two major types of lung cancer are small cell lung cancers and Non-small cell lung cancers (NSCLC).It is found that 85% of the lung cancers are NSCLC [9-10]. The NSCLC are further classified as large cell carcinoma, adenocarcinoma and squamous cell carcinomas. Adenocarcinomas start in the cells that secrete typical substance along with mucus [11]. Lung cancers are diagnosed mostly with people who smoke [12]. Lung cancers are also visible in people who do not smoke [13] which is due to second hand smoking. It is more common in ladies whose husbands smoke. Adenocarcinoma occurs in the exterior of the lung and it may be detected before it spreads. Squamous cells are flat cells that rim the inner way of the airways within the lungs. The cancer occur in it is called as Squamous cell carcinomas. This cancer is mostly related to the his-tory of smoking. Any part of the lung may be affected by large cell carcinoma. It is a dangerous cancer which may develop and spread quickly, making the treatment tougher. Large cell neuroendocrine carcinoma is a subtype of large cell carcinoma, which is a quickgrowing lung cancer. The cancer that start form the lung only is called as lung cancer .The cancers which spread from different organs like breast, kidney and pancreas to the lungs are not lung cancers [14]. 1.2. Causes of Lung Cancer The main cause of present days deaths are due to Lung cancers in both women and men. Today, nine out of 10 lung cancer deaths are caused by smoking. The various causes of lung cancers are discussed below 1.2.1. Smoking Cigarettes [15] are packed with cancer-inflicting chemical substances. In addition they spoil the lungs' natural protection system. To protect the lungs, the airways of the lungs are lined with tiny hairs known as cilia. They sweep out pollution, bacteria, and viruses from the air while breathing. The cilia is affected by the tobacco smoke and stops working in doing its job. This is the main reason for the cancer-causing chemicals to build up in the lungs and. the cancers starts quietly. During the early stages, no symptoms or warning are found due to cancer .But when it get worse, the following symptoms may be noticed  Chest pain, during deep breathing  Cough that won't go away  Bloody phlegm when Coughing  Wheezing or shortness of breath,  Fatigue 1.2.2. Second hand Smoke Even though the main cause of lung cancers is smoking, breathing in the second hand smoke released by a smoker is also danger. This second hand smoke which occur at domestic or at work place also may increase the risk of danger [16- 17]. It is found that woman‟s who are married to a smoking person are 20% to 30% more likely to get lung cancer when compared the woman‟s of non-smokers. 1.2.3. Risky work Some of the working environments may also be the cause of lung cancers . People who are working with the metals such as uranium, arsenic, and other chemical substances have more http://www.iaeme.com/IJEET/index.asp 9 editor@iaeme.com V. Kannan and V. Jagan Naveen chances for cancer and they should try to limit their exposure with such metals. People working with asbestos, for insulation, may also prone to lung cancer. 1.2.4. Radon Gas This natural radioactive gas [18-20] which can build up inside the houses may increase the risk of lung cancers. It is found to be the second leading cause of lung cancer in some countries. This gas can‟t be smelt or seen, but the presence of it can be identified using test kit. 1.2.5. Air Pollution Air pollution is also a cause for cancer, but it less prone when compared to smoking, but for good health it is better to avoid air pollutants [21-23]. It is found that pollutants from vehicles, power plants and factories also affect the lungs like second hand smoke does. Also it is cited that each and every car corporation is producing minimum of two hundred Cars per day in India which adds air pollution. 2. EARLY DETECTION OF LUNG CANCERS The method used to check or discover a disorder in an individual, when there is no signs and symptoms is called Screening. Initially X-Rays were studied for lung cancer screening. In recent years, CT test, MRI test has been taken to find people at a higher risk of getting lung cancer [24]. These scans can be used to locate the specific areas inside the lungs that can be cancerous. Research has shown that people at higher danger of lung cancers and screened with CT scans were saved at large, when compared to chest x-rays. For the people having higher risk of getting cancer may be screened yearly with CT or MRI scans. So that before the symptoms starts and it will help them in decreasing the risk of dying from lung cancer. Presently, Image processing methods are extensively used in a number of medical are-as. Since the early diagnosis of cancer can help in saving the life, Image processing methods are being developed in locating the cancerous tissues for treatment in the lungs, breast, and pancreas and so on. Image quality is very much important, which is obtained through X-Ray, CT and MRI Scans. The accuracy in locating the cancerous tumors may be obtained through Image processing methods such as image segmentation. 3. METHODOLOGY Lung cancers are found to be the most risky and widespread cancer in the world which cause more deaths. So there is a need for the early detection of the disease. Early detection can save the people from death by giving treatment at the early stage itself. The work is divided in to two stages 1. Image Filtering stage: Even though the images of the present age are more clear, in order to make the image better and accurate, image filtering is done to keep the image from noise, corruption or interference free. The Mean and median filters are used for filtering the images from noise. 2. Image Segmentation stage: In order to locate the cancerous issues in the lungs, a method called image segmentation is performed. Segmentation is a method of partitioning an image into regions which are greater meaningful for a specific cause. It is one of the steps to investigate and explain the image [25]. In the field of medicine, segmented images are often used to make it easy in analyzing the size, shape and borders of the selected organ of the patient. But, the process of image segmentation is not so easy as we think. The problems such as highlights, shadows, transparency, and item occlusion [26] are also to be considered during the segmentation process. Two different methods are used for segmenting the images in this research, they are Otsu‟s Thresholding approach and K-Means Clustering Segmentation approach. http://www.iaeme.com/IJEET/index.asp 10 editor@iaeme.com Detection of Lung Cancer using Image Segmentation The Otsu‟s Thresholding algorithm is based on a single intensity threshold .This threshold separate pixels into background and foreground pixels. The threshold is determined by minimizing the intra-class intensity variance and by maximizing the inter-class variance. In kmeans clustering, the objects are divided into several clusters mentioned by the number „K.‟ Here, the features or characteristics are compared, and all objects having similar characteristics are clustered together. The lung is a vital organ that has to be monitored often. For monitoring, the lungs Images may be obtained through any one of the imaging methods discussed. Then by using any one of the segmentation algorithms discussed, the cancerous tissues if any may be identified and treat-ed. The accuracy of segmentation methods plays an important role in identifying the lung disorders [27]. For evaluating the methods used for segmentation, some parameters of the processed images are calculated. Some of parameters used for evaluating the image processing techniques are signal to noise ratio (SNR), Peak signal to noise ratio (PSNR), timing run, histograms, Mean Square Errors (MSE), specificity, sensitivity, accuracy and so on. In this research the evaluation parameters considered are MSE [28], SNR [29] and PSNR[30]. The parameters of MSE, SNR and PSNR are evaluated using the equations as follows ∑ ∑ ∑ ∑ ∑ ∑ ) ) ) ) ) (1) (2) (3) Where, p(i, j) = the original image q(i, j)=the segmented output image MxN= the number of rows and columns in the images 4. RESULTS AND DISCUSSIONS The original images of X-RAY, CT and MRI of the abnormal lungs with tumour after the preprocessing are shown in Fig2a, Fig 3a, Fig. 4a respectively. Fig.2, Fig.3 and Fig.4 shows the segmentation of images using Otsu‟s segmentation algorithm. This method first converts the original colour image in to grayscale image and then works on it .The grayscale images of the lung with tumour are shown in Fig 2b, Fig3b and Fig4b respectively. The histograms of the grayscale images used for variance calculation are shown in Fig.2c, Fig.3c and Fig. 4c.Finally the segmented images using Otsu‟s segmentation algorithm is shown in Fig.2d, Fig.3d and Fig.4d. (a) Original X-Ray Image http://www.iaeme.com/IJEET/index.asp (b) Original Gray Scale Image 11 editor@iaeme.com V. Kannan and V. Jagan Naveen (c) Histogram (d) Segmented Image Figure 2 Otsu Segmentation of X-Ray Image (a) Original CT Image (b) Original Gray Scale Image (c) Histogram (d) Segmented Image Figure 3 Otsu Segmentation of CT Image (a) Original MRI Image http://www.iaeme.com/IJEET/index.asp (b) Original Gray Scale Image 12 editor@iaeme.com Detection of Lung Cancer using Image Segmentation (c) Histogram (d) Segmented Image Figure 4 Otsu Segmentation of MRI Image (a) Original X-Ray Image (c) Segmented Image Figure 5 K-Means Segmentation of X-Ray Image (a) Original CT Image (c) Segmented Image Figure 6 K-Means Segmentation of CT Image (a) Original MRI Image (c) Segmented Image Figure 7 K-Means Segmentation of MRI Image http://www.iaeme.com/IJEET/index.asp 13 editor@iaeme.com V. Kannan and V. Jagan Naveen Fig. 5, Fig. 6 and Fig. 7 shows the segmentation of images using K-Means Clustering algorithm. The segmented images using K-Means Clustering algorithm are shown in Fig. 5b, Fig.6b and Fig.7b respectively. It is found that by viewing the image itself, comparing the two image segmentation algorithms used in this research, K-Means Clustering algorithm shows a better result in segmenting the lung tumor. Also Otsu‟s segmentation algorithm has segmented to a certain extent on the CT and MRI images as shown in Fig2d and Fig3d and its performance is poor on X-Ray image as shown in Fig4d. The performance of the algorithms used are also evaluated by calculating the MSE, SNR and PSNR on the segmented images obtained from the two different methods.Table1 andTable2 shows the values of evaluation parameters for the images from the methods. Table 1 Otsu Segmentation of Images Image Type X-Ray Image CT Image MRI Image MSE 7154.94 9464.61 12704.75 SNR 5.39 4.74 4.37 PSNR 9.60 8.37 7.09 Table 2 K-Means Segmentation of Images Image Type X-Ray Image MSE 5859.47 SNR 6.26 PSNR 10.45 CT Image MRI Image 18334.38 17442.16 -0.53 3.00 5.50 5.71 It is observed from the results of the segmented images obtained from both Otsu‟s Thresholding and K-Means Clustering methods, better segmentation can be obtained using K-Means Clustering .Also lung cancer disease can be detected at early stages using K-Means Clustering algorithm and can increase the survival rate of the patients. 5. CONCLUSION The image segmentation techniques which may be useful for earlier disease detection for saving the precious human life is considered. Lung cancer is a disease widely spread on the smokers. As an early screening X-Ray, CT and MRI images are used. This research also concentrated on those images and included two image processing algorithms for image segmentations .The segmentation algorithms are Otsu‟s segmentation Algorithm and K-Means Clustering segmentation algorithm. It is found that by viewing the image itself and by comparing the two image segmentation algorithms used, K-Means Clustering algorithm shows a better result in segmenting the lung tumors irrespective of the type of images. Otsu‟s segmentation algorithm has segmented to a certain extent on the CT and MRI images and its performance is poor on X-Ray image. REFERENCES [1] A. A. Brindha, S. Indirani, and A. Srinivasan, Lung cancer detection using SVM algorithm and optimization techniques, Journal of Chemical and Pharmaceutical Sciences, vol. 9, no. 4, 2016. http://www.iaeme.com/IJEET/index.asp 14 editor@iaeme.com Detection of Lung Cancer using Image Segmentation [2] M. Kurkure and A. +akare, Introducing automated system for lung cancer detection us-ing Evolutionary Approach, International Journal of Engineering and Computer Science, vol. 5, no. 5, pp. 16736–16739, 2016. [3] B. Rani, A. K. Goel, and R. Kaur, A modified approach for lung cancer detection using bacterial forging optimization algorithm, International Journal of Scientific Research Engineering and Technology, vol. 5, no. 1, 2016. [4] Charles S, Dela Cruz, Lynn. T. Tanoue and Richard A Mattay, Lung Cancer: Epidemiology, Etiology and Prevention, Clinics inchest medicine, 32, 4, 2011. [5] Murray N, Coy P, Pater JL, Hodson I, Arnold A, Zee BC, Payne D, Kostashuk EC, Evans WK, Dixon P, et al. Importance of timing for thoracic irradiation in the combined modality treatment of limited-stage small-cell lung cancer. The national cancer institute of canada clinical trials group. Journal of clinical oncology: official journal of the American Society of Clinical Oncology. 11(2):336–344. 1993 [6] Tucker WD, Burns B. Anatomy, Thorax, Heart Pulmonary Arteries, Stat Pearls ,Stat Pearls Publishing; Treasure Island (FL): Dec 9, 2018. [7] Burlew JT, Banks KP. Stat Pearls, StatPearls Publishing; Treasure Island (FL): Feb 16, Anatomy, Thorax, Mediastinal Lymph Nodes. 2019. [8] Donley ER, Loyd JW. StatPearls, Stat Pearls Publishing; Treasure Island (FL): Feb 10, Anatomy, Thorax, Wall Movements. 2019 [9] S. Dearden, J. Stevens, Y.-L. Wu & D. Blowers, Mutation incidence and coincidence in non small-cell lung cancer: meta-analyses by ethnicity and histology (mutMap), Annals of Oncology 00: 1–6, 2013 doi:10.1093/annonc/mdt205 [10] Irene Ferrer, Jon Zugazagoitia, Stephan Herbertz, William John, Review of KRAS-Mutant non-small cell lung cancer: From biology to therapy, Lung Cancer, Volume 124, October Pages 53-64. 2018 [11] Kim H, Goo JM, Suh YJ, Park CM, Kim YT. Implication of total tumor size on the prognosis of patients with clinical stage IA lung adenocarcinomas appearing as part-solid nodules: Does only the solid portion size matter? Eur Radiol. Mar; 29 (3):1586-1594. 2019 [12] Muhammad Furrukh, Tobacco Smoking and Lung Cancer Perception-changing facts, Sultan Qaboos Univ Med J. Aug; 13(3): 345–358. 2013 [13] Jonathan M. Samet, Erika Avila-Tang, Paolo Boffetta, Lindsay M. Hannan, Susan OlivoMarston, Michael J. Thun, and Charles M. Rudin, Lung Cancer in Never Smokers: Clinical Epidemiology and Environmental Risk Factors, Clin Cancer Res. Sep 15; 15(18): 5626–5645. 2009 [14] Tracey A. Martin, Lin Ye, Andrew J. Sanders, Jane Lane, and Wen G. Jiang, Cancer Invasion and Metastasis: Molecular and Cellular Perspective, Metastatic Cancer: Clinical and Biological Perspectives edited by Rahul Jandial. Landes Bioscience. ©2013 [15] Sibu P Saha, Deepak K Bhalla, Thomas F Whayne, and CG Gairola, Cigarette smoke and adverse health effects: An overview of research trends and future needs, Int J Angiol. Autumn; 16(3): 77–83. 2007 [16] Kofi Asomaning, David P. Miller, Geoffrey Liu, John C. Wain, Thomas J. Lynch, Li Su, and David C. Christiani, Second hand smoke, age of exposure and lung cancer risk, Lung Cancer. Jul; 61(1): 13–20. 2008 [17] Leslie Stayner, James Bena, Annie J. Sasco, Randall Smith, Kyle Steenland, Michaela Kreuzer and Kurt Straif, Lung Cancer Risk and Workplace Exposure to Environmental Tobacco Smoke, Am J Public Health. March; 97(3): 545–551. 2007 http://www.iaeme.com/IJEET/index.asp 15 editor@iaeme.com V. Kannan and V. Jagan Naveen [18] IA RC Monographs on the Evaluation of Carcinogenic Risks to Hu-mans. Lung. In: Ionizing Radiation, Part 1: X- and Gamma (y)-Radiation, and Neutrons. Vol 75. Lyon, World Health Organization/IARC, pp. 253–254. 2000 [19] S Darby et.al. Radon in homes and risk of lung cancer: collaborative analysis of individual data from 13 European case-control studies, BMJ; 330:223. 2005 [20] Krewski D, Lubin JH, Zielinski JM, et al. A combined analysis of North American casecontrol studies of residential radon and lung cancer. J Toxicol Environ Health Part A; 69: 533–597. 2006 [21] A J Cohen and C A Pope, Lung cancer and air pollution, Environ Health Perspect. Nov; 103 (Suppl 8): 219–224. 1995 [22] Sandrah P Eckel, Myles Cockburn, Yu-Hsiang Shu, Huiyu Deng, Frederick W. Lurmann, Lihua Liu, and Frank D Gilliland, Air Pollution Affects Lung Cancer Survival, Thorax. Oct; 71(10): 891–898. 2016 [23] Xu X, Ha S, Kan H, Hu H, Curbow BA, Lissaker CT. Health effects of air pollution on length of respiratory cancer survival. BMC Public Health; 13(1):1–9. 2013 [24] N. Hollings, P. Shaw, Diagnostic Imaging of Lung Cancer, European Respiratory Journal 19: 722-742. 2002 [25] Sharma N, Aggrawal L M, “Automated medical image segmentation techniques”, Journal of Medical Physics, Jan-Mar, vol 35(1), pp 3-14. 2010 [26] Gonzalez R.C., Woods R.E., "Digital Image Processing using MATLAB", Upper Saddle River, NJ Prentice Hall, 2008. [27] Mokhled S. Al-Tarawneh, “Lung Cancer Detection Using Image Processing Techniques”, Leonardo Electronic Journal of Practices and Technologies, Issue 20, January- June 2012, p.147-158. ISSN 1583-1078 [28] Moreno et al. Towards no-reference of peak signal to noise ratio estimation based on chromatic induction model International Journal of Advanced Computer Science and Applications, 2013 [29] Zuzana Krbcova, Jaromir Kukal, Relationship between entropy and SNR changes in image enhancement, EURASIP Journal on Image and Video Processing. (1), 2017 [30] Wang, Zhou, Bovik and Alan C. Modern image quality assessment (USA: Morgan & Claypool). 2006 [31] Sreedhar T and Dr. Sathappan S, A Comparative Analysis of Image Segmentation Techniques. International Journal of Computer Engineering and Technology, 9(5), pp. 6469. 2018 [32] Regonda Nagaraju and M. Janga Reddy, Review of Medical Image Segmentation with Statistical Approach - State of the Art and Analysis. International Journal of Computer Engineering & Technology, 8(2), pp.36–55. 2017 [33] Amel H. Abbas, Aryan A. Kareem, Mohammed Y. Kami, Breast Cancer Image Segmentation Using Morphological Operations, International Journal of Electronics and Communication Engineering and Technology, 6(4), pp. 8–14. 2015 http://www.iaeme.com/IJEET/index.asp 16 editor@iaeme.com