1. easier for the Physician to interpret,
1. ABSTRACTInrecent years, Ulcerative Colitis (UC) is most common and severe type of cancerand to detect its stages and severity is most challenging job. UlcerativeColitis (UC) is a common intestinal complication which causes polyps in therectum which may develop further into cancer. UC causes deaths of about half a millionpeople every year.
Colonoscopy images are obtained by process calledcolonoscopy. There are various shapes of polyps at various stages. Detection ofstages of UC is impossible by naked eyes. In an optical colonoscopy course, theendoscopist looks for colon polyps.
Hyperplastic polyp is benign lesion;adenomatous polyp is likely to become cancerous. Hence, it is common practiceto remove all of the identified polyps and send them to subsequent histologicalanalysis. So we are proposing a technique in which detection of infectious areacan be done with help of medical image processing. So location of infectiousarea, blood clotting can be detected with help of color impact. Shape andtexture detection is used for checking type of polyp. The technique can beuseful in designing the treatment of UC and will help in prevention of disease.2. INTRODUCTION2.
1 MEDICAL IMAGE PROCESSINGMedicalimaging is the technique and process of creating visual representations ofthe interior of a body for clinical analysis and medical intervention. Medicalimaging seeks to reveal internal structures hidden by the skin and bones, aswell as to diagnose and treat disease. Medical imaging also establishes adatabase of normal anatomy and physiology to make itpossible to identify abnormalities. Medical imaging is the technique andprocess of creating visual representations of the interior of a body forclinical analysis and medical intervention, as well as visual representation ofthe function of some organs or tissues.
Advantagesof Digital Processing for Medical Applications are Digital data will not changewhen it is reproduced any number of times and retains the originality of thedata. It offers a powerful tool to physicians by easing the search forrepresentative images, displays images immediately after acquiring, enhancementof images to make them easier for the Physician to interpret, quantifyingchanges over time, providing a set of images for teaching to demonstrateexamples of diseases or features in any image also quick comparison of images.2.2 ULCERATIVE COLITISAchronic, inflammatory bowel disease that causes inflammation in the digestivetract. Ulcerative colitis (Colitis ulcerous, UC) is a form of inflammatorybowel disease (IBD) that causes inflammation and ulcers in the colon. Thedisease is a type of colitis, which is a group of diseases that causeinflammation of the colon, the largest section of the large intestine, eitherin segments or completely. Due to stress and other factors like acidic reactionetc.
there exists this disease with increasing number. More than 1 millioncases per year (India) are found. Treatment can help, but this condition can’tbe cured.
In this disease it requires a medical diagnosis Lab tests or imagingalways required Chronic. UC can last for years or be lifelong. Ulcerativecolitis is usually only in the innermost lining of the large intestine (colon)and rectum. Forms range from mild to severe. Having ulcerative colitis puts apatient at increased risk of developing colon cancer. Symptoms include rectalbleeding, bloody diarrhea, abdominal cramps and pain.
Treatment includesmedication and surgery. Figure1. Colonoscopy Images.
Endoscopicimage of a bowel section known as the sigmoid colon afflicted with ulcerativecolitis. The internal surface of the colon is blotchy and broken in places. Thedisease has most important 4 stages and the last of them leads to cancer. Theaffected part has various textures in various stages. Since these patterns areimpossible to detect with naked eyes by the doctor, PIT pattern and TextureRecognition techniques are used. The most important algorithms used are relatedto various feature extraction techniques.
Figure2. Ages affected in UC.2.3 COLONOSCOPYColonoscopyor coloscopy is the endoscopic examination of the large bowel and the distalpart of the small bowel with a CCD camera or a fiber optic camera on a flexibletube passed through the anus. It can provide a visual diagnosis (e.g.
ulceration, polyps) and grants the opportunity for biopsy or removal ofsuspected colorectal cancer lesions. Colonoscopy can remove polyps as small asone millimeter or less. Once polyps are removed, they can be studied with theaid of a microscope to determine if they are precancerous or not. It can takeup to 15 years for a polyp to turn cancerous. Colonoscopy screening preventsapproximately two thirds of the deaths due to colorectal cancers on the leftside of the colon, and is not associated with a significant reduction in deathsfrom right-sided disease.Figure3. ColonoscopyThe colonoscopy isperformed by a doctor experienced in the procedure and lasts approximately30-60 minutes.
Medications will be given into your vein to make youfeel relaxed and drowsy. You will be asked to lie on your left side on theexamining table. During a colonoscopy, the doctor uses a colonoscope, along, flexible, tubular instrument about 1/2-inch in diameter that transmits animage of the lining of the colon so the doctor can examine it for anyabnormalities. The colonoscope is inserted through the rectum and advanced tothe other end of the large intestine. 2.
4STAGES OF ULCERATIVE COLITISPit pattern: The motivationbehind computer based pit pattern classification is to assist the physician inanalyzing the colon images taken with a colonoscope just in time. Thus aclassification can already be done during the colonoscopy and therefore thismakes a fast classification possible.Figure 4. Pit Pattern 2.5 IMPORTANCE OF THE RESEARCH · 10 to 15 years for a polyp to develop into colourectal cancer.· When cancer has spread outside the colon or rectum, survivalrates are lower.· Risk factors – overweight or obese, physically inactive,diet, Smoking, heavy alcohol use· Colourectal cancer risk factors you cannot change : Beingolder, personal and family history, Having type 2 diabetes· Factors with unclear effects on colourectal cancer risk :Night shift work, Previous treatment for certain cancers.
3. LITERATURE SURVEYColorectalcancer is largest death causing disease in world. So to analyzing this diseasemedical treatment which is suggested is colonoscopy 15. Kudo et al. 5 haddiscussed about patterns which are being produced by colonoscopy results.
Theydivided pit patterns into seven principal types: (1) normal round pit; (2)small round pit; (3) small asteroid pit; (4) large asteroid pit; (5) oval pit;(6) gyrus-like pit; and (7) non-pit. So further distinguish done by authors 916 14 on this Kudo’s pit patterns. Pablo Mesejo et al.9 discuss aboutclassification of colonoscopy videos are classified in 3 classes as Adenoma,Hyperplastic and serrated adenoma. Here machine learning algorithms are used,it will help clinicians in virtual biopsy of hyperplastic, serrated adenoma andadenoma.
So technique is developed for diagnosis gastrointestinal lesions fromregular colonoscopy. So by using this technique systematic biopsy for suspectedhyperplastic tissues also 3D shape features improves classification accuracy. Also many techniques analyses aboutNBI images 16 3 7 17 for classification. Toru Tamaki et al. 16divides NBI images into 3 classes as Type A, B, C3. For classification localrecognition method as Bag-of-Words is introduced along with SVM classifier.Local features are considered and checks result for recognition checking.
HaoChun Wang et al.3 classifies according to Classification of Regional Feature(CoRF) which is extension of sparse matrix and vector quantization for featuredetection and segmentation. Mineo Iwatate et al. 7 discuss about efficiencyof magnifying chromoendoscopy and magnifying colonoscopy with NBI fordetection, histological prediction, estimation of depth of early colourectalcancer and future perspective. Here, NBI International Colourectal Endoscopic(NICE) classification is introduced here. Yasushi Sano et al.
17 discussesabout the work by Japan NBI Expert Team (JNET) where they discussed about Sano,Hiroshima, Showa, And Jike Classifications Based on The Findings Of NBIMagnifying Endoscopy. Also they discussed about Universal NBI MagnifyingEndoscopic Classification of Colorectal Tumors: Japan NBI Expert Team (JNET)Classification which is universal solution which has overcome the problemsraised by previous methods. The JNET classification combining previousclassifications to give common diagnostic criteria to promote academic progressof NBI. For detection of polyp fromcolonoscopy 4 14 various techniques are introduced. Ju Lynn Ong et al. 4gives idea about features of image like geometric feature, colon wallextraction constrction of probability density functions(PDF), comparison ofshape distribution. Here K-L divergence used for comparison between PDF forspecific image and previous database.
Yuan Shen and Christopher L. Wyatt 14uses feature extraction method for Computer Aided polyp Detection (CAPD) onbasis of principal curvature, Gaussian curvature and mean curvature, shapeindex, curvedness, maximum and minimum polyp radius etc. Principle ComponentAnalysis is used with wrapper method. For feature extraction of imagesvarious techniques12613 used for extracting features. Adegoke,B.O.
et al. 1 surveyed about medicalimage feature extraction. They researched about CBIR (Content- Based ImageRetrieval). The algorithms used in these systems are commonly divided intothree types as Extraction, Selection and Classification.
Different available medical image featureextraction had been studied in this paper. G.Nagarajan et al.2 proposes minimum description length principle basedgenetic algorithm (GA) approach for the selection of dimensionality reduced setof features. There are 3 phases are developed as for the extraction of thefeatures are Texton based contour gradient extraction algorithm, Intrinsicpattern extraction algorithm and modified shift invariant featuretransformation algorithm. For second phase to identify the potential featurevector GA based feature selection is done, with help of “Branch and Bound Algorithm” and “ArtificialBee Colony Algorithm”. In the third phase to improve the performance of thehybrid content based medical image retrieval system diverse density basedrelevance feedback method is used.
For this algorithm they used techniques suchas Intrinsic pattern extraction algorithm using PCA. The branch and boundalgorithm is used to give reduced feature vector. M.VASANTHA et al. 6researched about breast cancer where proposes an image classifier to classifythe mammogram images. For preprocessing they used low pass filter to removenoise.
In this Work intensity histogram features and Gray Level Co-OccurrenceMatrix(GLCM) features are Extracted. For classification we used J48 classifier,a decision tree classifier based on C4.5, from WEKA to train and test thefeatures. Seyyid Ahmed Medjahed et al.13 has done a comparative Study of Feature ExtractionMethods in Images Classification, in which they had discussed about FeatureExtraction Techniques and classifiers on the Cal-tech 101 image dataset. Theperformance measures used to evaluate and analyze the results are:classification accuracy rate, Precision, Recall, F_measure, G_mean, AUC and theRoc Curve. Similarly some techniques have beendeveloped for x-ray results1012 which also helpful for feature extraction.Randa Hassan Ziedan et al.
10 proposesa technique for classification of x-ray images, where they discussed aboutfeature extraction techniques GLCM, LBP, Canny and BoW. Also a comparativestudy for this techniques for x-ray images. Seyyed Mohammad Mohammadi etal.12 also proposed shaped texture feature extraction for x-ray images. Inresearch Novel Shape Texture feature is proposed with help of histogramadjustment, Noise removal, Edge and boundary extraction, phase congruencycomputation, Gabor Transformation, shape-texture feature extraction with helpof classifiers as Euclidean Distance, PNN( probabilistic Neural network) andSVM. SaimaRathore et al. 11 discusses about colon cancer detection techniques as regionbased segmentation methodologies, classification and segmentation, graph basedtechniques, automated diagnosis system etc.
Similarly Shiva K Ratuapli etal.15 gives idea about post colonoscopy follow-up with help of previousscreening and surveillance colonoscopy. Mohammad Sohrabinia et al. 8 usesdifferent image analysis and processing methods in order to extract informationcontent needed to update large scale maps.4. PROPOSED METHODIn this project, we areproposing a method for polyp detection. 1) Red color filtering – · on loaded image so by filtering most red saturated area is onlymaintained.· It can be considered by clinicians as most infected area.
2) RoI selection – · From loaded image, Region of Interest is selected.3) Shape/Texture detection –· Edge detection and shape detection is done.· For edge detection, we have analyzed canny edge detector and sobeledge detection which will give edge of RoI. 4) Classifier – · Classifier will be useful in classifying Shape/Texture accordingto classes.
5) Final result – · Final result will be combination of red colour filtering and RoIanalysis. 4.1 Canny Edge DetectorThealgorithm runs in 5 separate steps: 1.Smoothing: Blurring of the image to remove noise. 2.
Finding gradients: The edges should be marked where the gradients of the imagehas large magnitudes. 3.Non-maximum suppression: Only local maxima should be marked as edges. 4.
Double thresholding: Potential edges are determined by thresholding. 5.Edge tracking by hysteresis: Final edges are determined by suppressing all edgesthat are not connected to a very certain (strong) edge.
4.2 Sobel Edge DetectorFigure7. Sobel Edge Detector4.3 Red colour filteringFigure8. Red Colour Filtering5. CONCLUSION Withhelp of this project, we are trying to take advantage and make a project for asocial cause which will surely help doctors with less expertise and in remoteareas to detect the stage and the probable infections in the colon image. Itwill help doctors to detect the abnormalities which are not able to be seen bythe naked eye.Theproposed techniques will surely help doctor to get faster result so that timewhich is spent on post colonoscopy will be avoided.
Also, in rural area orunprivileged area where lack of technology is challenge then this technologycan give idea about disease severeness. 6. ACKNOWLEDGEMENT I take thisopportunity to thank my project guide Prof.Dr.
Siddhivinayak Kulkarni, my project co-guide Prof. Deepali Javale and ME project coordinator Prof. Jayshree Ghorpade-Aher for their valuable guidance and for providingall the necessary facilities, which were indispensable in the completion ofthis seminar report. I am also thankful to Head of the Department Dr.Bharati Dixit and all the staff members of the Computer Department of MAEER’S M.I.
T COLLEGE OF ENGINEERING, PUNE.I would alsolike to thank the institute for providing the required facilities, Internetaccess and important books. REFERENCES1 Adegoke,B.O, Ola, B.O. and Omotayo, M.E, “Reviewof Feature Selection Methods in Medical Image Processing”, IOSR Journal ofEngineering (IOSRJEN), Vol.
04, Issue 01 (January. 2014), ||V4|| PP 01-052 G.Nagarajan, R.I.
Minu, B Muthukumar, V.Vedanarayanan .D.Sundarsingh, “Hybrid Genetic Algorithm for Medical Image Feature Extractionand selection”, In Procedia Computer Science 85 ( 2016 ), 455 – 4623 Hao Chun Wang, Wei Ming Chen, Yen Pin Lin, Wei Chih Shen,”Tumor Detecting in Colonoscopic Narrow-Band Imaging Data”, In 2012 IEEEInternational Symposium on Intelligent Signal Processing and CommunicationSystems (ISPACS 2012) November 4-7, 2012, 564 – 568.4 Ju Lynn Ong, Abd-Krim Seghouane, Kevin Osborn,” PolypDetection In Ct Colonography Based On Shape CharacteristicsAnd Kullback-LeiblerDivergence”, 978-1-4244-2003-2/08/$25.00 ©2008 IEEE5 Kudo, S.
, Hirota, S., Nakajima, T., Hosobe, S., Kusaka, H.,Kobayashi, T.
, Himori, M., Yagyuu, A.,1994, “Colourectal tumours and pit pattern”. In Journal of Clinical Pathology47, 880–885.6 M.
VASANTHA, DR.V.SUBBIAH BHARATHI, R.DHAMODHARAN, “MedicalImage Feature, Extraction, Selection And Classification”, International Journalof Engineering Science and Technology, Vol.
2(6), 2010, 2071-20767 Mineo Iwatate, Taro Ikumoto, Santa Hattori, Wataru Sano,Yasushi Sano, and Takahiro Fujimori, “NBI and NBI Combined with MagnifyingColonoscopy”, In Hindawi Publishing Corporation, Diagnostic and TherapeuticEndoscopy, Volume 2012, Article ID 173269, 11 pages, doi:10.1155/2012/1732698 Mohammad Sohrabinia, Saeid Sadeghian, Dadfar Manavi,”Application of Image Processing and Image Analysis Methods ForLarge Scale MapRevision”, In the International Archives of the Photogrammetry, Remote Sensingand Spatial Information Sciences. Vol.
XXXVII. Part B4. Beijing 2008,1351-1354.9 Pablo Mesejo, Daniel Pizarro, Armand Abergel, OlivierRouquette, Sylvain Beorchia, et al.”Computer-Aided Classification ofGastrointestinal Lesions in Regular Colonoscopy”. In IEEE Transactions onMedical Imaging, Institute of Electrical and Electronics Engineers, 2016,35(9), pp.
2051 – 2063. 10 RandaHassan Ziedan, Mohamed A. Mead, Ghada S. Eltawel, ” Selecting the AppropriateFeature Extraction Techniques for Automatic Medical Images Classification”, In International Journal ofEmerging Engineering Research and Technology Volume 4, Issue 5, May 2016, PP1-911 Saima Rathore, Mutawarra Hussain, Ahmad Ali, Asifullah Khan,”A Recent Survey on Colon Cancer Detection Techniques”,ieeexplore.ieee.
org/iel7/8857/4359833/06565986.pdf12 Seyyed Mohammad Mohammadi, Mohammad Sadegh Helfroush, Kamran kazemi,”Novel ShapeTexture Feature Extraction for medical X-Ray image classification” , InInternational journal of Innovative Computing, Information and Control, Volume8, Number 1(B), January 2012, pp 659-67613 SeyyidAhmed Medjahed, “A Comparative Study of Feature Extraction Methods in ImagesClassification”, In I.J. Image, Graphics and Signal Processing,2015, 3, 16-2314 Shen, Y. and Wyatt, C. L., “Open implementation of featureextraction methods for computer aided polyp detection with principal componentanalysis,” in 11th International Conference on Medical Image Computing andComputer Assisted Intervention, 141-147 (2008).
15 Shiva K. Ratuapli, Suryakanth R. Gurudu, Mary A. Atia,MichaelD. Crowell, Sarah B. Umar, M.
Edwyn Harrison, Jonathan A. Leighton, andFrancisco C. Ramirez, “Post colonoscopy Follow-up Recommendations: Comparisonwith and without Use of Polyp Pathology”, In HindawiPublishing Corporation, Diagnostic and erapeutic Endoscopy, Volume 2014,Article ID 683491, 7 pages, http://dx.doi.org/10.1155/2014/68349116 Toru Tamaki, Junki Yoshimuta, Misato Kawakami, BisserRaytchev, Kazufumi Kaneda, Shigeto Yoshida, Yoshito Takemura, Keiichi Onji, RieMiyaki, Shinji Tanaka, “Computer-Aided Colorectal Tumor Classification in NBIEndoscopy Using Local Features”, In Medical Image Analysis, December 17, 201317 Yasushi Sano, Shinji Tanaka, Shin-ei Kudo, Shoichi Saito,Takahisa Matsuda, Yoshiki Wada, Takahiro Fujii, Hiroaki Ikematsu, ToshioUraoka, Nozomu Kobayashi, Hisashi Nakamura, Kinichi Hotta, Takahiro Horimatsu,Naoto Sakamoto, Kuang-I Fu, Osamu Tsuruta, Hiroshi Kawano, Hiroshi Kashida,Yoji Takeuchi, Hirohisa Machida, Toshihiro Kusaka, Naohisa Yoshida, IchiroHirata, Takeshi Terai, Hiro-o Yamano, Kazuhiro Kaneko, Takeshi Nakajima, TakuSakamoto, Yuichiro Yamaguchi, Naoto Tamai, Naoko Nakano, Nana Hayashi, ShiroOka, Mineo Iwatate, Hideki Ishikawa,Yoshitaka Murakami, Shigeaki Yoshida andYutaka Saito, “Narrow-band imaging (NBI) magnifying endoscopic classificationof colorectal tumors proposed by the Japan NBI Expert Team”, DigestiveEndoscopy 2016; 28: 526–53318 https://www.gstatic.com/healthricherkp/pdf/ulcerative_colitis_en_IN.pdf19 https://www.webmd.com/colorectal-cancer/colonoscopy-what-you-need-to-know#1