A recent survey on cancer detection and diagnosis using image processing techniques.
INTRODUCTIONProcessing of image using mathematical formulation or notations is simply called as image processing where the image is taken as a input, it can be processed using mathematical functions or computerized algorithms and produces an output in form of images or any parameter depends on image. This process can be broadly classified as analog and digital image processing. In analog image processing the image can be taken as an input in the form of waves or signals and in digital image processing the input image can be any form like a digital signals or binary representation etc. The digital processing has more scope than analog. In recent years all the image processing taken under the digital image processing.
A. Need of image processing in medical field:
To produce an accurate result or prediction about an action some processing are needed. As like that in day to day life different number of diseases are grown and spread in and around the world, the doctors are dealing patients in normal stage, malignant, more suspicious symptoms of cancer cells. They are using digital images such as X-ray, MRI or some other scanned images to predict whether that particular patient have disease or not, if the disease is present they must identify the level of disease. It is very big challenge to differentiate suspicious cells from normal to identify the disease, in such a situation the image processing techniques are used to improve the accuracy of prediction.
Literature Review:
Mokhled S et al discussed image processing techniques used in several medical areas for image improvement in earlier detection and treatment stages, where the time factor is very important to discover the abnormality issues in target images, especially in various cancer tumors such as lung cancer, breast cancer, etc. Image quality and accuracy is the core factors of this research, image quality assessment as well as improvement are depending on the enhancement stage where low pre-processing techniques is used based on Gabor filter within Gaussian rules. Following the segmentation principles, an enhanced region of the object of interest that is used as a basic foundation of feature extraction is obtained. Relying on general features, a normality comparison is made. In this research, the main detected features for accurate images comparison are pixels percentage and mask-labeling. [1]
Lung cancer has been attracting the attention of medical and sciatic communities in the latest years because of its high prevalence allied with the difficult treatment. Statistics from 2008 indicate that lung cancer, throughout world, is the one that attacks the greatest number of people. Early detection of lung cancer is very important for successful treatment. There are few methods available to detect cancerous cells. Bhagyashri G. Patildiscussed two methods of segmentation such as thresholding and watershed to detect the cancer cell and too find out better approach out of them. [2]
Tobias Christian Cahoon et al. describe the segmentation with fuzzy models and classification by the crisp k-nearest neighbor algorithm for assisting brest cancer detection in digital mamograms. The supervised and unsupervised methods of segmentation such as k-nn and fuzzy c-means in digital mamograms will have high misclassification rates when only intensity is used as the discriminating feature. [3]
Lung cancer seems to be the common cause of death among people throughout the world. Although Computed Tomography (CT) can be more efficient than X-ray. However, problem seemed to merge due to time constraint in detecting the present of lung cancer regarding on the several diagnosing method used. Hence, a lung cancer detection system using image processing is used to classify the present of lung cancer in a CT-images. In this study, MATLAB have been used through every procedures made. In image processing procedures, process such as image pre-processing, segmentation and feature extraction have been discussed by Vijay A. Gajdhane et al. also get more accurate results by using various enhancement and segmentation techniques. [5]
Hossein Ghayoumi Zadeh et all discussed an image analysis approach for automated detection, segmentation, and classification of particular cells, specially the cancer cells from normal cells is introduced. The number of defected cells and their position are identified using image processing. The results of this analysis are useable in designing a neural network for more accurate analysis. The particular cells segregation is the most important property of this work. Using the LABVIEWSoftware gave practical usages to this image processing system because it can communicate with other equipments used in this system and controls them in order to have an automatic system and demonstrates the potential effectiveness of such a system on diagnostic tasks that require the classification of individual cells. [4]
Vibin kumar jain et al. designed ROI system to detect the lungs caner from a MRI image, taken from a particular angle. Which shows a big spot in left part of lung that may be suspicious object, from which some part is extracted that is our ROI. Another sample of normal image is also extracted. We compare the intensity of value of both image samples and observe that cancer infected flash image has very much variation in intensity values at many palaces while normal flash image don't show any big variation in intensity values in image. [6]
Dishna sharma et al. proposed the automated Computer Aided Diagnosing (CAD) system is proposed in this paper for detection of lung cancer form the analysis of computed tomography images. To produce a successful Computer Aided Diagnosis system, several problems has to be resolved. In recent years the image processing mechanisms are used widely in several medical areas for improving earlier detection and treatment stages, in which the time factor is very important to discover the disease in the patient as possible as fast, especially in various cancer tumors such as the lung cancer, breast cancer. This system generally first segments the area of interest (lung) and then analyzes the separately obtained area for nodule detection in order to diagnosis the disease. Initially, the basic image processing techniques such as Erosion, Median Filter, Dilation, Outlining, and Lung Border Extraction are applied to the CT scan image in order to detect the lung region. Then the segmentation algorithm is applied in order to detect the cancer nodules from the extracted lung image. After segmentation, rule based technique is applied to classify the cancer nodules. Finally, a set of diagnosis rules are generated from the extracted features. For experimentation of the proposed technique, the CT images are obtained from a NIH/NCI Lung Image Database Consortium (LIDC) dataset that provides the chance to do the suggested research. DICOM (Digital Imaging and Communications in Medicine) has become a standard for medical imaging. Its purpose is to standardize digital medical imaging and data for easy access and sharing. There are many commercial viewers that support DICOM image format and can read metadata. [7]
Aravind kumar Tiwari et al. discussed prediction of lung cancer with the structure of cancer cell, where most of the cells are overlapped each other. The image processing techniques are mostly used for prediction of lung cancer and also for early detection and treatment to prevent the lung cancer. To predict the lung cancer various features are extracted from the images therefore, pattern recognition based approaches are discussed. Here, a comprehensive review for the prediction of lung cancer by previous researcher using image processing techniques is presented. [8]
Image Processing Stages:
In image processing, to clarify the disease it is necessary to classify the input images under several stages such as Image acquisition, Preprocessing, Processing and Feature extraction.
A. Image Acquisition:
Image acquisition is the initial stage in all the image processing work flow, where an input image can be taken from any resources. The image chosen as input is completely unprocessed and it may take by any hardware devices. The main aim of this is to have an image as a source of input operates within such controlled and measured guidelines so that an abnormality can be identified easily. From this base image the quality will be improved after applying few image processing techniques.
B. Image Preprocessing:
An original image take from acquisition make have few error, in an image error in the sense the data or detail present in the image may not visible clearly. In other way the captured picture by hardware devices may at lowest level of abstraction. The scope of image preprocessing is to improve or enhance the detail or data of an image for further processing. Simply this is a stage where an image quality can be improved for upcoming stages.
This image processing can be done by the following two stages noise removal, image enhancement.
1) Noise Removal:
The sample input image may distorted with several unwanted data may called as noise. In this preprocessing stage those data must be removed with affecting the original image.
2) Image enhancement:
Several methods are used to enhance the quality of image with out affecting its originality are
* Average filter
* Median filter
* Gabor filter
* Auto enhancement
A. Average filter:
Average filter is also called as Mean filter, this type of filtering method can be suitable for removing noise in photograph and also it reduces the intensity value among its neighbour pixels to remove the noise. Basically the image is represented by pixels in matrix form where a mean value can be calculated among all the neighbourhood and it finally sampled.
B. Median filter:
As like mean filter an average can be calculated among all the neignbourhood, it is sampled and all other pixels are replaced by this median value for sampling to remove the noise.
C. Gobarfilter:
Once the input images are sampled it must be classified for this type of classification few of the features are set as condition such as color, shape, texture, size etc., The gobar filter are used for edge based and texture based classification.
D. Auto enhancement:
Few computerised algorithms are used to enhance some quality measures in image like color, sharpness of an image, clarity, low noise and intensity values, such types of enhancement are known as auto enhancement.
3) Image segmentation:
Image segmentation is part of work in image processing work flow where the extracted, enhanced images are taken as input, it can be segmented by any of the following method
* Thresholding
* Clustering methods
* Histogram
A. Thresholding:
The simplest segmentation is thresholding method, like mean and median filter a common threshold value can be set as region or boundary other pixel values in an image can be compared with threshold and finally the decision can be made.
Simply this is based on threshold where the gray scale image can be easily converted into binary image.
B. Clustering methods:
A computerised algorithm can be used at this stage to group the images which are all have common features called clusters. The basic cluster algorithm is,
* Set cluster centers either randomly or based on some other method
* Calculate the distance that is squared or absolute difference between a pixel and a cluster center, assign each pixel in the image to the cluster that minimizes the distance between the pixel and the cluster center
* Re-compute the cluster centers by averaging all of the pixels in the cluster
* Repeat steps 2 and 3 until convergence or none of the pixels change its clusters is attained
C. Histogram:
Histogram-based image segmentation is one of the simplest and most often used segmentation techniques. It uses the histogram to select the gray levels for grouping pixels into regions. In a simple image there are two entities: the background and the object. The background is generally one gray level and occupies most of the image. Therefore, its gray level is a large peak in the histogram. The object or subject of the image is another gray level, and its gray level is another, smaller peak in the histogram.
4 Feature extraction:
From the input segmented images the features are derived which must be non-redundant and informative to further processing. This process is used reduce amount of resources needed to classify or identify the problems in image. Many number of techniques are used for this feature extraction such as
* Edge detection
* Corner detection
* Blob detection
* Ridge detection
* Thresholding
Conclusion:
A flow of image processing is discussed earlier in this paper, through that flow the input images are processed to enhance the quality which make the classification and identification process more easier. A series of filters will be used including average filters, Median, Auto enhancement and Gobar filter to remove noise. The research has also identified several medical limitations and contributions that can be done in future works.
REFERENCES
[1.] Croisille, C. et al., 1995. "Pulmonary nodules: Improved detection with vascular segmentation and extraction with spiral CT", Radiology, 197: 397-401.
[2.] Tozaki, T. et al., 1996. "Three-dimensional analysis of lung area using thin slice CT images", Medical Imaging Proc SPIE, 2709: 2-11.
[3.] Giger, M.L. et al., 1994. "Computerized detection of pulmonary nodules in computed tomography images", Investigative Radiology, 29(4): 459-465.
[4.] Toshioka, S. et al., 1997. "Computer aided diagnosis system for lung cancer based on helical CT images", Image Processing: KM Hanson, Proc SPIE 3034: pp: 975-984.
[5.] Toriwaki, J. et al., 1977. "Fundamental properties of the gray weighted distance transformation", Trans IEICE Japan, J60-D(12): 1101-1108.
[6.] Cavouras, D. et al., 1992. "Image analysis methods for the solitary pulmonary nodule characterization by computed tomography". Eur J Radiol., 14: 169-172.
[7.] McNitt-Gray, M.F. et al., 1997. "Characterization of solitary pulmonary nodules using features extracted from high resolution CT images". Radiology, 205(P): 395.
[8.] McNitt-Gray, M.F. et al., 1998. "The application of image analysis techniques to distinguish benign from malignant solitary pulmonary nodules imaged on CT". Proc SPIE Med Imag 3338.
[9.] Kawata, Y. et al., 1997. "Classification of pulmonary nodules in thin section CT images based on shape characterization". IEEE Int Conf Image Proc, 3: 528-531.
[10.] Summers, R.M. et al., 1998. "Automatic detection of endobronchial lesions with virtual bronchoscopy: comparison of two methods". Proc SPIE Med Imag 3338.
[11.] Reeves, A.P. et al., "Characterization of three-dimensional shape and size changes of pulmonary nodules over time from helical CT images". Radiology, 205(P): 396.
[12.] Yankelevitz, D.F. et al., 1997. "Can computer-aided morphologic analysis of solitary pulmonary nodules predict pathology?" Radiology, 205(P): 529.
[13.] Brain Tumors and Brain Cancer, 2013. Retrieved from: http://www.cedars-sinai.edu/Patients/HealthConditions/Brain -Tumors-and-Brain-Cancer.aspx, Retrieved on: 3.
[14.] American Brain Tumor Association, Facts and statistics, 2010. Retrieved from: http://www.abta.org/sitefiles/pdflibrary/ABTA-FactsandStatistics2010.pdf, Retrieved on: 3.
[15.] Central Brain Tumor Registry of the United State, 2010. CBTRUS statistical report: Primary brain and central nervous system tumors diagnosed in the United States in 2004-2006. Retrieved from: http://www.cbtrus.org/2010-NPCR-SEER/CBTRUS, Retrieved on: 3 December 2013.
[16.] Nagalkar, V.J., S.S Asole, "Brain Tumor Detection Using Digital Image Processing Based On Soft Computing", Journal Of Signal And Image Processing, 3(3): 102-105.
[17.] Dahab, D.A. et al., 2012. "Automated Brain Tumor Detection and Identification Using Image Processing and Probabilistic Neural Network Techniques", International Journal of Image Processing and Visual Communication ISSN2319-1724 1: 2.
(1) Gomathi S, (2) Radha C, (3) Saranya P, (4) Elavenil M, (5) Ilakkiya K S
(1) Departmenof Computer Science and Engineering, (3) Department of MCA, SNS college of Technology, Muthayammal Engineering College Address
(2) Assistant Professor, SNS College of Technology, Coimbatore.
(4) Assistant Professor, Muthayammal Engineering College, Rasipuram. Address Including Country Name
(5) Assistant Professor, Muthayammal Engineering College, Rasipuram.
Received 28 January 2017; Accepted 22 March 2017; Available online 28 April 2017
Address For Correspondence:
Gomathi S, Departmen of Computer Science and Engineering, #3Department of MCA, SNS college of Technology, Muthayammal Engineering College Address
E-mail: gomkct@gmail.com
Caption: Fig. 1: Architecture for image processing stages
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Author: | Gomathi, S.; Radha, C.; Saranya, P.; Elavenil, M.; Ilakkiya, K.S. |
|---|---|
| Publication: | Advances in Natural and Applied Sciences |
| Article Type: | Survey |
| Date: | Apr 30, 2017 |
| Words: | 2626 |
| Previous Article: | Privacy preserving access control on data aggregation for wireless sensor networks. |
| Next Article: | Literature survey on multiple image watermarking techniques with genetic algorithm. |
| Topics: | |