The liver cancer rate is increasing year by year. Generally, liver cancer?s death rate is very high because the disease causes no symptoms, so it?s often not caught until it’s in final stages. We propose methods for liver cancer detection which is based on concepts of neural network. Wavelet and glcm features extracted from input image. This glcm features classify to dataset and then get result is normal or abnormal stages. Many techniques have been developed for the detection of liver tumor utilizing the aberrant lesion size and shape. The novel methodology for the detection and diagnosis of liver tumor is additionally proposed in this paper for the detection and diagnosis of liver tumor.
Early diagnosis is vital in treating a disease. Diagnosis relies on the skills, experience, and knowledge of the practicing physicians, although human errors may sometimes occur. Recently, various artificial intelligence-based methods are being increasingly used for liver disorder diagnosis to assist doctors in the diagnosis of patients. Extended cirrhosis and statuses liver disease may cause the appearance of malignant or benign tumors in the liver. Any abnormality in the liver adversely affects the rest of the body, as well as the general health of the patient Fatty liver disease (FLD) occurs when the human body produces a considerable amount of fat or does not efficiently metabolize fat. This excess fat is stored in the liver cells, where it accumulates triglycerides in the blood, causing FLD It is believed that FLD involves the pathogenesis of various common disorders, such as Type II diabetes and cardiovascular diseases. The liver cancer rate is increasing year by year. Generally, liver cancer?s death rate is very high because the disease causes no symptoms, so it?s often not caught until it’s in final stages. We propose methods for liver cancer detection which is based on concepts of neural network. Wavelet and glcm features extracted from input image. This glcm features classify to dataset and then get result is normal or abnormal stages
- Principal Component Analysis
- DCT ?and shape features
- KNN and SVM classifier
Drawbacks of Exisitng System
- High Computational load and poor discriminatory power.
- SVM is slow training for large feature set.
- Less accuracy in classification
- DTCWT and GLCM Features
- NN Classifier
- K-means Clustering
- The segmentation algorithm Proves to be simple and effective
- Grey scale Co-occurrence matrix performed well in NN
- Better texture and edge representation
- Segmentation provides better clustering efficiency
Block Diagram Explanation
Dual-tree complex wavelet transforms (DT-CWT)
The dual-tree complex wavelet transform (CWT) is a relatively recent enhancement to the discrete wavelet transform (DWT), with important additional properties: It is nearly shift invariant and directionally selective in two and higher dimensions. It achieves this with a redundancy factor of only 2d for d-dimensional signals, which is substantially lower than the undecimated DWT. The multidimensional (M-D) dual-tree CWT is no separable but is based on a computationally efficient, separable filter bank (FB). The theory behind the dual-tree transforms shows how complex wavelets with good properties can be designed, and illustrates a range of applications in signal and image processing.
In statistical texture analysis, texture features are computed from the statistical distribution of observed combinations of intensities at specified positions relative to each other in the image. According to the number of intensity points (pixels) in each combination, statistics are classified into first-order, second order and higher-order statistics.
?The Gray Level Co occurrence Matrix (GLCM) method is a way of extracting second order statistical texture features. The approach has been used in a number of applications, Third and higher order textures consider the relationships among three or more pixels. These are theoretically possible but not commonly implemented due to calculation time and interpretation difficulty.
Clustering can be taken into consideration the most crucial unsupervised learning problem so,? it offers with finding a form in a collection of unlabeled statistics. A cluster is therefore a set of items which might be ?comparable? amongst them and are ?extraordinary? to the items belonging to exceptional clusters
Clustering algorithms can be categorized as listed underneath
- Exclusive Clustering
- Overlapping Clustering
- Hierarchical Clustering
- Probabilistic Clustering
In the primary case records are grouped in a one-of-a-kind way, in order that if a positive datum belongs to a unique cluster then it could not be blanketed in every other cluster. On the alternative the second kind, the overlapping clustering, uses fuzzy units to cluster information, simply so everything can also moreover furthermore belong to two or extra clusters with extremely good stages of club. In this case, statistics can be related to the perfect club fee. A hierarchical clustering algorithm is primarily based completely totally on the union the various 2 nearest clusters. The starting scenario is determined out via setting every datum as a cluster. After some iteration it reaches the final clusters preferred.
- 4 GB of RAM
- 500 GB of Hard disk
- MATLAB 2014a
 Upadhyay, Y. and Wasson, V. 2014. “Analysis of Liver MR Images for Cancer Detection using Genetic Algorithm”. International Journal of Engineering Research and General Science. Vol.2, No.4, PP: 730-737.
? Kumar, P. Bhalerao, S. 2014. “Detection of Tumor in Liver Using Image Segmentation and Registration Technique”. IOSR Journal of Electronics and Communication Engineering (IOSR-JECE). Vo.9, No.2, PP: 110-115.
? Selle, D.; Spindler, W.; Preim, B. and Peitgen, H. O. 2000. “Mathematical Methods in Medical Imaging: Analysis of Vascular Structures for Liver Surgery Planning”. PP: 1-21.
 Zimmer, C. and Olivo-Marin, J. C. 2005. “Coupled Parametric Active Contours”. Transactions on pattern Analysis and Machine Intelligence. Vol.27, No.11, PP: 1838-1841.
? Chitra, S. and Balakrishnan, G. 2012. “Comparative Study for Two Color Spaces HSCbCr and YCbCr in Skin Color Detection”. Applied Mathematical Sciences. Vol.6, No.85, PP: 4229 ? 4238.