CLUSTERING APPROCH Cancer is usually characterized as a heterogeneous disease with different types and subtypes. Cancer is not a single disease, but rather many related diseases that involve uncontrolled cellular growth and reproduction. It is one of the leading causes of death in the developed world and second in the developing world, almost a million people around the globe are losing the fight against cancer every year. The early diagnosis and prognosis of a cancer type have become inexorable in cancer research, as it can facilitate the quality of life of patients. Pancreas cancer is one of the most common malignant tumors with the highest morbidity and mortality, which is a considerable threat to people's health and life [1]. Advanced Pancreas cancer is likely to metastasize, which may lead to corresponding symptoms in patients with great pain, and are even life-threatening. This phenomenon is referred to as the distant metastasis of this cancer, which is represented by the M staging in the TNM staging system. Computed Tomography (CT) imaging is a widely used method for the evaluation of tumor prognosis. The previous system designed a clinical and radiomic features based feature extraction and Support Vector Machine (SVM) with Stochastic Gradient Descent (SGD) methods for diagnosis of distant metastasis of cancer. Feature Selection Via concave minimization (FSV) was used to select effective features.
Drawbacks
The use of the CT image is non enhanced; there is no enhancement of
the image or any obvious features. It has only less result of classification accuracy Efficient segmentation method is required to improve the prediction accuracy
Work 1 Proposed system
1. The CT images of patients is taken as an input
2. Pre-processing Adaptive median filtering is used to reduce the noises in these images 3. Segmentation Segmentation is performed by using modified Markov Random Field (MRF) model. 4. Feature extraction The following features are extracted A complete cancer tumor radiomic features set included its volume, texture Gabor and wavelet features. The patient's clinical characteristics were added as reference features, including the patient's age, smoking status, tumor classification, and T and N staging 5. Feature selection GLCM is used to select an efficient feature. It has a strong ability to find the most optimistic results for optimal design. 6. Classification
SVM with the SGD method is utilized for classification
Advantages
High classification accuracy
High precision and recall
Lower computational complexity
Work 2 Proposed:
1. CT images are taken as an input.
2. Adaptive median filtering is used to reduce the noises in these images. 3. After elimination of blurring and noise and increasing contrast, morphological operations are used for revealing details in enhancement operations. 4. After the morphological operations, region of interest is extracted based on the segmentation. Segmentation is performed by using Region Based Neural Networks 5. Edge detection using Improved canny edge detector Canny's aim was to discover the optimal edge detection algorithm which reduces the probability of detecting false edge, and gives sharp edges. 6. Feature extraction The following features are extracted A complete cancer tumor radiomic features set included its volume, texture Gabor and wavelet features. The patient's clinical characteristics were added as reference features, including the patient's age, smoking status, tumor classification, and T and N staging 7. Classification Enhanced Transductive Support Vector Machines (ETSVM) is utilized for classification
Advantages
High classification accuracy
High precision and recall Lower computational complexity