Automated lung tumor detection and diagnosis in CT Scans using texture feature analysis and SVM

Abstract

CT scans are an important tool in the diagnosis of lung tumors in medicine. This work presents an automated system for lung tumor diagnosis on CT scans. Scans are automatically segmented using marker-based watershed transformation, which successfully segments hardly separable, lung wall adjunct tumors. The scans are further analyzed in a sliding window approach using Haralick features and a Support Vector Machine classifier to detect and classify benign and malignant tumors. This novel approach for classification was tested using the LUNGx Challenge dataset [1] and achieved exceptional results while utilizing a minimal training set. Communication

Particle Swarm Optimization and Texture Analysis Image Processing Techniques using MRI Images to Detect Brain Tumor in Human

“…The second part is performing the texture analysis, for which we will extract the segmented image and using those collected features the classifiers are used to make the classify the tumor. [22] Primarily the aim is to analyze every MRI image and after processing, the machine will provide the decision by saying whether the input image is healthy brain or unhealthy brain. The block of the process is required, the MIR as an entry, preprocess image used by FCM, to make segmentation, collect the features from it and finally a classification of the MRI images as shown in the figure 1.…”

“…The block of the process is required, the MIR as an entry, preprocess image used by FCM, to make segmentation, collect the features from it and finally a classification of the MRI images as shown in the figure 1. These can be a roadblock for further processing and segmentation because they represent separate components that are also recognized as such by the connected component algorithm, even though they are not part of the tissue that needs to be examined so they leads to inefficient detection which can prove fatal for someone's life [22]. These artifacts are removed from the image background in a first preprocessing step by performing the median filtering operation on the MRI image.…”

“…The next step for biomedical image analysis after preprocessing and artifacts removal we will do segmentation which will breaks down image into multiple segments and allow easy analysis and helps in extracting meaning full information [22] [1]. Previously, Clustering is used as a method for segmentation which was comparatively better as compared to other methods like Otsu, thresholding [5] [10]and other but then it was seen that there were certain disadvantages of this method like sensitive to noise, computationally expensive and determination of membership function was not easy too [25].…”

Robust Image Forgery Detection Using Point Feature Analysis

“…3) SVM Classifier: Support Vector Machine (SVM) is an efficient and optimal classifier commonly used with machine learning systems, and neural networks [28], [2]. In our system we only have two classes original and fake.…”

Instance Segmentation Model Created from Three Semantic Segmentations of Mask, Boundary and Centroid Pixels Verified on GlaS Dataset

“…S SEMANTIC segmentation is the most important computer vision task in biomedical applications and any improvement of it may result in saved lives [1], [2]. Combining multiple models is a well known technique to improve segmentation.…”