Classification of Liver Cancer Diseases using Image Processing technique

Abstract

The early detection of liver cancer saves the time of a doctor and life of a patient. The proposed system collects Microscopic images as input from the patients and preprocesses them to extract features. Once the feature extraction stage is completed the classification of the image need to be done on them. The proposed system uses the classifier support vector machine (SVM) technique to classify the images into their respective classes. The classifier in the proposed system uses the normal approach of classification i.e. a classifier has normally two stages one is training and then testing. Each of these classifiers goes through both these stages. Firstly, the training stage involves the system learning on the images and their respective category which is already known from the expert advice. In this way a series of images are given in the form of an input with their actual category. The classifier learns from this and then in the testing phase a new image is given for classification to the system. The system uses the prior knowledge which it has learnt during the training phase to predict the category for the image. The image is classified into diseased or not. If the Image is diseased then it is classified into type of liver cancer. The classification of type of liver cancer saves the life of a patient and time of a doctor.

Note: Funding Statement: No funding applied.

Declaration of Interests: No competing interest.

Keywords: Microscopic Images, Support Vector Machine (SVM), Feature Extraction, Training, Testing