Colitis is an inflammation of the colon's inner lining that is commonly brought on by an infection or an allergic reaction. The patient's general health may be significantly impacted by how severe this illness is. Although there is no known cure for this illness, deep learning-based early disease diagnosis and severity grading can stop the disease from progressing to more dangerous phases. In this paper, we offer a technique for classifying colon illnesses based on SVM and convolutional neural networks. When used to diagnose four different types of colitis illnesses, the model demonstrated good and strong outcomes when compared to earlier studies. With the aid of artificial intelligence, this proposed method can facilitate the provision of quick medical care that will hasten diagnosis times. Due to the availability of these technologies in all fields, particularly in the classification of medicine, a variety of algorithms with varying performances using artificial intelligence and deep learning techniques have therefore emerged. These algorithms serve an important and effective role in the ability to detect and diagnose. Images of colon disorders So, the main objective of the study is to keep learning about the models whose function it is to solve the classification problem. The suggested model's classification accuracy was 92.31% and it effectively categorized nine different types’ colitis diseases using CNN and SVM approach. Images that don't fall into any of the four categories can be found using the suggested model. The term "unknown images" refers to these pictures. This research aims to develop a computer aided diagnostic system that can classify colitis. In this article, we provide support vector machine (SVM) classifiers for patient-level colitis diagnosis and deep convolutional neural networks approaches for lesion-level colitis identification on routine abdominal CT scans.