Abstract: Medical imaging has become a vital part in the early detection, diagnosis and treatment of cancer. The reason: the remarkable power of medical imaging in providing physicians with sight and insight about human disease and physiology. In this role, imaging is used to: Screen, diagnose, and stage cancer; Guide cancer treatments; Determine if a treatment is working; Monitor cancer recurrence; and Facilitate medical research, particularly in such critical areas as drug discovery and therapeutic innovation to improve patient care. The cancer (lung, breast, or brain) biological database which contains the medical images classifies the digital films into three categories: normal, benign and malignant. The normal ones are those characterizing a healthy patient (non-nodules); nodules can be either benign (non-cancerous) or malignant (cancer). This paper presents a comprehensive study of various medical image processing and classification techniques used for predicting various types of cancer in early stage.

Keywords: Computer Aided Diagnosis Systems, Image Mining, Image Recognition, Neural Networks, Classification, Image matching, Image retrieval.