Computer-Aided Diagnosis System for Early Detection of Malignant Tissues

Cancer is the most general term collectively used for a variety of diseases that can affect any part of the body. Malignant tumors and neoplasms are other words that are used for cancer. Cancer is the quick development of aberrant cells that proliferate outside of their normal borders and have the potential to move to other organs and infiltrate nearby bodily parts and is the leading cause of death worldwide. Medical experts may use Computer-Aided Diagnosis (CAD) as a second opinion and to save their valuable time from manual analysis. Practitioners use various modalities such as X-ray, computed tomography (CT), and magnetic resonance imaging (MRI). X-ray images are helpful in early diagnosis but for effective diagnosis and to pinpoint the malignant issues, MRI and CT images are preferred. Even though MR images are fine-grained and detailed images of organs, during the acquisition process some rician and gaussian noise may creep up. It is necessary for a diagnostic system to denoise the images before segmentation and classification to attain better accuracy for the models. This paper aims to provide insights into the machine learning algorithms used by various CAD systems, which aim to detect and segment various types of cancers such as prostate, cervix, pancreas, colon, and kidney.