Vol. 26 Issue 1 2023
  1. Divya Daniel1, C. Shoba Bindu2, P. Dileep Kumar Reddy3

1Department of CSE, JNTUA College of Engineering, Ananthapuramu, Andhra Pradesh, India

2Department of CSE, JNTUA College of Engineering, Ananthapuramu, Andhra Pradesh, India

3Department of CSE, Narsimha Reddy Engineering College (Autonomous), Secunderabad, Telangana, India,


All cancers stem from underlying cellular DNA abnormalities and might appear at any time in a person’s life. Because of the wide range of resulting genetic and phenotypic variations among afflicted individuals, the search for viable therapeutics is time-consuming and resource- intensive. Due to the rarity of samples and the abundance of input parameters, cancer datasets are notoriously difficult to work with when attempting to construct reliable predictors for classifying patients into risk groups. This article discusses four different types of cancer, lung cancer, breast cancer, pancreatic cancer and kidney cancer. To better understand how to categorize cancer patients into those at high and low risk, this research employs machine learning and deep learning algorithms for prediction, relying on a mix of supervised, unsupervised and self-supervised learning methodologies. This study’s findings add to the growing body of evidence supporting the use of integrated learning algorithm and the combination of genetic and clinical data in the support of Clinical oncology decision-making.

INDEX TERMS: Cancer, Pancreatic cancer, renal cancer and machine learning or deep learning algorithms.