1. A neural network framework for predicting Adenocarcinoma cancer using high-throughput gene expression data.
- Author
-
Mohanty, Subhra, Nayak, Debasish Swapnesh Kumar, and Swarnkar, Tripti
- Subjects
- *
GENE expression , *MACHINE learning , *PREIMPLANTATION genetic diagnosis , *ARTIFICIAL intelligence , *SOMATIC mutation , *ADENOCARCINOMA - Abstract
Machine learning is a part of artificial intelligence and its optimization techniques are allows system to learn from past example and identify pattern from large dataset. This criteria are well suited for researchers for doing their research on medical application, genomic measurements and it cause to machine learning techniques are used in cancer diagnosis and detection. Recent year machine learning has been applied to cancer prediction and prognosis. This latter approach growing trend towards an artificial neural network. ANN ensemble is a learning platform in which a number of neurons work together to solve a problem. Our approach is used to identify adenocarcinoma cancer diseases from gene expression data samples obtained from bodies that are to be diagnosed in this research. Last decades Next-generation sequencing (NGS) technology has been expanded with advantages in the reliability, data interpretation and costs, sequencing chemistry. Such large improvements can make NGS clinically practical, and NGS technology is presently being developed for use in the field of oncolo-gy. Various clinical applications are discussed, including DNA-sequencing-based mutation detection in inherited cancer syndromes, RNA-sequencing-based detection of spliceogenic variants, DNA-sequencing-based risk modi-fiers and application for pre-implantation genetic diagnosis, and cancer somatic mutation analysis. In this study, neural network classifiers were used to determine patient carcinoma cancer status. Applied neural network architecture is based on a two-level architecture, with each individual network having two outputs based on the normal cell sample or cancer cell sample according to our dataset. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF