Pathway toDetect Cancer Tumor byGenetic Mutation

Title

Pathway toDetect Cancer Tumor byGenetic Mutation

Creator

Mohanty A.; Prusty A.R.; Dasig D.

Description

Cancer detection is one of the challenging tasks due to the unavailability of proper medical facilities. The survival of cancer patients depends upon early detection and medication. The main cause of the disease is due to several genetic mutations which form cancer tumors. Identification of genetic mutation is a time-consuming task. This creates a lot of difficulties for the molecular pathologist. A molecular pathologist selects a list of gene variations to analyze manually. The clinical evidence strips belong to nine classes, but the classification principle is still unknown. This implementation proposes a multi-class classifier to classify genetic mutations based on clinical evidence. Natural language processing analyzes the clinical text of evidence of gene mutations. Machine learning algorithms like K-nearest neighbor, linear support vector machine, and stacking models are applied to the collected text dataset, which contains information about the genetic mutations and other clinical pieces of evidence that pathology uses to classify the gene mutations. In this implementation, nine genetic variations have been taken, considered a multi-class classification problem. Here, each data point is classified among the nine classes of gene mutation. The performance of the machine learning models is analyzed on the gene, variance, and text features. The gene, variance, and text features are analyzed individually with univariate analysis. Then K-nearest neighbor, linear support vector machine, and stacking model are applied to the combined features of a gene, variance, and text. In the experiment, support vector machine gives better results as compared to other models because this model provides fewer misclassification points. Based on the variants of gene mutation, the risk of cancer can be detected, and medications can be given. This chapter will motivate the readers, researchers, and scholars of this field for future investigations. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

Source

Studies in Computational Intelligence, Vol-1132, pp. 171-187.

Date

2024-01-01

Publisher

Springer Science and Business Media Deutschland GmbH

Coverage

Mohanty A., Computer Science and Engineering, CHRIST (Deemed to be) University, Bengaluru, India; Prusty A.R., DGT, RDSDE, NSTI(W), West Bengal, Kolkata, India; Dasig D., Graduate Studies College of Science and Computer Studies, De La Salle University, Dasmarinas, Philippines

Rights

Restricted Access

Relation

ISSN: 1860949X

Format

Online

Language

English

Type

Book chapter

Identifier

https://doi.org/10.1007/978-981-99-8853-2_11

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85186487454&doi=10.1007%2f978-981-99-8853-2_11&partnerID=40&md5=79ad8883838ab7d8b8fc38107f3d6d5f