Supervising Healthcare Schemes Using Machine Learning in Breast Cancer and Internet of Things (SHSMLIoT)

Lamba, Monika; Munjal, Geetika; Gigras, Yogita

doi:10.1002/9781119792468.ch11

Cited by 3 publications

(1 citation statement)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Histological Grade (HG) is a well-described prognostic factor that reflects the morphological characteristics of the tumour as well as the clinical behaviour of the disease. Diagnosticians have long recognised this morphological multiplicity, which is replicated in histological grades with dissimilar microscopic appearances and correlated with medical outcomes [3] [4]. HG, which stands for the morphologic assessment of tumour genetic traits, is a well-established prognostic factor that has been successful in generating significant evidence regarding the clinical behaviour of the disease [5] [6].…”

Section: Introductionmentioning

confidence: 99%

Two-Stage Gene Selection Technique For Identifying Significant Prognosis Biomarkers In Breast Cancer

Lamba,

Munjal,

Gigras

2024

IJCDS

View full text Add to dashboard Cite

One crucial stage in the data preparation procedure for breast cancer classification involves extracting a selection of meaningful genes from microarray gene expression data. This stage is crucial because it discovers genes whose expression patterns can differentiate between different types or stages of breast cancer. Two highly effective algorithms, CONSISTENCY-BFS and CFS-BFS, have been developed for gene selection. These algorithms are designed to identify the genes that are most crucial in distinguishing between different types and stages of breast cancer by analysing large volumes of genetic data. A noteworthy advancement is a refined 2-Stage Gene Selection technique specifically designed for predicting subtypes in breast cancer. The initial phase of the 2-Stage Gene Selection (GeS) approach relies on the CFS-BFS algorithm, which plays a crucial role in effectively eliminating unnecessary, distracting, and redundant genes. The initial filtering process plays a crucial role in simplifying the dataset and identifying the genes that have the highest potential to shed light on the category of breast cancer. The CONSISTENCY-BFS algorithm guarantees that only the most pertinent genes are retained by further refining the gene selection process. This stage is essential for eliminating any remaining uncertainty and enhancing the overall efficiency of the algorithm. This innovative approach represents a significant advancement in the field of bioinformatics as it offers a more accurate and targeted method for selecting genes based on their relevance to breast cancer classification. When the 2-Stage GeS is constructed using Hidden Weight Naive Bayes, remarkably, it yields more precise and dependable outcomes. The indicators that demonstrate positive outcomes encompass recollection, accuracy, f-score, and fallout rankings. The Kaplan-Meier Survival Model was employed to further validate the top four genes, namely E2F3, PSMC3IP, GINS1, and PLAGL2. Presumably, precision therapy will specifically focus on targeting the genes E2F3 and GINS1.

show abstract