Cargando…

DenseNet_ HybWWoA: A DenseNet-Based Brain Metastasis Classification with a Hybrid Metaheuristic Feature Selection Strategy

Brain metastases (BM) are the most severe consequence of malignancy in the brain, resulting in substantial illness and death. The most common primary tumors that progress to BM are lung, breast, and melanoma. Historically, BM patients had poor clinical outcomes, with limited treatment options includ...

Descripción completa

Detalles Bibliográficos
Autor principal: Alshammari, Abdulaziz
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10216652/
https://www.ncbi.nlm.nih.gov/pubmed/37239025
http://dx.doi.org/10.3390/biomedicines11051354
_version_ 1785048350027939840
author Alshammari, Abdulaziz
author_facet Alshammari, Abdulaziz
author_sort Alshammari, Abdulaziz
collection PubMed
description Brain metastases (BM) are the most severe consequence of malignancy in the brain, resulting in substantial illness and death. The most common primary tumors that progress to BM are lung, breast, and melanoma. Historically, BM patients had poor clinical outcomes, with limited treatment options including surgery, stereotactic radiation therapy (SRS), whole brain radiation therapy (WBRT), systemic therapy, and symptom control alone. Magnetic Resonance Imaging (MRI) is a valuable tool for detecting cerebral tumors, though it is not infallible, as cerebral matter is interchangeable. This study offers a novel method for categorizing differing brain tumors in this context. This research additionally presents a combination of optimization algorithms called the Hybrid Whale and Water Waves Optimization Algorithm (HybWWoA), which is used to identify features by reducing the size of recovered features. This algorithm combines whale optimization and water waves optimization. The categorization procedure is consequently carried out using a DenseNet algorithm. The suggested cancer categorization method is evaluated on a number of factors, including precision, specificity, and sensitivity. The final assessment findings showed that the suggested approach exceeded the authors’ expectations, with an F1-score of 97% and accuracy, precision, memory, and recollection of 92.1%, 98.5%, and 92.1%, respectively.
format Online
Article
Text
id pubmed-10216652
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-102166522023-05-27 DenseNet_ HybWWoA: A DenseNet-Based Brain Metastasis Classification with a Hybrid Metaheuristic Feature Selection Strategy Alshammari, Abdulaziz Biomedicines Article Brain metastases (BM) are the most severe consequence of malignancy in the brain, resulting in substantial illness and death. The most common primary tumors that progress to BM are lung, breast, and melanoma. Historically, BM patients had poor clinical outcomes, with limited treatment options including surgery, stereotactic radiation therapy (SRS), whole brain radiation therapy (WBRT), systemic therapy, and symptom control alone. Magnetic Resonance Imaging (MRI) is a valuable tool for detecting cerebral tumors, though it is not infallible, as cerebral matter is interchangeable. This study offers a novel method for categorizing differing brain tumors in this context. This research additionally presents a combination of optimization algorithms called the Hybrid Whale and Water Waves Optimization Algorithm (HybWWoA), which is used to identify features by reducing the size of recovered features. This algorithm combines whale optimization and water waves optimization. The categorization procedure is consequently carried out using a DenseNet algorithm. The suggested cancer categorization method is evaluated on a number of factors, including precision, specificity, and sensitivity. The final assessment findings showed that the suggested approach exceeded the authors’ expectations, with an F1-score of 97% and accuracy, precision, memory, and recollection of 92.1%, 98.5%, and 92.1%, respectively. MDPI 2023-05-04 /pmc/articles/PMC10216652/ /pubmed/37239025 http://dx.doi.org/10.3390/biomedicines11051354 Text en © 2023 by the author. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Alshammari, Abdulaziz
DenseNet_ HybWWoA: A DenseNet-Based Brain Metastasis Classification with a Hybrid Metaheuristic Feature Selection Strategy
title DenseNet_ HybWWoA: A DenseNet-Based Brain Metastasis Classification with a Hybrid Metaheuristic Feature Selection Strategy
title_full DenseNet_ HybWWoA: A DenseNet-Based Brain Metastasis Classification with a Hybrid Metaheuristic Feature Selection Strategy
title_fullStr DenseNet_ HybWWoA: A DenseNet-Based Brain Metastasis Classification with a Hybrid Metaheuristic Feature Selection Strategy
title_full_unstemmed DenseNet_ HybWWoA: A DenseNet-Based Brain Metastasis Classification with a Hybrid Metaheuristic Feature Selection Strategy
title_short DenseNet_ HybWWoA: A DenseNet-Based Brain Metastasis Classification with a Hybrid Metaheuristic Feature Selection Strategy
title_sort densenet_ hybwwoa: a densenet-based brain metastasis classification with a hybrid metaheuristic feature selection strategy
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10216652/
https://www.ncbi.nlm.nih.gov/pubmed/37239025
http://dx.doi.org/10.3390/biomedicines11051354
work_keys_str_mv AT alshammariabdulaziz densenethybwwoaadensenetbasedbrainmetastasisclassificationwithahybridmetaheuristicfeatureselectionstrategy