Segmentation-Based Classification Deep Learning Model for Breast Cancer Detection using Mammogram images DOI
Ankita Sinha,

Manjusha Pandey,

M. Nazma B. J. Naskar

et al.

Published: Dec. 1, 2023

Breast cancer has emerged as a leading cause of mortality, responsible for an extensive number deaths in recent years. The current imaging-based diagnostic methods adopted the detection breast cancer, such mammography, shown inadequate effectiveness clinical environments due to their tendency significant mistake rates. This paper introduces effective methodology that uses segmentation based on deep learning classifiers classification order perform automated, productive, and precise diagnosis cancer. In enhance best model combination, hybrid approach was employed, integrating models with VGG-19 models. performance proposed evaluated using several statistical metrics, accuracy, precision, recall, f1-score, receiver operating characteristics (ROC), along cross-entropy loss function. showed outstanding results compared other segmentation-based techniques. research concluded UNet method, improved classifier, had improvement 2.25% (Pre-Trained) model.

Language: Английский

BC-QNet: A Quantum-Infused ELM Model for Breast Cancer Diagnosis DOI
Anas Bilal, Azhar Imran, Xiaowen Liu

et al.

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 175, P. 108483 - 108483

Published: April 24, 2024

Language: Английский

Citations

25
Progressive Approaches in Oncological Diagnosis and Surveillance: Real‐Time Impedance‐Based Techniques and Advanced Algorithms DOI
Viswambari Devi Ramaswamy, Michael Keidar

Bioelectromagnetics, Journal Year: 2025, Volume and Issue: 46(1)

Published: Jan. 1, 2025

ABSTRACT Cancer remains a formidable global health challenge, necessitating the development of innovative diagnostic techniques capable early detection and differentiation tumor/cancerous cells from their healthy counterparts. This review focuses on confluence advanced computational algorithms with noninvasive, label‐free impedance‐based biophysical methodologies—techniques that assess biological processes directly without need for external markers or dyes. elucidates diverse array state‐of‐the‐art technologies, illuminating distinct electrical signatures inherent to cancer vs tissues. Additionally, study probes transformative potential these modalities in recalibrating personalized treatment paradigms. These offer real‐time insights into tumor dynamics, paving way precision‐guided therapeutic interventions. By emphasizing quest continuous vivo monitoring, herald pivotal advancement overarching endeavor combat globally.

Language: Английский

Citations

1
Revolutionizing breast ultrasound diagnostics with EfficientNet-B7 and Explainable AI DOI Creative Commons

M. Latha,

P. Santhosh Kumar,

R. Roopa Chandrika

et al.

BMC Medical Imaging, Journal Year: 2024, Volume and Issue: 24(1)

Published: Sept. 2, 2024

Breast cancer is a leading cause of mortality among women globally, necessitating precise classification breast ultrasound images for early diagnosis and treatment. Traditional methods using CNN architectures such as VGG, ResNet, DenseNet, though somewhat effective, often struggle with class imbalances subtle texture variations, to reduced accuracy minority classes malignant tumors. To address these issues, we propose methodology that leverages EfficientNet-B7, scalable architecture, combined advanced data augmentation techniques enhance representation improve model robustness. Our approach involves fine-tuning EfficientNet-B7 on the BUSI dataset, implementing RandomHorizontalFlip, RandomRotation, ColorJitter balance dataset The training process includes stopping prevent overfitting optimize performance metrics. Additionally, integrate Explainable AI (XAI) techniques, Grad-CAM, interpretability transparency model's predictions, providing visual quantitative insights into features regions influencing outcomes. achieves 99.14%, significantly outperforming existing CNN-based approaches in image classification. incorporation XAI enhances our understanding decision-making process, thereby increasing its reliability facilitating clinical adoption. This comprehensive framework offers robust interpretable tool detection cancer, advancing capabilities automated diagnostic systems supporting processes.

Language: Английский

Citations

7
Integration of ultrasound and mammogram for multimodal classification of breast cancer using hybrid residual neural network and machine learning DOI
Kushangi Atrey, Bikesh Kumar Singh, Narendra Kuber Bodhey

et al.

Image and Vision Computing, Journal Year: 2024, Volume and Issue: 145, P. 104987 - 104987

Published: March 12, 2024

Language: Английский

Citations

4
A Dirichlet Distribution-Based Complex Ensemble Approach for Breast Cancer Classification from Ultrasound Images with Transfer Learning and Multiphase Spaced Repetition Method DOI
Osman Güler

Deleted Journal, Journal Year: 2025, Volume and Issue: unknown

Published: April 29, 2025

Breast ultrasound is a useful and rapid diagnostic tool for the early detection of breast cancer. Artificial intelligence-supported computer-aided decision systems, which assist expert radiologists clinicians, provide reliable results. Deep learning methods techniques are widely used in field health diagnosis, abnormality detection, disease diagnosis. Therefore, this study, deep ensemble model based on Dirichlet distribution using pre-trained transfer models cancer classification from images proposed. In experiments were conducted Ultrasound Images Dataset (BUSI). The dataset, had an imbalanced class structure, was balanced data augmentation techniques. DenseNet201, InceptionV3, VGG16, ResNet152 with fivefold cross-validation. Statistical analyses, including ANOVA test Tukey HSD test, applied to evaluate model's performance ensure reliability Additionally, Grad-CAM (Gradient-weighted Class Activation Mapping) explainable AI (XAI), providing visual explanations decision-making process. spaced repetition method, commonly improve success learners educational sciences, adapted artificial intelligence study. results training as input further training, previously learned information. use method led increased reduced times. weights obtained trained into system different variations. proposed achieved 99.60% validation accuracy demonstrating its effectiveness classification.

Language: Английский

Citations

0
Breast Cancer Detection Redefined: Integrating Xception and EfficientNet-B5 for Superior Mammography Imaging DOI Creative Commons

N. Talukdar,

Amulya Kakati,

Upasana Barman

et al.

Deleted Journal, Journal Year: 2025, Volume and Issue: unknown, P. 100038 - 100038

Published: May 1, 2025

Language: Английский

Citations

0
Graph neural network-based breast cancer diagnosis using ultrasound images with optimized graph construction integrating the medically significant features DOI Creative Commons

Sadia Sultana Chowa,

Sami Azam, Sidratul Montaha

et al.

Journal of Cancer Research and Clinical Oncology, Journal Year: 2023, Volume and Issue: 149(20), P. 18039 - 18064

Published: Nov. 20, 2023

An automated computerized approach can aid radiologists in the early diagnosis of breast cancer. In this study, a novel method is proposed for classifying tumors into benign and malignant, based on ultrasound images through Graph Neural Network (GNN) model utilizing clinically significant features.Ten informative features are extracted from region interest (ROI), radiologists' markers. The significance evaluated using density plot T test statistical analysis method. A feature table generated where each row represents individual image, considered as node, edges between nodes denoted by calculating Spearman correlation coefficient. graph dataset fed GNN model. configured ablation study Bayesian optimization. optimized then with different thresholds getting highest performance shallow graph. consistency validated k-fold cross validation. impact ROIs handcrafted tumor classification comparing model's Histogram Oriented Gradients (HOG) descriptor entire image. Lastly, clustering-based performed to generate new filtered graph, considering weak strong relationships nodes, similarities.The results indicate that threshold value 0.95, achieves accuracy 99.48%, precision recall 100%, F1 score 99.28%, reducing number 85.5%. 86.91%, no HOG features. Different values Spearman's experimented compared. No differences observed previous graph.The might effective diagnosing learning pattern

Language: Английский

Citations

7
HBNet: an integrated approach for resolving class imbalance and global local feature fusion for accurate breast cancer classification DOI
Barsha Abhisheka, Saroj Kr. Biswas, Biswajit Purkayastha

et al.

Neural Computing and Applications, Journal Year: 2024, Volume and Issue: 36(15), P. 8455 - 8472

Published: Feb. 22, 2024

Language: Английский

Citations

2
Intelligent Breast Mass Classification Approach Using Archimedes Optimization Algorithm with Deep Learning on Digital Mammograms DOI Creative Commons
Mohammed Basheri

Biomimetics, Journal Year: 2023, Volume and Issue: 8(6), P. 463 - 463

Published: Oct. 1, 2023

Breast cancer (BC) has affected many women around the world. To accomplish classification and detection of BC, several computer-aided diagnosis (CAD) systems have been introduced for analysis mammogram images. This is because by human radiologist a complex time-consuming task. Although CAD are used to primarily analyze disease offer best therapy, it still essential enhance present integrating novel approaches technologies in order provide explicit performances. Presently, deep learning (DL) outperforming promising outcomes early BC creating executing convolutional neural networks (CNNs). article presents an Intelligent Mass Classification Approach using Archimedes Optimization Algorithm with Deep Learning (BMCA-AOADL) technique on Digital Mammograms. The major aim BMCA-AOADL exploit DL model bio-inspired algorithm breast mass classification. In approach, median filtering (MF)-based noise removal U-Net segmentation take place as pre-processing step. For feature extraction, utilizes SqueezeNet AOA hyperparameter tuning approach. detect classify mass, applies belief network (DBN) simulation value system studied MIAS dataset from Kaggle repository. experimental values showcase significant compared other algorithms maximum accuracy 96.48%.

Language: Английский

Citations

6
Machine Learning for Early Breast Cancer Detection DOI
N. A. Chowdhury, Lulu Wang, Linxia Gu

et al.

Journal of Engineering and Science in Medical Diagnostics and Therapy, Journal Year: 2024, Volume and Issue: 8(1)

Published: July 26, 2024

Abstract Globally, breast cancer (BC) remains a significant cause to female mortality. Early detection of BC plays an important role in reducing premature deaths. Various imaging techniques including ultrasound, mammogram, magnetic resonance imaging, histopathology, thermography, positron emission tomography, and microwave have been employed for obtaining images (BIs). This review provides comprehensive information different modalities publicly accessible BI sources. The advanced machine learning (ML) offer promising avenue replace human involvement detecting cancerous cells from BIs. article outlines various ML algorithms (MLAs) which extensively used identifying BIs at the early stages, categorizing them based on presence or absence malignancy. Additionally, addresses current challenges associated with application MLAs identification proposes potential solutions.

Language: Английский

Citations

2