Enhanced Brain Tumor Segmentation and Size Estimation in MRI Samples using Hybrid Optimization DOI
Ayesha Agrawal, Vinod Maan

Data & Metadata, Journal Year: 2023, Volume and Issue: 2, P. 408 - 408

Published: Dec. 26, 2023

The area of medical imaging specialization, specifically in the context brain tumor segmentation, has long been challenged by inherent complexity and variability structures. Traditional segmentation methods often struggle to accurately differentiate between diverse types tissues within brain, such as white matter, grey cerebrospinal fluid, leading suboptimal results identification delineation. These limitations necessitate development more advanced precise techniques enhance diagnostic accuracy treatment planning. In response these challenges, proposed study introduces a novel approach that combines Grey Wolf Optimization Cuckoo Search Fuzzy C-Means (FCM) framework. integration GWO CS is designed leverage their respective strengths optimizing tissues. This hybrid was rigorously tested across multiple Magnetic Resonance Imaging (MRI) datasets, demonstrating significant enhancements over existing methods. observed 4,9 % improvement accuracy, 3,5 increase precision, 4,5 higher recall, 3,2 less delay, 2,5 better specificity segmentation. implications advancements are profound. By achieving precision method can substantially aid early diagnosis accurate staging tumors, eventually effective planning improved patient outcomes. Furthermore, FCM process sets new benchmark imaging, paving way for future investigation field

Language: Английский

SPBTGNS: Design of an Efficient Model for Survival Prediction in Brain Tumour Patients using Generative Adversarial Network with Neural Architectural Search Operations DOI Creative Commons

Ruqsar Zaitoon,

Sachi Nandan Mohanty, Deepthi Godavarthi

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 140847 - 140869

Published: Jan. 1, 2024

The landscape of medical imaging, particularly in brain tumor analysis and survival prediction, necessitates advancements due to the inherent complexities life-threatening nature tumors. Existing methodologies often struggle with precision efficiency, predominantly limitations handling diverse intricate image datasets. This research presents a novel approach that aims improve accuracy prediction patients tumours, leveraging Generative Adversarial Network (GAN) integrated Neural Architectural Search (NAS) operations. model employs Adaptive Computation Time (ACT) Transformer, method crucial for dynamically adjusting number transformer layers based on complexity input sets. feature is beneficial imaging adapting varying data samples. integration Squeeze-and-Excitation Networks (SENet) enables recalibrate features channel-wise, significantly enhancing sensitivity pivotal MRI images. Furthermore, application Google's AutoML Vision Edge offers efficient neural architecture hyperparameter optimization, specifically tuned Efficient Architecture (ENAS) utilized discover high-performance models lower computational demands, critical aspect where resource constraints are common different use cases. also incorporates customized loss functions, Weighted Cross-Entropy Loss, addressing class imbalance datasets by emphasizing rarer types. Spatial Dropout Batch Normalization as regularization techniques generalization reduce overfitting risks. model's efficacy was validated Br35H, Kaggle Brain Tumor Dataset, IEEE Data Port Dataset Databases, exhibiting notable improvement over existing methods: 5.9% better precision, 6.5% higher accuracy, 4.9% recall analysis. In analysis, demonstrated 8.5% 8.3% among other improvements. These enhancements underscore capability providing more accurate, efficient, reliable predictions patients, potentially revolutionizing diagnosis prognostication clinical settings.

Language: Английский

Citations

6
Advancements in deep learning techniques for brain tumor segmentation: A survey DOI Creative Commons

C. Umarani,

Shantappa G. Gollagi, Shridhar Allagi

et al.

Informatics in Medicine Unlocked, Journal Year: 2024, Volume and Issue: 50, P. 101576 - 101576

Published: Jan. 1, 2024

Citations

4
Brain Tumor Segmentation and Classification using MRI: Modified Segnet Model and Hybrid Deep Learning Architecture with Improved Texture Features DOI

Palleti Venkata Kusuma,

S. Chandra Mohan Reddy

Computational Biology and Chemistry, Journal Year: 2025, Volume and Issue: 117, P. 108381 - 108381

Published: Feb. 18, 2025

Language: Английский

Citations

0
Accurate brain tumor region segmentation using local intensity deviation based ResNet50 DOI

K. V. V. Kumar,

D. Mabuni

Journal of Ambient Intelligence and Humanized Computing, Journal Year: 2025, Volume and Issue: unknown

Published: May 2, 2025

Language: Английский

Citations

0
Intelligent Multi-Grade Brain Tumor Identification in MRI: A Metaheuristic-Based Uncertain Set Framework DOI
Saravanan Alagarsamy, Vishnuvarthanan Govindaraj,

A. Shahina

et al.

IEEE Transactions on Artificial Intelligence, Journal Year: 2024, Volume and Issue: 5(11), P. 5381 - 5391

Published: Aug. 12, 2024

Language: Английский

Citations

2
Brain Tumor Diagnosis Using Deep Learning: A Literature Review DOI

S. Bhuvaneswari,

Joel G. Thomas,

S Nithish

et al.

Published: Feb. 22, 2024

Brain tumor diagnosis is a critical task in the field of medical imaging, with potential to significantly impact patient outcomes and treatment planning. The use deep learning has been more well-known within last ten years as method for improving automating detection categorization brain tumors. goal this literature review present thorough overview application

Language: Английский

Citations

1
Brain tumor segmentation and classification using transfer learning based CNN model with model agnostic concept interpretation DOI
A. Maria Nancy,

R. Maheswari

Multimedia Tools and Applications, Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 18, 2024

Language: Английский

Citations

1
Brain Tumor Classification Using Pretained Deep Convolutional Neural Networks DOI Open Access

M. Meena,

U. Balaswetha,

M. Harini

et al.

International Journal of Health Sciences and Research, Journal Year: 2024, Volume and Issue: 14(5), P. 176 - 180

Published: May 7, 2024

Due to their complexity and sensitivity, classifying brain diseases is a very difficult task. Because tumors are serious sometimes fatal, early detection diagnosis essential for developing an efficient treatment plan. A vital medical imaging tool, magnetic resonance (MRI) allows the detailed, non-invasive visualization of internal structures brain. When it comes diagnosing treating tumors, plays critical role. Starting with dataset preprocessing, method applies MRI scans clinical data from people different conditions, including cases non-tumors. Training testing sets make up dataset. tumor requires number processes, feature extraction, classification, image post-processing. For images, system makes use Convolutional Neural Networks Long Short-Term memory (LSTM) pre-trained model using approach transfer learning. The proposed framework not only uses improve performance training better but also thresholding accuracy augmentation increasing images in Preliminary outcome shows that family models Hybrid algorithm performs than previous CNN architectures because scale all dimensions depth, width, resolution constant ratio compound coefficient. results demonstrated by scaling baseline architecture able capture complicated features thus overall improved. Key words: Brain convolutional neural network, imaging, deep learning,

Language: Английский

Citations

0
Enhanced Brain Tumor Segmentation and Size Estimation in MRI Samples using Hybrid Optimization DOI
Ayesha Agrawal, Vinod Maan

Data & Metadata, Journal Year: 2023, Volume and Issue: 2, P. 408 - 408

Published: Dec. 26, 2023

The area of medical imaging specialization, specifically in the context brain tumor segmentation, has long been challenged by inherent complexity and variability structures. Traditional segmentation methods often struggle to accurately differentiate between diverse types tissues within brain, such as white matter, grey cerebrospinal fluid, leading suboptimal results identification delineation. These limitations necessitate development more advanced precise techniques enhance diagnostic accuracy treatment planning. In response these challenges, proposed study introduces a novel approach that combines Grey Wolf Optimization Cuckoo Search Fuzzy C-Means (FCM) framework. integration GWO CS is designed leverage their respective strengths optimizing tissues. This hybrid was rigorously tested across multiple Magnetic Resonance Imaging (MRI) datasets, demonstrating significant enhancements over existing methods. observed 4,9 % improvement accuracy, 3,5 increase precision, 4,5 higher recall, 3,2 less delay, 2,5 better specificity segmentation. implications advancements are profound. By achieving precision method can substantially aid early diagnosis accurate staging tumors, eventually effective planning improved patient outcomes. Furthermore, FCM process sets new benchmark imaging, paving way for future investigation field

Language: Английский

Citations

0