MULTforAD: Multimodal MRI Neuroimaging for Alzheimer’s Disease Detection Based on a 3D Convolution Model DOI Open Access
Walaa N. Ismail,

Fathimathul Rajeena P. P.,

Mona A. S. Ali

et al.

Electronics, Journal Year: 2022, Volume and Issue: 11(23), P. 3893 - 3893

Published: Nov. 24, 2022

Alzheimer’s disease (AD) is a neurological that affects numerous people. The condition causes brain atrophy, which leads to memory loss, cognitive impairment, and death. In its early stages, tricky predict. Therefore, treatment provided at an stage of AD more effective less damage than later stage. Although common condition, it difficult recognize, classification requires discriminative feature representation separate similar patterns. Multimodal neuroimage information combines multiple medical images can classify diagnose accurately comprehensively. Magnetic resonance imaging (MRI) has been used for decades assist physicians in diagnosing disease. Deep models have detected with high accuracy computing-assisted diagnosis by minimizing the need hand-crafted extraction from MRI images. This study proposes multimodal image fusion method fuse neuroimages modular set preprocessing procedures automatically convert neuroimaging initiative (ADNI) into BIDS standard classifying different data subjects normal controls. Furthermore, 3D convolutional neural network learn generic features capturing AlD biomarkers fused images, resulting richer information. Finally, conventional CNN three classifiers, including Softmax, SVM, RF, forecasts classifies extracted traits healthy brain. findings reveal proposed efficiently predict progression combining high-dimensional characteristics public sources range 88.7% 99% outperforming baseline when applied MRI-derived voxel features.

Language: Английский

AML‐Net: Attention‐based multi‐scale lightweight model for brain tumour segmentation in internet of medical things DOI Creative Commons
Muhammad Zeeshan Aslam, Basit Raza, Muhammad Faheem

et al.

CAAI Transactions on Intelligence Technology, Journal Year: 2024, Volume and Issue: unknown

Published: Jan. 17, 2024

Abstract Brain tumour segmentation employing MRI images is important for disease diagnosis, monitoring, and treatment planning. Till now, many encoder‐decoder architectures have been developed this purpose, with U‐Net being the most extensively utilised. However, these require a lot of parameters to train semantic gap. Some work tried make lightweight model do channel pruning that made small receptive field which compromised accuracy. The authors propose an attention‐based multi‐scale called AML‐Net in Internet Medical Things overcome above issues. This consists three are trained different scale input along previously learned features diminish loss. Moreover, designed attention module replaced traditional skip connection. For module, six experiments were conducted, from dilated convolution spatial performed well. has convolutions relatively large followed by extract global context encoder low‐level features. Then fine combined decoder's same layer high‐level perform experiment on low‐grade‐glioma dataset provided Cancer Genome Atlas at least Fluid‐Attenuated Inversion Recovery modality. proposed 1/43.4, 1/30.3, 1/28.5, 1/20.2 1/16.7 fewer than Z‐Net, U‐Net, Double BCDU‐Net CU‐Net respectively. authors’ gives results IoU = 0.834, F 1‐score 0.909 sensitivity 0.939, greater CU‐Net, RCA‐IUnet PMED‐Net.

Language: Английский

Citations

9
Arouse-Net: Enhancing Glioblastoma Segmentation in Multi-Parametric MRI with a Custom 3D Convolutional Neural Network and Attention Mechanism DOI Creative Commons
Haiyang Li, Xiaozhi Qi, Ying Hu

et al.

Mathematics, Journal Year: 2025, Volume and Issue: 13(1), P. 160 - 160

Published: Jan. 4, 2025

Glioblastoma, a highly aggressive brain tumor, is challenging to diagnose and treat due its variable appearance invasiveness. Traditional segmentation methods are often limited by inter-observer variability the lack of annotated datasets. Addressing these challenges, this study introduces Arouse-Net, 3D convolutional neural network that enhances feature extraction through dilated convolutions, improving tumor margin delineation. Our approach includes an attention mechanism focus on edge features, essential for precise glioblastoma segmentation. The model’s performance benchmarked against state-of-the-art BRATS test dataset, demonstrating superior results with over eight times faster processing speed. integration multi-modal MRI data novel evaluation protocol developed offer robust framework medical image segmentation, particularly useful clinical scenarios where datasets limited. findings research not only advance field analysis but also provide foundation future work in development automated tools tumors.

Language: Английский

Citations

1
multiPI-TransBTS: A multi-path learning framework for brain tumor image segmentation based on multi-physical information DOI
Hongjun Zhu,

Jeffrey Huang,

Kuo Chen

et al.

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 191, P. 110148 - 110148

Published: April 10, 2025

Language: Английский

Citations

1
DenseUNet+: A novel hybrid segmentation approach based on multi-modality images for brain tumor segmentation DOI Creative Commons
Halit ÇETİNER, Sedat Metlek

Journal of King Saud University - Computer and Information Sciences, Journal Year: 2023, Volume and Issue: 35(8), P. 101663 - 101663

Published: July 23, 2023

Segmentation of brain tumors is great importance for patients in clinical diagnosis and treatment. For this reason, experts try to identify border regions special using multimodal images from magnetic resonance imaging systems. In some images, may be intertwined. As a result, situation leads make incomplete or wrong decisions. This paper presents DenseUNet+, new deep learning-based approach perform segmentation with high accuracy images. the DenseUNet+ model, data four different modalities were used together dense block structures. Afterward, linear operations applied these then concatenate operation was performed. The results obtained way transferred decoder layer. proposed method also compared state-of-the-art (SOTA) studies same dataset by dice jaccard metrics BraTS2021 FeTS2021 datasets. result comparison, evaluation 95% 88%, respectively, 86% 87% performance values FeTS2021, respectively. It has been determined that are better than many SOTA tumor methods.

Language: Английский

Citations

21
GMetaNet: Multi-scale ghost convolutional neural network with auxiliary MetaFormer decoding path for brain tumor segmentation DOI
Yao Lu,

Yankang Chang,

Zhouzhou Zheng

et al.

Biomedical Signal Processing and Control, Journal Year: 2023, Volume and Issue: 83, P. 104694 - 104694

Published: Feb. 20, 2023

Language: Английский

Citations

20
Residual attention based uncertainty-guided mean teacher model for semi-supervised breast masses segmentation in 2D ultrasonography DOI
Muhammad Umar Farooq, Zahid Ullah, Jeonghwan Gwak

et al.

Computerized Medical Imaging and Graphics, Journal Year: 2023, Volume and Issue: 104, P. 102173 - 102173

Published: Jan. 9, 2023

Language: Английский

Citations

19
MDFF-Net: A multi-dimensional feature fusion network for breast histopathology image classification DOI Open Access
Cheng Xu, Ke Yi, Nan Jiang

et al.

Computers in Biology and Medicine, Journal Year: 2023, Volume and Issue: 165, P. 107385 - 107385

Published: Aug. 16, 2023

Language: Английский

Citations

18
Brain tumor segmentation with missing MRI modalities using edge aware discriminative feature fusion based transformer U-net DOI
B. Jagadeesh, G. Anand Kumar

Applied Soft Computing, Journal Year: 2024, Volume and Issue: 161, P. 111709 - 111709

Published: May 10, 2024

Language: Английский

Citations

8
Enhancing brain tumor segmentation in MRI images: A hybrid approach using UNet, attention mechanisms, and transformers DOI Creative Commons
Tat-Bao-Thien Nguyen,

Thien-Qua T. Nguyen,

Hieu-Nghia Nguyen

et al.

Egyptian Informatics Journal, Journal Year: 2024, Volume and Issue: 27, P. 100528 - 100528

Published: Aug. 31, 2024

Accurate brain tumor segmentation in MRI images is crucial for effective treatment planning and monitoring. Traditional methods often encounter challenges due to the complexity variability of shapes textures. Consequently, there a growing need automated solutions assist healthcare professionals tasks, improving efficiency reducing workload. This study introduces an innovative method accurately segmenting tumors by employing refined 3D UNet model integrated with Transformer. The goal leverage self-attention mechanisms enhance capabilities. proposed combines Contextual Transformer (CoT) Double Attention (DA) architectures. CoT extended format baseline exploit intricate contextual details images. DA blocks skip connections aggregate distribute long-range features, emphasizing inter-dependencies within expanded spatial scope. Experimental results demonstrate superior performance compared current state-of-the-art methods. With its ability segment delineate 3D, our promises be powerful tool medical image processing optimization, saving time systems.

Language: Английский

Citations

8
Diagnostic biomarker discovery from brain EEG data using LSTM, reservoir-SNN, and NeuCube methods in a pilot study comparing epilepsy and migraine DOI Creative Commons
Samaneh Alsadat Saeedinia,

Mohammad Reza Jahed‐Motlagh,

Abbas Tafakhori

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: May 9, 2024

Abstract The study introduces a new online spike encoding algorithm for spiking neural networks (SNN) and suggests methods learning identifying diagnostic biomarkers using three prominent deep network models: BiLSTM, reservoir SNN, NeuCube. EEG data from datasets related to epilepsy, migraine, healthy subjects are employed. Results reveal that BiLSTM hidden neurons capture biological significance, while SNN activities NeuCube dynamics identify channels as biomarkers. achieve 90 85% classification accuracy, achieves 97%, all pinpointing potential like T6, F7, C4, F8. research bears implications refining classification, analysis, early brain state diagnosis, enhancing AI models with interpretability discovery. proposed techniques hold promise streamlined brain-computer interfaces clinical applications, representing significant advancement in pattern discovery across the most popular addressing crucial problem. Further is planned how can these predict an onset of states.

Language: Английский

Citations

7