XAI-MRI: An Ensemble Dual-Modality Approach for 3D Brain Tumor Segmentation Using Magnetic Resonance Imaging DOI Creative Commons
Ahmeed Suliman Farhan, Muhammad Khalid, Umar Manzoor

и другие.

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Ноя. 21, 2024

Brain tumor segmentation from Magnetic Resonance Images (MRI) presents significant challenges due to the complex nature of brain tissues. This complexity makes distinguishing tissues healthy difficult, mainly when radiologists perform manual segmentation. Reliable and accurate is crucial for effective grading treatment planning. In this paper, we proposed a novel ensemble dual-modality approach 3D using MRI. Initially, individual U-Net models are trained evaluated on single MRI modalities (T1, T2, T1ce, FLAIR) establish each modality's performance. Subsequently, U-net combinations best-performing exploit complementary information improve accuracy. The most then integrated into an model, designed capture strengths modality combination. Finally, suggested by combining two pre-trained dual-modalities enhance Experimental results demonstrate model significantly improved performance, achieving Dice Coefficient 97.73% Mean IoU 60.08% BraTS2020 dataset. Our that outperforms traditional single-modality models, providing more robust method study underscores potential precision reliability MRI-based code publicly available at: https://github.com/Ahmeed-Suliman-Farhan/Ensemble-Dual-Modality-Approach .

Язык: Английский

XAI-MRI: an ensemble dual-modality approach for 3D brain tumor segmentation using magnetic resonance imaging DOI Creative Commons
Ahmeed Suliman Farhan, Muhammad Khalid, Umar Manzoor

и другие.

Frontiers in Artificial Intelligence, Год журнала: 2025, Номер 8

Опубликована: Фев. 19, 2025

Brain tumor segmentation from Magnetic Resonance Images (MRI) presents significant challenges due to the complex nature of brain tissues. This complexity poses a challenge in distinguishing tissues healthy tissues, particularly when radiologists rely on manual segmentation. Reliable and accurate is crucial for effective grading treatment planning. In this paper, we proposed novel ensemble dual-modality approach 3D using MRI. Initially, individual U-Net models are trained evaluated single MRI modalities (T1, T2, T1ce, FLAIR) establish each modality's performance. Subsequently, U-net combinations best-performing exploit complementary information improve accuracy. Finally, introduced by combining two pre-trained dual-modalities enhance Experimental results show that model enhanced result achieved Dice Coefficient 97.73% Mean IoU 60.08%. The illustrate outperforms single-modality models. Grad-CAM visualizations implemented, generating heat maps highlight regions provide useful clinicians about how made decision, increasing their confidence deep learning-based systems. Our code publicly available at: https://github.com/Ahmeed-Suliman-Farhan/Ensemble-Dual-Modality-Approach.

Язык: Английский

Процитировано

0
Advanced brain tumor segmentation using DeepLabV3Plus with Xception encoder on a multi-class MR image dataset DOI
Shoffan Saifullah, Rafał Dreżewski, Anton Yudhana

и другие.

Multimedia Tools and Applications, Год журнала: 2025, Номер unknown

Опубликована: Фев. 21, 2025

Язык: Английский

Процитировано

0
Brain tumor segmentation with deep learning: Current approaches and future perspectives DOI
Akash Verma, Arun Kumar Yadav

Journal of Neuroscience Methods, Год журнала: 2025, Номер unknown, С. 110424 - 110424

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0
X‐SCSANet: Explainable Stack Convolutional Self‐Attention Network for Brain Tumor Classification DOI Creative Commons

Rahad Khan,

Rafiqul Islam

International Journal of Intelligent Systems, Год журнала: 2025, Номер 2025(1)

Опубликована: Янв. 1, 2025

Brain tumors are devastating and shorten the patient’s life. It has an impact on physical, psychological, financial well‐being of both patients family members. Early diagnosis treatment can reduce patients’ chances survival. Detecting diagnosing brain cancers using MRI scans is time‐consuming requires expertise in that domain. Nowadays, instead traditional approaches to tumor analysis, several deep learning models used assist professionals mitigate time. This paper introduces a stack convolutional self‐attention network extracts important local global features from freely available scan dataset. Since medical domain one most sensitive fields, end‐users should put their trust model before automating classification. Therefore, Grad‐CAM method been updated better explain model’s output. Combining improves classification performance, with suggested reaching accuracy 96.44% relevant The proposed precision, specificity, sensitivity, F1‐score reported as 96.5%, 98.83%, 96.44%, 96.4%, respectively. Furthermore, layers’ insights examined acquire deeper knowledge decision‐making process.

Язык: Английский

Процитировано

0
Modal Feature Supplementation Enhances Brain Tumor Segmentation DOI

Kaiyan Zhu,

Weiye Cao,

Jianhao Xu

и другие.

International Journal of Imaging Systems and Technology, Год журнала: 2025, Номер 35(3)

Опубликована: Апрель 3, 2025

ABSTRACT For patients with brain tumors, effectively utilizing the complementary information between multimodal medical images is crucial for accurate lesion segmentation. However, features across different modalities remains a challenging task. To address these challenges, we propose modal feature supplement network (MFSNet), which extracts modality simultaneously using both main and an auxiliary network. During this process, supplements of network, enabling tumor We also design enhancement module (MFEM), cross‐layer fusion (CFFM), edge (EFSM). MFEM enhances performance by fusing from networks. CFFM additional contextual adjacent encoding layers at scales, are then passed into corresponding decoding layers. This aids in preserving more details during upsampling. EFSM improves deformable convolution to extract boundary features, used final output layer. evaluated MFSNet on BraTS2018 BraTS2021 datasets. The Dice scores whole tumor, core, enhancing regions were 90.86%, 90.59%, 84.72%, 92.28%, 92.47%, 86.07%, respectively. validates accuracy segmentation, demonstrating its superiority over other networks similar type.

Язык: Английский

Процитировано

0
An Efficient Approach of Brain Tumor Detection & Extraction using BWT with Auto Enhance Technique DOI Open Access
Nilesh Bhaskarrao Bahadure, Sudhanshu Gonge, Jagdish Chandra Patni

и другие.

Procedia Computer Science, Год журнала: 2025, Номер 258, С. 4105 - 4116

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
Optimizing Cancer Patient Classification Forecasting With Bayesian Pattern Recognition DOI Open Access
Praowpan Tansitpong

International Journal of Healthcare Information Systems and Informatics, Год журнала: 2024, Номер 19(1), С. 1 - 21

Опубликована: Авг. 14, 2024

This research examines patterns in cancer treatment by analyzing electronic medical record (EMR) data, with the goal of optimizing healthcare provision and improving patient outcomes. The study aims to apply Bayesian prediction models regression analysis determine posterior probability comorbidities forecast arrivals. implemented algorithms allow for customization techniques, resulting enhanced effectiveness therapy improved decision-making delivery. Utilizing approaches analyze EMR data provides insights into intricacies related expenses. application this could be useful enhance information systems informatics using data-driven improve care practices operational efficiency hospital settings.

Язык: Английский

Процитировано

0
XAI-MRI: An Ensemble Dual-Modality Approach for 3D Brain Tumor Segmentation Using Magnetic Resonance Imaging DOI Creative Commons
Ahmeed Suliman Farhan, Muhammad Khalid, Umar Manzoor

и другие.

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Ноя. 21, 2024

Brain tumor segmentation from Magnetic Resonance Images (MRI) presents significant challenges due to the complex nature of brain tissues. This complexity makes distinguishing tissues healthy difficult, mainly when radiologists perform manual segmentation. Reliable and accurate is crucial for effective grading treatment planning. In this paper, we proposed a novel ensemble dual-modality approach 3D using MRI. Initially, individual U-Net models are trained evaluated on single MRI modalities (T1, T2, T1ce, FLAIR) establish each modality's performance. Subsequently, U-net combinations best-performing exploit complementary information improve accuracy. The most then integrated into an model, designed capture strengths modality combination. Finally, suggested by combining two pre-trained dual-modalities enhance Experimental results demonstrate model significantly improved performance, achieving Dice Coefficient 97.73% Mean IoU 60.08% BraTS2020 dataset. Our that outperforms traditional single-modality models, providing more robust method study underscores potential precision reliability MRI-based code publicly available at: https://github.com/Ahmeed-Suliman-Farhan/Ensemble-Dual-Modality-Approach .

Язык: Английский

Процитировано

0