Tumor classification with MRI (Magnetic Resonance Imaging) is critical, as it consumes an enormous amount of time. Furthermore, this detection method is complicated due to the similarity of both ...abnormal and normal brain tissues. For earlier treatment planning and clinical assessment of brain tumors, automatic segmentation and classification process using medical images are very challenging. Computerized medical imaging aids clinicians in providing critical therapies to patients while allowing faster decision-making. This work focus on efficient segmentation and classification using machine learning (ML) models motivated by diagnosing tumor growth and treatment processes. To achieve efficient brain tumor detection, different stages in the proposed methodology are pre-processing, segmentation, extraction, selection and classification. Initially, blur-removal is done using NMF (Normalized Median Filter) for image smoothening and quality enhancement. Then segmentation is done using binomial thresholding method. The next step is feature extraction, which is the fusion of GLCM (Gray level co-occurrence matrix), and SGLDM (Spatial Grey Level Dependence Matrix) techniques. Harris hawks optimization (HHO) algorithm is used for feature selection. Finally, KSVM-SSD is used for effective and accurate classification. Here, the brain tumor is classified as benign and malignant using KSVM (Kernel Support Vector Machine) and further classification of the malignant tumor as low, medium, and high using social ski driver (SSD) optimization algorithm. The simulation/implementation tool used here is the PYTHON platform. The performance is analyzed on multiple datasets such as BRATS 2018, 2019 and 2020. Hence, it is proved that the segmentation and classification outcomes are superior compared to existing methods with precision, accuracy, recall, and F1 score. The superiority of the proposed KSVM-SSD model is identified in terms of classification accuracy tested on the BRATS datasets with accuracy as 99.2%, 99.36% and 99.15%, respectively for 2018, 2019 and 2020 BRATS datasets. Higher detection accuracy offers timely and proper diagnosis that can save the lives of people. Hence, these outcomes on tumor detection and classification signifiy improved performance when compared to baseline models.
We have developed six convolutional neural network (CNN) models for finding optimal brain tumor detection system on high‐grade glioma and low‐grade glioma lesions from voluminous magnetic resonance ...imaging human brain scans. Glioma is the most common form of brain tumor. The models are constructed based on the different combinations and settings of hyperparameters with conventional CNN architecture. The six models are two layers with five epochs, five layers with dropout, five layers with stopping criteria (FLSC), FLSC and dropout (FLSCD), FLSC and batch normalization (FLSCBN), and FLSCBN and dropout. The models were trained and tested with BraTS2013 and whole brain atlas data sets. Among them, FLSCBN model yielded the best classification results for brain tumor detection. Experimental results revealed that our deep learning approach was better than the conventional state‐of‐art methods.
This paper proposes a novel approach, BTC-SAGAN-CHA-MRI, for the classification of brain tumors using a SAGAN optimized with a Color Harmony Algorithm. Brain cancer, with its high fatality rate ...worldwide, especially in the case of brain tumors, necessitates more accurate and efficient classification methods. While existing deep learning approaches for brain tumor classification have been suggested, they often lack precision and require substantial computational time.The proposed method begins by gathering input brain MR images from the BRATS dataset, followed by a pre-processing step using a Mean Curvature Flow-based approach to eliminate noise. The pre-processed images then undergo the Improved Non-Sub sampled Shearlet Transform (INSST) for extracting radiomic features. These features are fed into the SAGAN, which is optimized with a Color Harmony Algorithm to categorize the brain images into different tumor types, including Gliomas, Meningioma, and Pituitary tumors. This innovative approach shows promise in enhancing the precision and efficiency of brain tumor classification, holding potential for improved diagnostic outcomes in the field of medical imaging. The accuracy acquired for the brain tumor identification from the proposed method is 99.29%. The proposed BTC-SAGAN-CHA-MRI technique achieves 18.29%, 14.09% and 7.34% higher accuracy and 67.92%,54.04%, and 59.08% less Computation Time when analyzed to the existing models, like Brain tumor diagnosis utilizing deep learning convolutional neural network with transfer learning approach (BTC-KNN-SVM-MRI); M3BTCNet: multi model brain tumor categorization under metaheuristic deep neural network features optimization (BTC-CNN-DEMFOA-MRI), and efficient method depending upon hierarchical deep learning neural network classifier for brain tumour categorization (BTC-Hie DNN-MRI) respectively.
Accurate brain tumor segmentation is crucial for clinical assessment, follow-up, and subsequent treatment of gliomas. While convolutional neural networks (CNN) have become state of the art in this ...task, most proposed models either use 2D architectures ignoring 3D contextual information or 3D models requiring large memory capacity and extensive learning databases. In this study, an ensemble of two kinds of U-Net-like models based on both 3D and 2.5D convolutions is proposed to segment multimodal magnetic resonance images (MRI). The 3D model uses concatenated data in a modified U-Net architecture. In contrast, the 2.5D model is based on a multi-input strategy to extract low-level features from each modality independently and on a new 2.5D Multi-View Inception block that aims to merge features from different views of a 3D image aggregating multi-scale features. The Asymmetric Ensemble of Asymmetric U-Net (AE AU-Net) based on both is designed to find a balance between increasing multi-scale and 3D contextual information extraction and keeping memory consumption low. Experiments on 2019 dataset show that our model improves enhancing tumor sub-region segmentation. Overall, performance is comparable with state-of-the-art results, although with less learning data or memory requirements. In addition, we provide voxel-wise and structure-wise uncertainties of the segmentation results, and we have established qualitative and quantitative relationships between uncertainty and prediction errors. Dice similarity coefficient for the whole tumor, tumor core, and tumor enhancing regions on BraTS 2019 validation dataset were 0.902, 0.815, and 0.773. We also applied our method in BraTS 2018 with corresponding Dice score values of 0.908, 0.838, and 0.800.
•We treat an automatic 3D/2D MRI images using intelligent metaheuristic such as PSO based on FPGA to obtain better results either qualitatively or quantitatively. The benefits of this work can be ...spread into 4 levels.•Algorithmic level: The FODPSO algorithm represents the best optimization method compared to other metaheuristics and it facilitates the search for the optimal threshold.•Implementation level: The use of hardware FPGA target reduce the computation time for multilevel segmentation based on pipeline which increases the clock frequency and the data processing rate.•Design level: The use of VSG tool allows a single design for validation and for automatic generation of the implementable architecture.•Data level: The use of 2D algorithms to segment 3D data for MRI images will greatly reduce the intensity of computation and the degree of parallelism and so reduce the execution time.
Magnetic resonance imaging is among the advanced diagnostic testing tools for brain health issues. This method captures a series of detailed head images. These images are then printed and diagnosed by a specialist doctor to demonstrate differences in the brain tissue. Accordingly, additional diagnostic information can be given to determine the extent of the damage and the appropriate treatment methods. In this paper, and in order to facilitate the work of the specialist doctor and help him, we propose an automated hardware architecture for 3D/2D segmentation on MRI images to diagnose differences in brain tissue. For this, we used the metaheuristic technique based on Particle Swarm Optimization (PSO); for which we proposed improvements both for the velocity and position equations and for the fitness function. The goal of the work is to develop a real time automatic system for MRI images segmentation with improved metrics such as accuracy, sensitivity, specificity, dice metrics, execution time and resources utilization. The proposed hardware architecture was synthetized and then co-simulated using Matlab-Vivado System (VSM) for Field Programmable Gate Array (FPGA). Results show that our 3D segmentation method benefited from 2D segmentation with 95.39% accuracy rate and 87.97% DSC similarity (for 5-level segmentation) with 4.57 ms execution time for the case of BraTS 2013 dataset of brain MRI Images.
Brain tumor segmentation is an important direction in medical image processing, and its main goal is to accurately mark the tumor part in brain MRI. This study proposes a brand new end-to-end model ...for brain tumor segmentation, which is a multi-scale deep residual convolutional neural network called mResU-Net. The semantic gap between the encoder and decoder is bridged by using skip connections in the U-Net structure. The residual structure is used to alleviate the vanishing gradient problem during training and ensure sufficient information in deep networks. On this basis, multi-scale convolution kernels are used to improve the segmentation accuracy of targets of different sizes. At the same time, we also integrate channel attention modules into the network to improve its accuracy. The proposed model has an average dice score of 0.9289, 0.9277, and 0.8965 for tumor core (TC), whole tumor (WT), and enhanced tumor (ET) on the BraTS 2021 dataset, respectively. Comparing the segmentation results of this method with existing techniques shows that mResU-Net can significantly improve the segmentation performance of brain tumor subregions.
Graphical abstract
Brain tumor segmentation (BTS) from magnetic resonance imaging (MRI) scans is crucial for the diagnosis, treatment planning, and monitoring of therapeutic results. Thus, this research work proposes a ...novel graph momentum fully convolutional network with a modified Elman spike neural network (MESNN) for BTS and overall survival prediction (OSP). Initially, the introduced graph momentum fully convolutional network segments the brain tumor as enhanced tumor, the tumor core, and the whole tumor from the pre‐processed MRI scans. Second, the texture, intensity, shape, and wavelet features were extracted from the segmented tumors. Then, the horse herd optimization algorithm is utilized to minimize the feature's dimensionality. Finally, the OSP is performed by the MESNN which classifies the survival prediction of a patient as long‐term, mid‐term, and short‐term. The achieved segmentation accuracy of proposed method is 97% and the survival prediction's average RMSE is 215.5.
The surging use of medical AI algorithms and their hardware integration is transforming healthcare by improving non-invasive medical analysis with early disease detection, advanced segmentation, and ...classification. However, realizing comprehensive and accurate medical analysis through efficient AI-based tools necessitates a fundamental requirement - extensive multimodal data for training deep learning models. Handling this extensive data volume demands significant hardware resources, including multi-node training, to address the substantial computational requirements essential for accelerating model development. Hence, the challenge is two-fold: Achieving high accuracy while upholding a computationally inexpensive solution. To navigate this challenge, we propose a novel and efficient solution: a lightweight predictive tool for medical image classification developed by combining a Radiomics-based Random Forest model with MobileViT transformer, tailored for mobile applications. This approach ensures enhanced accuracy and reproducibility along with hardware flexibility. Our proposed method is exemplified by its superior performance in the BraTS2021 challenge, surpassing current state-of-the-art models with the best AUROC of 0.64 and 0.63 on both public and private test datasets respectively. The success of our approach highlights the potential of hybrid models in diverse medical applications beyond image classification.
Summary
Brain tumor is caused by the growth of abnormal cells, which forms a mass and affects the brain functions. The existing methods did not provide sufficient accuracy with high computational ...complexity. Therefore, in this manuscript, a Dropout AlexNet‐Extreme Learning optimized with Fast Gradient Descent optimization algorithm is proposed for detecting and classifying images of brain tumor. Here, the input magnetic resonance imaging images are taken from three datasets: BRATS dataset, ISLES dataset, and RemBRANDT Dataset. Then the imageries are preprocessed to remove the noise as well as improve the superiority of the images. The image features are extracted using the Gray‐Level Co‐Occurrence Matrix methods. The extracted features are given to the Dropout AlexNet‐XtremeLearning Machine architecture for classification. Finally, the DrpXLM classifier classifying the brain images as benign, malignant, and normal. The simulation is implemented in MATLAB. For BRATS dataset, the proposed strategy achieves34.64%, 45.36%, and 33.32% higher accuracy for benign, 37.85%, 28.94%, and 56.74% higher accuracy for malignant and 46.76%, 38.96%, and 44.86% better accuracy for normal compared with the existing methods, like Gaussian filters and long short‐term memory based brain tumor detection (GF‐LSTM‐BTD), Shannon's‐Entropy and Social‐Group‐Optimization based brain tumor detection (SE‐SGO‐BTD), Alex and Google networks with softmax layer based brain tumor detection (AGN‐SOFT‐BTD).