Investigation and Fine Tuning of Hyper Parameters on U-Net Model for Segmentation of Glioma (original) (raw)
Related papers
A review of deep learning models (U-Net architectures) for segmenting brain tumors
Highly accurate tumor segmentation and classification are required to treat the brain tumor appropriately. Brain tumor segmentation (BTS) approaches can be categorized into manual, semi-automated, and full-automated. The deep learning (DL) approach has been broadly deployed to automate tumor segmentation in therapy, treatment planning, and diagnosing evaluation. It is mainly based on the U-Net model that has recently attained state-of-the-art performances for multimodal BTS. This paper demonstrates a literature review for BTS using U-Net models. Additionally, it represents a common way to design a novel U-Net model for segmenting brain tumors. The steps of this DL way are described to obtain the required model. They include gathering the dataset, pre-processing, augmenting the images (optional), designing/selecting the model architecture, and applying transfer learning (optional). The model architecture and the performance accuracy are the two most important metrics used to review the works of literature. This review concluded that the model accuracy is proportional to its architectural complexity, and the future challenge is to obtain higher accuracy with lowcomplexity architecture. Challenges, alternatives, and future trends are also presented.
A U-Net-Based CNN Model for Detection and Segmentation of Brain Tumor
Computers, Materials & Continua
Human brain consists of millions of cells to control the overall structure of the human body. When these cells start behaving abnormally, then brain tumors occurred. Precise and initial stage brain tumor detection has always been an issue in the field of medicines for medical experts. To handle this issue, various deep learning techniques for brain tumor detection and segmentation techniques have been developed, which worked on different datasets to obtain fruitful results, but the problem still exists for the initial stage of detection of brain tumors to save human lives. For this purpose, we proposed a novel U-Net-based Convolutional Neural Network (CNN) technique to detect and segmentizes the brain tumor for Magnetic Resonance Imaging (MRI). Moreover, a 2-dimensional publicly available Multimodal Brain Tumor Image Segmentation (BRATS2020) dataset with 1840 MRI images of brain tumors has been used having an image size of 240 × 240 pixels. After initial dataset preprocessing the proposed model is trained by dividing the dataset into three parts i.e., testing, training, and validation process. Our model attained an accuracy value of 0.98 % on the BRATS2020 dataset, which is the highest one as compared to the already existing techniques.
Automated Brain Tumor Diagnosis Using Deep Residual U-Net Segmentation燤odel
Computers, Materials & Continua
Automated segmentation and classification of biomedical images act as a vital part of the diagnosis of brain tumors (BT). A primary tumor brain analysis suggests a quicker response from treatment that utilizes for improving patient survival rate. The location and classification of BTs from huge medicinal images database, obtained from routine medical tasks with manual processes are a higher cost together in effort and time. An automatic recognition, place, and classifier process was desired and useful. This study introduces an Automated Deep Residual U-Net Segmentation with Classification model (ADRU-SCM) for Brain Tumor Diagnosis. The presented ADRU-SCM model majorly focuses on the segmentation and classification of BT. To accomplish this, the presented ADRU-SCM model involves wiener filtering (WF) based preprocessing to eradicate the noise that exists in it. In addition, the ADRU-SCM model follows deep residual U-Net segmentation model to determine the affected brain regions. Moreover, VGG-19 model is exploited as a feature extractor. Finally, tunicate swarm optimization (TSO) with gated recurrent unit (GRU) model is applied as a classification model and the TSO algorithm effectually tunes the GRU hyperparameters. The performance validation of the ADRU-SCM model was tested utilizing FigShare dataset and the outcomes pointed out the better performance of the ADRU-SCM approach on recent approaches.
Exploring the U-Net++ model for Automatic Brain Tumor Segmentation
IEEE Access
The accessibility and potential of deep learning techniques have increased considerably over the past years. Image segmentation is one of the many fields which have seen novel implementations being developed to solve problems in the domain. U-Net is an example of a popular deep learning model designed specifically for biomedical image segmentation, initially proposed for cell segmentation. We propose a variation of the U-Net++ model, which is itself an adaptation of U-Net, and evaluate its brain tumor segmentation capabilities. The proposed approach obtained Dice Coefficient scores of 0.7192, 0.8712, and 0.7817 for the Enhancing Tumor, Whole Tumor and Tumor Core classes of the BraTS 2019 challenge Validation Dataset. The proposed approach differs from the standard U-Net++ model in a number of ways, including the loss function, number of convolutional blocks, and method of employing deep supervision. Data augmentation and post-processing techniques were also implemented and observed to substantially improve the model predictions. Thus, this article presents a novel adaptation of the U-Net++ architecture, which is both lightweight, and performs comparably with peer-reviewed work evaluated on the same data.
Modified U-Net for Automatic Brain Tumor Regions Segmentation
2019 27th European Signal Processing Conference (EUSIPCO), 2019
Novel deep learning based network architectures are investigated for advanced brain tumor image classification and segmentation. Variations in brain tumor characteristics together with limited labelled datasets represent significant challenges in automatic brain tumor segmentation. In this paper, we present a novel architecture based on the U-Net that incorporates both global and local feature extraction paths to improve the segmentation accuracy. The results included in the paper show superior performance of the novel segmentation for five tumor regions on the large BRATs 2018 dataset over other approaches.
BRAIN CANCER SEGMENTATION IN MRI USING FULLY CONVOLUTIONAL NETWORK WITH THE U-NET MODEL
IEEE Xplore, 2024
The manual segmentation of brain tumors from magnetic resonance (MR) images represents a formidable challenge, imposing significant demands on the time and expertise of medical professionals. This study addresses the complexity of sematic segmentation in brain tumor detection, acknowledging the necessity for meticulous preprocessing and post-processing procedures. The proposed approach leverages the power absolutely Fully Convolutional Network with the U-Net model architecture, emphasizing the critical role of segmentation in cases where accurate and timely clinical diagnosis is pivotal for patient survival. The intricacies of brain tumor detection demand an advanced neural network architecture capable of discerning subtle details in MR images. By employing a FCN, the main aim is to streamline the segmentation process, mitigating the burden on healthcare practitioners. The incorporation of the U-Net model enhances the network's ability to capture intricate spatial features, ensuring a comprehensive understanding of the tumor boundaries. This research underscores the significance of leveraging deep learning techniques in medical imaging, particularly in the condition of brain tumor detection. The proposed FCN with U-Net architecture not only demonstrates robust segmentation capabilities but also addresses the need for expeditious and accurate clinical diagnoses. The findings contribute to the ongoing efforts for bettering medical image quality analysis, offering a potential breakthrough in the realm of neuro imaging and facilitating improved patient outcomes.
Brain Tumor Segmentation of MRI Images Using Processed Image Driven U-Net Architecture
Computers
Brain tumor segmentation seeks to separate healthy tissue from tumorous regions. This is an essential step in diagnosis and treatment planning to maximize the likelihood of successful treatment. Magnetic resonance imaging (MRI) provides detailed information about brain tumor anatomy, making it an important tool for effective diagnosis which is requisite to replace the existing manual detection system where patients rely on the skills and expertise of a human. In order to solve this problem, a brain tumor segmentation & detection system is proposed where experiments are tested on the collected BraTS 2018 dataset. This dataset contains four different MRI modalities for each patient as T1, T2, T1Gd, and FLAIR, and as an outcome, a segmented image and ground truth of tumor segmentation, i.e., class label, is provided. A fully automatic methodology to handle the task of segmentation of gliomas in pre-operative MRI scans is developed using a U-Net-based deep learning model. The first step...
Brain Tumor Segmentation Using Modified U-Net
Research Square (Research Square), 2022
Planning and evaluating the extent of the spread of brain tumors for the treatment is the main challenge. Magnetic resonance imaging (MRI) has been shown as a great diagnostic tool for brain tumors without harmful radiation. We have found in using the traditional U-net, that there are some errors in identifying the tumor, so we wanted to develop this method so that it overcomes some errors in determining the tumor in the traditional method. Methods: We demonstrate a way to segment the brain tumor using modify U-net networks. Then compared with traditional Unet, modified U-net, k-means, and thresholding segmentation. The experiments were carried out using three multimodal brain tumor image segmentation datasets (FIGSHARE database), which contain 3929 abnormal (with a tumor) and normal brain MRI images, 100 images obtained from "The Cancer Imaging Archive (TCIA)" and BraTS 2019 challenge which include 4600 cases (normal and abnormal) of HGG (high-grade glioma) and LGG(low-grade glioma). Results: Through this study, the modified U-net achieved higher accuracy than other methods. Conclusion: Determining the tumor inside the human brain is an important matter to preserve human life. We used old and modern techniques and modified the basic form of the traditional U-net. The above-mentioned techniques were compared and
Enhancement of the U-net Architecture for MRI Brain Tumor Segmentation
Lecture Notes in Networks and Systems, 2021
The magnetic resonance images (MRI) brain tumor segmentation is one of the most difficult medical images segmentation, and it has many challenges because the tumor has no specific shape or size, not found in a specific place of the brain and contains three sub-regions (full tumor FT, tumor core TC, and enhanced tumor ET). The manual segmentation is extremely difficult and prone to mistakes. In this research, the semantic segmentation is used by implementing the U-net model, which is a fully convolutional network (FCN) algorithm on BraTS 2018 that contains four modalities (T1, T2, T1c, Flair). The U-net model is used two time: The first is using 9-layers U-net with some enhancements on the original model for segment the full tumor, and the second is using 7-layers U-net model for segment the tumor core and enhanced tumor. The results were promised by segmenting the brain tumor within three sub-regions (full tumor, tumor core, and enhanced tumor), and the results were evaluated using standard brain tumor segmentation metrics. The proposed system achieves mean dice similarity coefficient metric; it is 0.87, 0.76, and 0.71 for full tumor, tumor core, and enhanced tumor, respectively. Additionally, the median dice similarity coefficient metric is 0.90, 0.84, and 0.80 for full tumor, tumor core, and enhanced tumor, respectively.
Healthcare
Brain tumors (BTs) are an uncommon but fatal kind of cancer. Therefore, the development of computer-aided diagnosis (CAD) systems for classifying brain tumors in magnetic resonance imaging (MRI) has been the subject of many research papers so far. However, research in this sector is still in its early stage. The ultimate goal of this research is to develop a lightweight effective implementation of the U-Net deep network for use in performing exact real-time segmentation. Moreover, a simplified deep convolutional neural network (DCNN) architecture for the BT classification is presented for automatic feature extraction and classification of the segmented regions of interest (ROIs). Five convolutional layers, rectified linear unit, normalization, and max-pooling layers make up the DCNN’s proposed simplified architecture. The introduced method was verified on multimodal brain tumor segmentation (BRATS 2015) datasets. Our experimental results on BRATS 2015 acquired Dice similarity coeffi...