Brain Tumor and Haemangioma Discrimination Framework Using Unsupervised Feature Selection

Since magnetic resonance imaging (MRI) has superior soft tissue contrast, contouring (brain) tumor accurately by MRI images is essential in medical image processing. Segmenting tumor accurately is immensely challenging, since tumor and normal tissues are often inextricably intertwined in the brain. It is also extremely time consuming manually. Late deep learning techniques start to show reasonable success in brain tumor segmentation automatically. The purpose of this study is to develop a new region-of-interest-aided (ROI-aided) deep learning technique for automatic brain tumor MRI segmentation. The method consists of two major steps. Step one is to use a 2D network with U-Net architecture to localize the tumor ROI, which is to reduce the impact of normal tissue’s disturbance. Then a 3D U-Net is performed in step 2 for tumor segmentation within identified ROI. The proposed method is validated on MICCAI BraTS 2015 Challenge with 220 high Gliomas grade (HGG) and 54 low Gliomas grade (LGG) patients’ data. The Dice similarity coefficient and the Hausdorff distance between the manual tumor contour and that segmented by the proposed method are 0.876 ±0.068 and 3.594±1.347 mm, respectively. These numbers are indications that our proposed method is an effective ROI-aided deep learning strategy for brain MRI tumor segmentation, and a valid and useful tool in medical image processing.

Gamma knife treatment is a clinically effective, safe, and cost-effective adjunctive therapy for primary malignant brain tumors. For most brain metastases, radiosurgery is the treatment of choice and will result in effective tumor control in more than 90% of treated tumors.

Purpose One of the foremost research disciplines in medical image processing is to identify tumors, which is a challenging task practicing traditional methods. To overcome this, various research studies have been done effectively. Design/methodology/approach Medical image processing is evolving swiftly with modern technologies being developed every day. The advanced technologies improve medical fields in diagnosing diseases at the more advanced stages and serve to provide proper treatment. Findings Either the mass growth or abnormal growth concerning the cells in the brain is called a brain tumor. Originality/value The brain tumor can be categorized into two significant varieties, non-cancerous and cancerous. The carcinogenic tumors or cancerous is termed as malignant and non-carcinogenic tumors are termed benign tumors. If the cells in the tumor are healthy then it is a benign tumor, whereas, the abnormal growth or the uncontrollable growth of the cell is indicated as malignant. To find the tumor the magnetic resonance imaging (MRI) is carried out which is a tiresome and monotonous task done by a radiologist. In-order to diagnosis the brain tumor at the initial stage effectively with improved accuracy, the computer-aided robotic research technology is incorporated. There are numerous segmentation procedures, which help in identifying tumor cells from MRI images. It is necessary to select a proper segmentation mechanism to detect brain tumors effectively that can be aided with robotic systems. This research paper focuses on self-organizing map (SOM) by applying the adaptive network-based fuzzy inference system (ANFIS). The execution measures are determined to employ the confusion matrix, accuracy, sensitivity, and furthermore, specificity. The results achieved conclusively explicate that the proposed model presents more reliable outcomes when compared to existing techniques.

Computer-aided research to improve image decoding is a long-standing theme in medical imaging. A variety of imaging techniques, including ultrasound imaging, The Magnetic Resonance Imaging (MRI), and Computed Tomography (CT), are generally depleted to estimate tumors in the prostate, lung, brain, breast, and liver. The study used MRI images of the brain to identify the tumors. Brain tumors are an almost common and cruel disease that can significantly shorten life expectancy. It is important to use MRI images to locate and classify contaminating tumors. There are many tumors including gliomas, meningiomas, pituitary tumors, and no tumors. One of the most difficult aspects of brain tumor assortment is the diagnosis and prevention of tumor type. Accurate tumor classification helps to assess disease progression and select therapeutic strategies. To resolve the issue, a Deep Belief Neural- Convolutional Neural Network (DeepBeliefCNN) method was proposed. At first phase, we preprocess the brain tumor MRI dataset by using 2D Wavelet Filter method. This method allows for the analysis of images at multiple resolutions, thus enabling detection of features at different scales. This effectively reduces noise in medical images while preserving important details. This improves the definition of tumor boundaries and other important features, helping in accurate diagnosis and analysis. Then preprocessing the dataset segment the preprocessed images based on Watershed method. It is a powerful image segmentation technique used to outline objects in images. When used for image segmentation of a brain tumor, it can help identify tumor boundaries within the brain. Markers are assigned to areas of interest in the image. These markers can be manually set by experts or generated automatically using techniques such as distance transforms and morphological functions. At last, the dataset is classify by using the DeepBeleifCNN method. The DeepBeleifCNN approach combines the hierarchical feature extraction capabilities of DBN with the spatial feature extraction capabilities of CNN. This allows a detailed understanding of brain tumor images and improves classification performance. This method successfully generalizes to new and unrecognized brain tumor images because of its capability to study healthy and biased features in together pre-training and fine-tuning stages. The investigational outcomes illustrate that our deployed methodology beats the existing method in accuracy, sensitivity, F1 score, specificity and error rate.

Dual vision Transformer-DSUNET with feature fusion for brain tumor segmentation

PurposeAutomatic segmentation of brain tumor from medical images is a challenging task because of tumor's uneven and irregular shapes. In this paper, the authors propose an attention-based nested segmentation network, named DAU-Net. In total, two types of attention mechanisms are introduced to make the U-Net network focus on the key feature regions. The proposed network has a deep supervised encoder–decoder architecture and a redesigned dense skip connection. DAU-Net introduces an attention mechanism between convolutional blocks so that the features extracted at different levels can be merged with a task-related selection.Design/methodology/approachIn the coding layer, the authors designed a channel attention module. It marks the importance of each feature graph in the segmentation task. In the decoding layer, the authors designed a spatial attention module. It marks the importance of different regional features. And by fusing features at different scales in the same coding layer, the network can fully extract the detailed information of the original image and learn more tumor boundary information.FindingsTo verify the effectiveness of the DAU-Net, experiments were carried out on the BRATS 2018 brain tumor magnetic resonance imaging (MRI) database. The segmentation results show that the proposed method has a high accuracy, with a Dice similarity coefficient (DSC) of 89% in the complete tumor, which is an improvement of 8.04 and 4.02%, compared with fully convolutional network (FCN) and U-Net, respectively.Originality/valueThe experimental results show that the proposed method has good performance in the segmentation of brain tumors. The proposed method has potential clinical applicability.

Tumor.

Methylation of the TERT promoter and risk stratification of childhood brain tumours: an integrative genomic and molecular study

Navigating the challenges of neuro-oncology in Africa: addressing diagnostic and treatment barriers in the region: a correspondence

Proteolytic Cleavage of the CD44 Adhesion Molecule in Multiple Human Tumors

Brain and central nervous system (CNS) tumors found in adolescents and young adults (AYA) are a distinct group of tumors that pose challenges not only to treatment but also to reporting. Overall, cancer that occurs in this age group is biologically distinct from those that occur in both younger and older age groups1,2 posing significant challenges for clinicians. The most commonly diagnosed histologies in AYA vary from those in both children age (0-14 years), and older adults (40+ years).3,4 Prognosis and expected survival also varies between younger and older adults, with those who are diagnosed with brain and CNS tumors at younger ages having significantly longer survival. Despite this survival advantage, recent analyses have reported that while cancer survival has been improving overall, AYA have not experienced these same increases in survival and in some cases may have worse survival than those cancers diagnosed in persons over age 40 years.5 This report provides an in depth analyses of the epidemiology of brain and CNS tumors in adolescents and young adults in the United States (US), and is the first report to provide histology-specific statistics in this population for both malignant and non-malignant brain and other CNS tumors. In 2006, the National Institutes of Health, the National Cancer Institute (NCI) and the LiveStrong Young Adult Alliance conducted a Progress Review Group to investigate AYA Oncology entitled Research and care imperatives for adolescents and young adults with cancer: A Report of the Adolescent and Young Adult Oncology Progress Review Group. This group established the standard age range for the AYA group as 15-39 years. This is the age range used by the Surveillance Epidemiology and End Results (SEER) program of the NCI, as well as in the 2015 CBTRUS Statistical Report.3,6 Brain tumors and other CNS tumors are less common in AYA than in older adults, but they have a higher incidence than brain tumors in children (age 0-14 years).3 Non-malignant tumors are significantly more common in AYA than children (Average annual age adjusted incidence in age 15-39 years: 6.17 per 100,000; age 0-14 years: 0.79 per 100,000), while malignant tumors are slightly more common in those age 0-14 years (Average annual age adjusted incidence in 15-39 years old: 3.26 per 100,000; 0-14 years old: 3.73 per 100,000). While a rare cancer overall, brain and CNS tumors are among the most common cancers occurring in this age group (4.4% of all cancers in those age 15-39 years as compared to 32.4% in children age 0-14 years, and 2.2% of cancers in adults age 40+ years).3,4,7 Malignant brain and CNS tumors are the 11th most common cancer and the 3rd most common cause of cancer death7,8 in the AYA population. Incidence rates of brain tumors overall as well as specific histologies vary significantly by age. It is, therefore, important to provide an accurate statistical assessment of brain and other CNS tumors in the adolescent and young adult population to better understand their impact on the US population and to serve as a reference for afflicted individuals, for researchers investigating new therapies and for clinicians treating patients.

BACKGROUND Population-based data on the various causes of death among Primary Brain and CNS tumor patients are lacking. We evaluated the causes of death for all eligible patients using the National Program of Cancer Registries (NPCR) data. METHODS The population-based cancer survival data collected by the Centers for Disease Control and Prevention’s National Program of Cancer Registries (NPCR) were used to analyze the causes of death for patients of all ages with primary brain and CNS tumors diagnosed between 2001 and 2016. Patients for whom the cause of death was not listed on the death certificate or whose state death certificate was not available were excluded. Additional analyses to identify factors associated with brain tumor-specific mortality for the most common malignant (Glioblastoma) and non-malignant (Meningioma) were performed using univariable and multivariable logistic regression analysis. RESULTS Major cause of death for patients with malignant tumors was death due to brain and other CNS tumors (49.29%), and for non-malignant tumors were other benign and malignant tumors (31.5%) and heart disease (17.9%). Overall mortality was 36.4% (n=331,953) in patients with Primary Brain and CNS Tumors during the study period. Specifically, 163,621 (49.29%) patients died due to brain and other CNS tumors. A significant proportion of patients with malignant tumors had brain tumor-specific mortality compared to non-malignant tumors (75.4% in malignant vs 4.2% in non-malignant). The factors associated with brain specific mortality in Glioblastoma patients were Age (p< 0.001), Race (p< 0.001) and Primary Site (p< 0.001). Further, the factors associated with brain specific mortality in Non-malignant Meningioma patients were Age (p< 0.001), Sex (p< 0.001), Race (p< 0.001) and Primary Site (P< 0.001). CONCLUSION Cause of death attributed to the brain tumor was significantly higher in malignant brain tumors compared to non-malignant brain tumors.

The core purpose of this paper is to compare the efficiency of two methods which are used to segment the brain tumor images. Brain tumor segmentation is an essential procedure for diagnose tumor in earlier stage. Generally, in medical imaging, segmentation of brain tumor images is executed manually in clinical practice. It is a time taking process and so manual brain tumor detection is complicated. To overcome this drawback an automatic brain tumor segmentation method is needed. Among several automatic brain tumor segmentation approaches, this paper investigates two methods and their performances are compared to observe the best method for brain tumor partition. The first method segments the brain tumor images using Local Independent Projection based Classification (LIPC). The second technique uses wavelet and Self Organization Map (SOM). To analyse the performance of these methods, several performance metrics are used. This work utilizes Precision Rate, Recall Rate, F-Measure, Sensitivity and Specificity to examine the efficiency. From the experimental outcomes it is shown that the Wavelet based SOM approach performs superior than the other method.

Aims This research gives insight into the various machine learning models like enhanced Support Vector Machines (SVM), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Artificial Neural Networks (ANN) in brain tumor recognition by medical imaging. This research provides an accurate model for allowing a better form of diagnostic method in neuro-oncology, with the help of precision, recall, and F1-score metrics. The present study, therefore, also provides a basis on which further predictive models for medical image analysis can be developed. Background This study is premised on the critical need for improved diagnostic tools within medical imaging in the fight against the prevalence of brain tumors. A model showing meaningful performance in the practices of brain tumor detection includes enhanced SVM, CNN, RNN, and ANN. The models have been evaluated based on their accuracy, precision, recall, and F1 score to investigate their performance and potential. Consequently, the models addressing the subject of neuro-oncological diagnostics were evaluated. Objective This study seeks to critically evaluate the performance of four different machine learning models: enhanced SVM, CNN, RNN, and ANN, in detecting a brain tumor. It will be determined from this study which model has the highest accuracy, precision, and recall in finding a brain tumor. It will then lead to the improvement of diagnostic techniques in neuro-oncology. Methods The methodology of this research involved a detailed assessment of four machine learning models: enhanced SVM, CNN, RNN, and ANN. Each model was evaluated based on accuracy, precision, recall, and F1 score metrics. The analysis focused on their ability to detect brain tumors from medical imaging data, examining the models' performance in identifying complex patterns within varied feature spaces. Results The outcome of this study reveals that the enhanced Support Vector Machine (SVM) model performed the highest compared to the other models, demonstrating an impressive 97.6% accuracy. In the case of CNN, it achieved 95.76% for effectively identifying hierarchical features. The RNN showed a good accuracy of 92.3%, which was pretty adequate for sequential data treatment. The ANN achieved a high accuracy of 88.77%. These findings describe the differences and strengths of both models and have possible applications in brain tumor detection. Conclusion This study conclusively established how much potential emerged for machine learning models to improve the detection capabilities of brain tumors. Addressing a performance perspective, the enhanced SVM ranked first. Again, this is proof of its critical importance as a tool in accurate diagnostic medicine. Based on these findings, further development of machine learning techniques in neuro-oncology will lead to an increase in diagnostic accuracy and treatment outcomes. It lays the fundamental foundation for betterment in any predictive model to be made in the future.

Abstract: A brain tumour is described by the presence of abnormal cells in the brain's tissues. Brain tumours can be benign (not cancerous) or malignant (cancerous). The malignant brain tumour is one of the leading and common cancers in the world. There are two types of tumours, primary tumours that develop in the brain and secondary tumours that start in another region of the body and then spread to the brain. The precise identification of the size and location of a brain tumour is crucial in the diagnosis of a brain tumour and is often diagnosed with magnetic resonance imaging (MRI). This book chapter discusses the major types of brain tumours and the advancements in computeraided approaches for detecting brain tumours. The manuscript gives an overview of various recent machine learning and medical image processing approaches developed recently for the identification and classification of brain tumours. Several medical image dataset available for the research works in this domain is also briefed in this article. The major research challenges which can be addressed by the researchers in the domain of brain tumour detection are also discussed in this article. Method: This manuscript mainly discusses various medical image processing methods to detect brain tumors. Results: The manuscript discusses the efficiency of the existing schemes and the key areas where further improvement is required Conclusion: This manuscript gives an overview of various categories of brain tumors, existing methods to detect brain tumors, the critical challenges in this domain and the medical image dataset available for the study.