Meningioma Tumor Research Articles

A randomized control trial for evaluation of transfusion related immuno-modulation in patients with meningioma

BackgroundBlood transfusion necessity in neurosurgery varies based on surgical type, blood loss, and patient anemia. Leukocytes in red blood cells (RBCs) component release pro-inflammatory cytokines during storage, contributing to transfusion-related immunomodulation (TRIM). Our aim was to examine the impact of the leukocyte content in transfused PRBCs on patients undergoing neurosurgery for meningioma tumours. Study design and methodsThis prospective randomized controlled trial conducted from 2018 to 2020 by dividing patients randomly into non-leukoreduced (NLR) (n = 65) and leuko-reduced (LR) (n = 65) groups based on PRBCs received during surgery and hospital stay. Hospital and ICU stays, mechanical ventilation duration, and postoperative bacterial infections were observed. Hematological parameters and cytokine levels (IL-10, INF-gamma, and FAS-L) were assessed at pre-transfusion, 24 h, and 7 days post-transfusion. Data analysis included Mann-Whitney U test, Friedman test, Fisher's chi-square test, with statistical significance at p < 0.05. ResultsIn our study, ICU and hospital stay duration showed no significant difference (p = 0.06) between groups. However, NLR group had longer mean mechanical ventilation (18 ± 40.1 h) than the LR group (12.8 ± 8.6 h). Both groups showed statistically significant increase in Fas-L level on days 1 and 7 (p < 0.05). The IL-10 levels rose 43% in the NLR group, while and decreased by 7% the LR group on day 1. On day 7, IL-10 increased by 75% in NLR and decreased by 40% in LR, with no significance (p > 0.05). ConclusionIn conclusion, leukoreduction appeared to offer some immune response protection in term of reducing mechanical ventilation timings and cytokine level changes.

Non-mitotic proliferation of malignant cancer cells revealed through live-cell imaging of primary and cell-line cultures

IntroductionAnti-mitosis has been a key strategy of anti-cancer therapies, targeting at a fundamental property of cancer cells, their non-controllable proliferation due to overactive mitotic divisions. For improved anti-cancer therapies, it is important to find out whether cancer cells can proliferate independent of mitosis and become resistant to anti-mitotic agents.ResultsIn this study, live-cell imaging was applied to both primary-cultures of tumor cells, and immortalized cancer cell lines, to detect aberrant proliferations. Cells isolated from various malignant tumors, such as Grade-III hemangiopericytoma, atypical meningioma, and metastatic brain tumor exhibit distinct cellular behaviors, including amoeboid sequestration, tailing, tunneling, nucleic DNA leakage, as well as prokaryote-like division such as binary fission and budding-shedding, which are collectively referred to and reported as ‘non-mitotic proliferation’ in this study. In contrast, benign tumors including Grade-I hemangiopericytoma and meningioma were not obvious in such behaviors. Moreover, when cultured in medium free of any anti-cancer drugs, cells from a recurrent Grade-III hemangiopericytoma that had been subjected to pre-operation adjuvant chemotherapy gradually shifted from non-mitotic proliferation to abnormal mitosis in the form of daughter number variation (DNV) and endomitosis, and eventually regular mitosis. Similarly, when treated with the anti-cancer drugs Epirubicin or Cisplatin, the cancer cell lines HeLa and A549 showed a shift from regular mitosis to abnormal mitosis, and further to non-mitosis as the dominant mode of proliferation with increasing drug concentrations. Upon removal of the drugs, the cells reversed back to regular mitosis with only minor occurrences of abnormal mitosis, accompanied by increased expression of the stem cell markers ALDH1, Sox, Oct4 and Nanog.ConclusionsThe present study revealed that various types of malignant, but not benign, cancer cells exhibited cellular behaviors indicative of non-mitotic proliferation such as binary fission, which was typical of prokaryotic cell division, suggesting cell level atavism. Moreover, reversible transitions through the three modes of proliferation, i.e., mitosis, abnormal mitosis and non-mitosis, were observed when anticancer drug concentrations were grossly increased inducing non-mitosis or decreased favoring mitosis. Potential clinical significance of non-mitotic proliferation in cancer drug resistance and recurrence, and its relationship with cancer stem cells are worthy of further studies.

Lumbar clear cell meningioma mimicking schwannoma 7 years after resection of the same type of intracranial tumor: a case report

BackgroundMeningioma is the second most common intradural extramedullary tumor, following schwannoma. Meningioma is primarily categorized as benign World Health Organization grade 1, but clear cell meningioma is grade 2 of the intermediate malignant category. Clear cell meningiomas are rare, accounting for less than 1% of all meningioma tumors. There is no previous report of multiple intraspinal clear cell meningiomas without dural attachment.Case presentationA 27-year-old Asian male patient presented with lower right extremity pain, and had undergone tumor resection for intracranial clear cell meningioma 7 years previously, with re-resection and radiotherapy for local tumor recurrence at our hospital’s department of neurosurgery being carried out 4 years previously. No recurrence was observed since then. Preoperative lumbar magnetic resonance imaging showed two tumors at the L1 and L4 levels, both mimicking schwannoma with well-defined margins, no dural tail sign and homogeneous internal contrast. Intraoperative findings on tumor resection showed two tumors contiguous with the right L2 and L5 roots, which were not attached to the dura mater, similar to a schwannoma. After gross total resection, the postoperative pathology revealed no nuclear SMARCE1 antibody staining. The patient was diagnosed with clear cell meningioma. The patient’s postoperative course went well, with no symptoms of nerve dropout and no recurrence 2 years after surgery. In this case, both lumbar lesions were well demarcated and spherical in shape, occurring with single roots. Tumor characteristics suggested a primary rather than a metastatic lesion. Clear cell meningioma is characterized by a SMARCE1 mutation and is different from other types of meningiomas.ConclusionTo the best of our knowledge, this is the first report of multiple intraspinal clear cell meningiomas without dural attachment at the lumbar spine after resection of intracranial clear cell meningioma. We speculate that the two tumors were de novo lesions on the basis of the features of the tumors, although they were detected 7 years after the resection of intracranial clear cell meningioma.

Metastasis infiltrating tumor to meningioma: a case report

BackgroundThere have been many reports of tumor-to-tumor metastasis, in which cancer metastasizes directly into meningiomas. However, metastasis infiltrating tumors in which cancer metastasizes around meningiomas are rare. Therefore, we report a case of metastasis originating from lung cancer that infiltrated meningioma.Case presentationA 79-year-old Japanese woman underwent head magnetic resonance imaging for brain metastasis screening before lung cancer surgery. At that time, asymptomatic meningioma of the left frontal region was accidentally found. Magnetic resonance imaging 6 months later revealed a lesion suspected to be a metastatic brain tumor close to the meningioma. Brain tumor resection was performed, and histopathological diagnosis was meningioma and metastatic brain tumor. Metastatic cancer had invaded the meningioma at the boundary between the brain tumor and metastasis.ConclusionsA sudden change in imaging findings on routine examination of meningiomas in patients with lung carcinoma may indicate a metastatic brain tumor. The form of cancer metastasis to meningioma is not limited to tumor-to-tumor metastasis, but also includes metastasis infiltrating tumors near the meningioma.

Brain tumor detection from images and comparison with transfer learning methods and 3-layer CNN

Health is very important for human life. In particular, the health of the brain, which is the executive of the vital resource, is very important. Diagnosis for human health is provided by magnetic resonance imaging (MRI) devices, which help health decision makers in critical organs such as brain health. Images from these devices are a source of big data for artificial intelligence. This big data enables high performance in image processing classification problems, which is a subfield of artificial intelligence. In this study, we aim to classify brain tumors such as glioma, meningioma, and pituitary tumor from brain MR images. Convolutional Neural Network (CNN) and CNN-based inception-V3, EfficientNetB4, VGG19, transfer learning methods were used for classification. F-score, recall, imprinting and accuracy were used to evaluate these models. The best accuracy result was obtained with VGG16 with 98%, while the F-score value of the same transfer learning model was 97%, the Area Under the Curve (AUC) value was 99%, the recall value was 98%, and the precision value was 98%. CNN architecture and CNN-based transfer learning models are very important for human health in early diagnosis and rapid treatment of such diseases.

Stereotactic radiosurgery in India: A nationwide survey of technology and quality assurance practices.

India is rapidly adopting advanced treatments like Stereotactic Radiosurgery (SRS). However, there is a paucity of data on SRS practice. The aim of study is to assess the current status of technology and practices of machine quality assurance (QA) and patient specific quality assurance for SRS in India. A survey questionnaire was designed using Google Forms and sent to chief/senior medical physicists across 220 radiotherapy centers in India on July 15, 2022. It contained questions on infrastructure availability, treatment planning, and QA. SRS was found to be extensively used for the treatment of brain metastases (99.3%), followed by meningioma (50.3%), acoustic neuroma (45.5%), and pituitary tumours (33.1%). The most commonly used photon energy and treatment technique were 6MV FFF and VMAT, respectively. A prescription isodose line ranging from 70% to 100% was selected by linac users. Most linac institutes verify pretreatment doses. There was a lack of uniformity in the analysis metrics such as Low Dose Threshold, Dose Difference, and Distance to Agreement. A survey revealed that the variety of SRS QA programs being followed at Indian radiotherapy centers. This is the first study to report the physics practice of SRS in India. The survey shows a need to carry out a postal dose audit for small static photon fields in India.

A mask R-CNN approach for detection and classification of brain tumours from MR images

ABSTRACT This study aims to tackle the implementation of a single deep learning (DL) technique to the brain tumour recognition, segmentation and classification problem based on MR images. We propose utilising the Mask R-CNN approach along with transfer learning models on images. This was done by assigning bounding boxes and successfully constructing a border on each image for the tumour volume in order to distinguish it from neighbouring tissues and structures. By processing the images in this way, a single DL model could accurately identify, classify and segment three different categories of brain tumours (meningioma, glioma or pituitary). ResNet-50 and ResNet-101 network architectures were trained for 1000 epochs, and their classification and segmentation performances were assessed. A higher classification success was achieved for the ResNet-101 backbone pretrained on COCO images (%75 accuracy) in the testing phase. ResNet-50, on the other hand, achieved a higher classification accuracy (%87) when used with the ImageNet dataset. The classification and segmentation successes of meningioma and pituitary tumours were comparable, while glioma tumours could be segmented at relatively lower success rates (41–73% for all models), despite similar classification performances (73–95%) with other tumour types.

Anterior transpetrosal approach and the tumor removal rate, postoperative neurological changes, and complications: experience in 274 cases over 33 years.

The authors report on the anterior transpetrosal approach (ATPA) and the results of surgeries performed over a 33-year period for petroclival tumors, including meningioma, trigeminal schwannoma, chordoma, and epidermoid tumor. They analyze early postoperative neurological changes, surgical complications, and trends over the decades. A retrospective analysis of 274 surgical cases that had undergone the ATPA from January 1984 to March 2017 was conducted. Data were collected from charts, clinical summaries, operative records, and operative videos. The analyzed parameters included patient diagnosis, tumor size, disease location, operation date, tumor removal rate, pre- and postoperative neurological symptoms (consciousness level, motor and sensory deficits of the limbs, sensory aphasia, and cranial nerve III-VIII injuries), surgical deaths, and radiologically recognized brain injuries after the operation (contusion, infarction, hemorrhage). Gross-total resection (GTR) was achieved in 53.5% of the 243 tumors with available data. The GTR rate for meningiomas (148 cases) was 54.1%. Trigeminal schwannomas had a high GTR rate of 87.1%, whereas chordomas had a low GTR rate of 14.3%. The rate of early neurological deterioration immediately after the ATPA, referred to as "early neurological change," was as follows: consciousness disturbance in 1.9% of cases (5 cases), improvement of hemiparesis in 45.0% of cases but deterioration in 8.1% of cases, sensory aphasia in 2.3% of cases due to temporal lobe injury, improvement of cerebellar symptoms in 39.3% of cases with rare deterioration (1.9% of cases), worsening of preoperative diplopia in 49.4% of patients and rarely improving, improvement of trigeminal symptoms in 19.1% of cases (mostly trigeminal neuralgia) among the 43.7% of patients who had them preoperatively, and deterioration of facial hypesthesia and/or paresthesia in 27.4% of cases. Early neurological deterioration was monitored in 183 patients for 6 months to determine the surgical complications of ATPA. Consciousness disturbance recovered in half of the cases but persisted in 3 (1.5%). Hemiparesis fully recovered in 63.2% of cases, resulting in a complication rate of 3.0%. The most frequent complication was diplopia (36.4%), with a complete remission rate of 26.4%. The second most frequent complication was facial hypesthesia (24.0%), with a recovery rate of 16.1%. Facial nerve palsy improved in 63.0% of cases and had a complication rate of 4.9%. Cerebellar symptoms showed complete recovery in all cases. The ATPA allows the removal of petroclival tumors extending into Meckel's cave and the middle fossa, making it preferred for dumbbell trigeminal schwannomas and meningiomas. However, the ATPA's aggressive tumor removal can risk a lower recovery of cranial nerve IV-VI deficits. For benign meningiomas, initial observation with regular follow-up is recommended. Surgery is appropriate for high-growth cases aiming for total removal, accompanied by a thorough explanation of the risks. If the risks are not accepted, subtotal removal can be considered, and radiosurgery is suggested for residual tumor.

Fractal-Based Analysis of Histological Features of Brain Tumors.

The structural complexity of brain tumor tissue represents a major challenge for effective histopathological diagnosis. Tumor vasculature is known to be heterogeneous, and mixtures of patterns are usually present. Therefore, extracting key descriptive features for accurate quantification is not a straightforward task. Several steps are involved in the texture analysis process where tissue heterogeneity contributes to the variability of the results. One of the interesting aspects of the brain lies in its fractal nature. Many regions within the brain tissue yield similar statistical properties at different scales of magnification. Fractal-based analysis of the histological features of brain tumors can reveal the underlying complexity of tissue structure and angiostructure, also providing an indication of tissue abnormality development. It can further be used to quantify the chaotic signature of disease to distinguish between different temporal tumor stages and histopathological grades.Brain meningioma subtype classifications' improvement from histopathological images is the main focus of this chapter. Meningioma tissue texture exhibits a wide range of histological patterns whereby a single slide may show a combination of multiple patterns. Distinctive fractal patterns quantified in a multiresolution manner would be for better spatial relationship representation. Fractal features extracted from textural tissue patterns can be useful in characterizing meningioma tumors in terms of subtype classification, a challenging problem compared to histological grading, and furthermore can provide an objective measure for quantifying subtle features within subtypes that are hard to discriminate.

Evaluating the clinical applicability of neural networks for meningioma tumor segmentation on 3D MRI

Magnetic resonance imaging (MRI), in both 2D and 3D configurations, serves as the leading non-invasive and non-ionizing technique for the detection and clinical assessment of brain tumors. 3D multiparametric MRIs (mpMRIs) offer enhanced spatial and biological context over other MRIs, which can help clinicians plan tailored treatment and therapy. Presently, surgical intervention remains the primary treatment method for brain tumors, requiring accurate segmentation of tumor tissue on MRI scans. However, manual tumor segmentation and traditional machine learning techniques are labor intensive and subject to observer bias. Deep learning models can provide enhanced accuracy and precision in tumor segmentation due to their ability to automatically extract intricate features and patterns from MRI scans. In this study, MRI scans from 358 meningioma patients were used to train a series of deep learning models to segment the tumor subregions. We hypothesized that a deep learning framework designed for meningioma tumor segmentation and trained utilizing 3D mpMRIs would demonstrate superior segmentation accuracy compared to a framework trained on 2D MRIs or single-sequence 3D MRIs. We compared the model predictions with the ground truth labels using similarity metrics to assess accuracy and clinical applicability. The model trained on 3D mpMRI scans showed reliable performance for meningioma subregion segmentation with a Dice Similarity Coefficient score of 0.91 and a median Sensitivity of 91.38%. It outperformed the other models due to the additional spatial context provided by 3D mpMRIs, suggesting it could be readily used in clinical practice to help guide treatment strategy.

Detection and classification of brain tumor using hybrid deep learning models

Accurately classifying brain tumor types is critical for timely diagnosis and potentially saving lives. Magnetic Resonance Imaging (MRI) is a widely used non-invasive method for obtaining high-contrast grayscale brain images, primarily for tumor diagnosis. The application of Convolutional Neural Networks (CNNs) in deep learning has revolutionized diagnostic systems, leading to significant advancements in medical imaging interpretation. In this study, we employ a transfer learning-based fine-tuning approach using EfficientNets to classify brain tumors into three categories: glioma, meningioma, and pituitary tumors. We utilize the publicly accessible CE-MRI Figshare dataset to fine-tune five pre-trained models from the EfficientNets family, ranging from EfficientNetB0 to EfficientNetB4. Our approach involves a two-step process to refine the pre-trained EfficientNet model. First, we initialize the model with weights from the ImageNet dataset. Then, we add additional layers, including top layers and a fully connected layer, to enable tumor classification. We conduct various tests to assess the robustness of our fine-tuned EfficientNets in comparison to other pre-trained models. Additionally, we analyze the impact of data augmentation on the model's test accuracy. To gain insights into the model's decision-making, we employ Grad-CAM visualization to examine the attention maps generated by the most optimal model, effectively highlighting tumor locations within brain images. Our results reveal that using EfficientNetB2 as the underlying framework yields significant performance improvements. Specifically, the overall test accuracy, precision, recall, and F1-score were found to be 99.06%, 98.73%, 99.13%, and 98.79%, respectively.

Topographic mapping of oxygen extraction and cerebral venous blood volume fraction in Brain tumors using quantitative BOLD: A case report

Background: While oxygen extraction fraction (OEF) reflects the underlying variations in cerebral brain oxygen metabolism, tissue voxels having elevated volume fraction of blood vessel network with deoxygenated blood, will apparently contribute to higher cerebral venous blood volume fraction (CVBVF). This Case report examines the difference in intra and peri-tumoral topographical patterns of OEF and CVBVF in cases of a meningioma tumor (Case-I) and a low- grade glioma (Case-II). Methods: Using a “static dephasing regime” BOLD model, we use the BOLD signal model containing parameters representing OEF and CVBVF. For each voxel in the region of interest, the parameters are solved by non-linearly fitting the signal model using paired differences between logarithms of the measured echo signal after inhomogeneity correction. Results: OEF and CVBVF maps in Case-I reveals an interesting phenomenon in the peritumoral parenchyma showing reduced OEF and increased CVBVF levels. The uniformly low CVBVF and elevated OEF in Case-II indicates that even with less density of vasculature, the region extracts higher amount of oxygen. Conclusion: While topographic mapping using qBOLD revealed elevated levels of intra-tumoral OEF for both cases, the pattern of CVBVF variation was uniformly low in Case-II.

Identification of meningioma tumor using recurrent neural networks

&lt;p&gt;By the calculations of national center for biotechnology information from COVID 19 pandemic, number of meningioma tumor patients are increasing in world. Identifying the meningioma tumor and its position in brain is not easy task by using deep neural networking based medical imaging. But it is needed to identify meningioma tumors in brain by using AI based medical imaging for the purpose of medical artificial intelligence technology innovation. Comparing to neural network results with recurrent neural network results can give accurate results. For identifying the patients’ present condition and prediction of future behavior by using recurrent neural network is need for us. Increase the accurate results for neural networking based medical imaging in health care is very expensive. By using recurrent neural networks (RNN) algorithm with many hidden layers for identification of tumor(s) in human brain with high accuracy by comparison of existing images in our data base with new unknown medical image with low cost. In this study first we are collecting the masks of skull from MRI image and dividing the masks to different types of datasets depending on age criteria like a child age, middle age and old age with two types male and female. Then we can get totally 6 types of datasets. All these masks of MRI images to binary imaging by using morphological erosion concept after that storing that masks in data sets then collect the new MRI image and comparing its mask part of skull with existing dataset in recurrent neural networks.&lt;/p&gt;

MEHW‐SVM multi‐kernel approach for improved brain tumour classification

AbstractThe human brain, the primary constituent of the nervous system, exhibits distinctive complexities that present considerable difficulties for healthcare practitioners, specifically in categorizing brain tumours. Magnetic resonance imaging is a widely favoured imaging modality for detecting brain tumours due to its extensive range of image characteristics and utilization of non‐ionizing radiation. The primary objective of the current investigation is to differentiate between three distinct classifications of brain tumours by introducing a novel methodology. The utilization of a combined feature extraction technique that integrates novel global grey level co‐occurrence matrix and local binary patterns is employed, thereby offering a comprehensive representation of the structural and textural information contained within the images. Principal component analysis is used to improve the model's efficiency for effective feature selection and dimensionality reduction. This study presents a novel framework incorporating four separate kernel functions, Minkowski–Gaussian, exponential support vector machine (SVM), histogram intersection SVM, and wavelet kernel, into a SVM classifier. The ensemble kernel employed in this study is specifically designed to classify glioma, meningioma, and pituitary tumours. Its implementation enhances the model's robustness and adaptability, surpassing the performance of conventional single‐kernel SVM approaches. This study substantially contributes to medical image classification by utilizing innovative kernel functions and advanced machine‐learning techniques. The findings demonstrate the potential for enhanced diagnostic accuracy in brain tumour cases. The presented approach shows promise in effectively addressing the intricate challenges associated with classifying brain tumours.

68 Ga-DOTATATE PET to Characterize Lesions in the Neuroaxis.

The differentiation of paragangliomas, schwannomas, meningiomas, and other neuroaxis tumors in the head and neck remains difficult when conventional MRI is inconclusive. This study assesses the utility of 68 Ga-DOTATATE PET/CT as an adjunct to hone the diagnosis. This retrospective study considered 70 neuroaxis lesions in 52 patients with 68 Ga-DOTATATE PET/CT examinations; 22 lesions (31%) had pathologic confirmation. Lesions were grouped based on pathological diagnosis and best radiologic diagnosis when pathology was not available. Wilcoxon rank sum tests were used to test for differences in SUV max among paragangliomas, schwannomas, and meningiomas. Receiver operator characteristic curves were constructed. Paragangliomas had a significantly greater 68 Ga-DOTATATE uptake (median SUV max , 62; interquartile range [IQR], 89) than nonparagangliomas. Schwannomas had near-zero 68 Ga-DOTATATE uptake (median SUV max , 2; IQR, 1). Intermediate 68 Ga-DOTATATE uptake was seen for meningiomas (median SUV max , 19; IQR, 6) and other neuroaxis lesions (median SUV max , 7; IQR, 9). Receiver operator characteristic analysis demonstrated an area under the curve of 0.87 for paragangliomas versus all other lesions and 0.97 for schwannomas versus all other lesions. Marked 68 Ga-DOTATATE uptake (>50 SUV max ) favors a diagnosis of paraganglioma, although paragangliomas exhibit a wide variability of uptake. Low to moderate level 68 Ga-DOTATATE uptake is nonspecific and may represent diverse pathophysiology including paraganglioma, meningioma, and other neuroaxis tumors but essentially excludes schwannomas, which exhibited virtually no uptake.

Dual-Level Augmentation Radiomics Analysis for Multisequence MRI Meningioma Grading.

Preoperative, noninvasive prediction of meningioma grade is important for therapeutic planning and decision making. In this study, we propose a dual-level augmentation strategy incorporating image-level augmentation (IA) and feature-level augmentation (FA) to tackle class imbalance and improve the predictive performance of radiomics for meningioma grading on Magnetic Resonance Imaging (MRI). This study recruited 160 consecutive patients with pathologically proven meningioma (129 low-grade (WHO grade I) tumors; 31 high-grade (WHO grade II and III) tumors) with preoperative multisequence MRI imaging. A dual-level augmentation strategy combining IA and FA was applied and evaluated in 100 repetitions in 3-, 5-, and 10-fold cross-validation. The best area under the receiver operating characteristics curve of our method in 100 repetitions was ≥0.78 in all cross-validations. The corresponding cross-validation sensitivities (cross-validation specificity) were 0.72 (0.69), 0.76 (0.71), and 0.63 (0.82) in 3-, 5-, and 10-fold cross-validation, respectively. The proposed method achieved significantly better performance and distribution of results, outperforming single-level augmentation (IA or FA) or no augmentation in each cross-validation. The dual-level augmentation strategy using IA and FA significantly improves the performance of the radiomics model for meningioma grading on MRI, allowing better radiomics-based preoperative stratification and individualized treatment.

Brain tumor classification and detection via hybrid alexnet-gru based on deep learning

Brain tumors are among the most dangerous types of brain cancer due to their aggressiveness. The development of aberrant brain tissue leads to brain tumors, which pose a major threat to people's health and welfare. Early detection of these malignancies is still fairly challenging, particularly when attempting to distinguish between various kinds including gliomas, meningioma, and pituitary tumors. The complexity of the brain's structure further increases the need for a speedy and accurate identification of irregularities. An innovative approach using a hybrid model that combines AlexNet and GRU (Gated Recurrent Unit) neural networks is presented to identify and characterize brain cancers using MRI data. The MRI images should first be sharpened and denoised using a non-local means filter to ensure the best input data. To avoid overfitting and achieve optimal model performance, the AlexNet architecture employs layers to extract specificfeatures from images. Model complexity and potential for generalization are regulated through hyperparameter modification. The GRU component resolves gradient vanishing in deep networks and uses the softmax activation function to categorize braintumorsinto four distinct classes. According to the findings, the model can categorize and diagnose brain cancers with 97% accuracy, 97.63% precision, a robust 96.78% recall rate, and an impressive 97.25% F1-Score. These findings show that theresearch has the potential to improve patient outcomes and create a more favorable healthcare environment in medical imaging and brain tumor detection.

NeuroML - Brain Tumor Classification using Machine Learning and Deep Learning

In this study we have proposed CNN model's efficacy in accurately classifying brain tumors into meningioma, glioma, and pituitary tumor categories, showcasing high sensitivity and specificity. The incorporation of deep learning techniques empowers the model to discern subtle and intricate patterns, contributing to heightened diagnostic precision. This study underscores the potential of advanced machine learning algorithms in medical imaging for specific brain tumor classification, offering a valuable tool for healthcare professionals. The research findings hold promise for improving the accuracy of neuro-oncological diagnoses, ultimately advancing patient care in the domain of brain tumor pathology. The study utilizes a diverse dataset comprising magnetic resonance imaging (MRI) scans of the brain, encompassing various tumor types and conditions. The preprocessing phase involves standardizing and augmenting the dataset to ensure optimal model training. A Convolutional Neural Network architecture is designed to automatically learn discriminative features from the input images, capturing intricate patterns indicative of tumor presence. Results demonstrate the proposed CNN model's efficacy in accurately classifying brain tumors into meningioma, glioma, and pituitary tumor categories, showcasing high sensitivity and specificity. The incorporation of deep learning techniques empowers the model to discern subtle and intricate patterns, contributing to heightened diagnostic precision. This study underscores the potential of advanced machine learning algorithms in medical imaging for specific brain tumor classification, offering a valuable tool for healthcare professionals. The research findings hold promise for improving the accuracy of neuro-oncological diagnoses, ultimately advancing patient care in the domain of brain tumor pathology. In our work, CNN gained an accuracy of 99.3 %, which is very compelling. The main aim of this paper is to distinguish between normal and abnormal pixels, based on texture based and statistical based features

EXTH-57. THE IGF-2 SIGNALING PATHWAY IS A PROMISING MENINGIOMA TREATMENT TARGET

Abstract Meningioma is the most common primary central nervous system tumor often causing serious complications while there is no effective medical treatment. We developed a tumor-derived primary culture model featuring tumor cells with retained PDG2S lineage marker as well as non-arachnoid cells mimicking the tumor cell microenvironment. We used this model to test the relevance of miRNA differentially expressed in meningioma tumor samples identified by small RNA sequencing. We found that in all meningiomas studied, there was activation of the IGF2/miR-483 locus with high expression of both miR-483-5p and IGF2. Inhibition of miR-483-5p reduced the growth of cultured meningioma cells, whereas a miR-483 mimic increased cell proliferation. Anti-IGF2 neutralizing antibodies reduced meningioma cell proliferation. Small molecule tyrosine kinase inhibitor blockade of the IGF2 receptor (IGF1R) resulted in rapid loss of viability of cultured meningioma tumor-derived cells. Autocrine IGF2 feed-back appears to be obligatory for meningioma tumor cell survival and growth. Inhibition of this circuit with IGF1R small molecule tyrosine kinase inhibitors or receptor-binding antibodies as well as anti-IGF2 neutralizing antibodies may have therapeutic potential for meningioma.

Privacy preserved collaborative transfer learning model with heterogeneous distributed data for brain tumor classification

AbstractCorrect identification of tumor in brain images is critical for treatment. In the medical domain, class distributions of recorded data could differ with locations and require high levels of privacy while collaboratively training the deep learning (DL) models for classifications. The main aim of this paper is to propose a privacy‐preserving collaborative model for the classification of brain tumor in heterogeneously distributed magnetic resonance imaging (MRI) images. In this paper, initially, an open‐source dataset has been acquired and analyzed as per the required competencies. The acquired dataset has four types of MRI images: pituitary tumor, meningioma tumor, glioma tumor, and no tumor. First, the acquired dataset was analyzed using DL and transfer learning algorithms. By applying implementations of basic algorithms, better algorithms were identified for further implementations in a federated learning ecosystem. DenseNet201‐based transfer learning was identified as a better neural network and further utilized for collaborative transfer learning implementations. Here, the paper also focused on developing a suitable system for a heterogeneous distributed tumor database. Heterogeneous data were converted from the available data by applying nonidentical data distribution. The study discovered that the federated DL models, involving multiple clients, exhibited superior performance compared to conventional pretrained models. The proposed framework possesses distinctive characteristics that distinguish it from existing classification methods for brain tumor identification, particularly in terms of ensuring data privacy for edge devices with limited resources. Due to these additional features, the framework stands as the optimal alternative solution for early diagnosis of brain tumor.