Advanced searches left 3/3

Brain Tumor - DOAJ

Summarized by Plex Scholar
Last Updated: 10 June 2022

* If you want to update the article please login/register

On the Performance of Deep Transfer Learning Networks for Brain Tumor Detection Using MR Images

"A brain tumor must be detected early in the development of the disease, or it may result in a chronic disease that cannot be treated if it is progressed. " A precise diagnosis of brain tumor can help to begin the right therapy, which ultimately reduces the patient's survival rate. Brain tumor detection from 2D Magnetic Resonance images is a commonly used technique for brain tumor detection. Using a combination of traditional classifiers to detect the brain tumor, several transfer learning based deep learning techniques are investigated in this paper. Cohen’s kappa, AUC, Jaccard, and Specificity were all tested to determine the correct results in terms of accuracy, precision, recall, F1-score, and specificity. The findings in this review are expected to be helpful in the search of a suitable method for deep transfer learning-based brain tumor detection. ".

Source link: https://doi.org/10.1109/ACCESS.2022.3179376


Brain Tumor Classification Based on Attention Guided Deep Learning Model

"Abstract Cancer is the second leading cause of death worldwide. " Brain tumors account for one out of every four cancer deaths. The convolutional neural network is one of the most common neural network frameworks for classifying images. To address the above problems, this paper introduces a novel brain tumor classification scheme that incorporates an attention sensor and a multipath network. While dismissing irrelevant information, an attention system is used to select the necessary data pertaining to the target area while leaving out irrelevant information. The data is broadcast to multiple channels by a multipath network before converting each channel and merging the results of all branches. ".

Source link: https://doi.org/10.1007/s44196-022-00090-9


Brain tumor image generation using an aggregation of GAN models with style transfer

"An interesting use of deep learning is synthetic data processing, particularly in the field of medical image analysis. " Another reason for using data augmentation strategies is class mismatch. Synthetic image generation in a variety of fields is possible thanks to generative Adversarial Networks. To this end, we have developed AGGrGAN, a aggregation of three base GAN models, with a Wasserstein GAN and a Deep Convolutional Generative Adversarial Network to produce synthetic MRI scans of brain tumors. Our suggested model effectively addresses the data unavailability of data availability and ability to recognize the information difference in several versions of the raw images. We've carried out all of the experiments on the two publicly available datasets, the brain tumor registry study and the Multimodal Brain Tumor Segmentation Challenge 2020 datasets. According to the two studies, the new model can produce fine-quality images with maximum Structural Similarity Index Measure scores of 0. 57 and 0. 83.

Source link: https://doi.org/10.1038/s41598-022-12646-y


A Novel Brain Tumor Detection and Coloring Technique from 2D MRI Images

"The first automated diagnosis of brain tumors in MRI scans is a difficult task. " For a long time, continuing research efforts have suggested a new strategy of replacing various grayscale anatomic regions of diagnostic images with appropriate colors that might solve the radiologists' challenges. The conversion of grayscale photographs into high-contrast color photos is proving difficult for raising various regions' u2019 contrasts. This paper explores common methods of separating different types of tissues within normal and abnormal regions by assigning colors to grayscale human brain MR images to distinguish different types of tissues. In addition, the colorization scheme based on luminance and pixel matrix after segmentation and ROI selection is also profitable due to increased PSNR and SSIM values, as well as visible contrast enhancement. Our new algorithm uses less processing overhead and takes less time than those of the industry's previously used color transfer system. ".

Source link: https://doi.org/10.3390/app12115744


Factors associated with delayed diagnosis among Filipino pediatric brain tumor patients: a retrospective review

"Aim: Determine delayed diagnosis based on the prediagnostic symptomatic interval among Filipino pediatric brain tumor patients and determine causal causes" are among the causes. Methods: From 2015 to 2019, data on Philippine General Hospital pediatric brain tumor patients was collected retrospectively. Conclusion: The delayed diagnosis of brain tumor patients in the Philippines is often related to age, tumor formation, and signs that are unusual in this condition.

Source link: https://doi.org/10.2217/cns-2022-0009


Segmentation for Multimodal Brain Tumor Images Using Dual-Tree Complex Wavelet Transform and Deep Reinforcement Learning

"The results show that the algorithm in this paper can effectively remove multimodal brain tumor image noise, and the segmented image has excellent detail and edges, and the segmented image has high similarity to the original photograph. ".

Source link: https://doi.org/10.1155/2022/5369516


Surgical Treatment of Radiation-Induced Late-Onset Scalp Wound in Patients Who Underwent Brain Tumor Surgery: Lessons from a Case Series

"The treatment of late-onset scalp wounds after radiation is painful, particularly in patients with a medical history of intracranial neoplasms. " We discuss our clinical experience with radiation-induced scalp wounds in this section, as well as a surgical approach for their care. The medical records of 13 patients with brain tumors who were treated for intractable scalp wounds after irradiation between January 2000 and August 2015 were retrospectively reviewed. Based on the u201d construction ladder, u201d, and according to the status of bone flap and scalp tissue, surgical treatment for a late-onset scalp wound was decided. In 11 patients, the bone flap was removed, allowing complete wound healing. 3 patients underwent cranioplasty using artificial components, but 2 patients were referred to hospitalization due to recurrent wound infections. Postirradiation scalp wounds are painful to handle and pose a significant risk of recurrence. The bone flap should be removed if osteoradionecrosis is suspected, and if osteoradionecrosis is suspected. ".

Source link: https://doi.org/10.1155/2022/3541254


Brain Tumor Segmentation from Magnetic Resonance Image using Optimized Thresholded Difference Algorithm and Rough Set

"At the first level, i. e. , an overlay image is created, which is the intensity average of all the brain area pixels segmented in the initial stage. " "Then is the second level of the thresholded difference between the brain area and the overlay image depending on the specified threshold. ".

Source link: https://doi.org/10.18421/TEM112-17


Znet: Deep Learning Approach for 2D MRI Brain Tumor Segmentation

"Background: Detection and segmentation of brain tumors using MR images is a difficult and rewarding task in the medical field. " Early diagnosis and localization of brain tumors can save lives and help physicians choose effective treatment strategies. Researchers in medical imaging have been attracted by deep learning methodologies due to their flexibility, success, and potential to aid in accurate diagnosis, prognosis, and medical therapy technologies. This paper introduces a novel method for separating 2D brain tumors in MR images by using deep neural networks and data processing techniques. Conclusions: Our experimental results revealed high values of the mean dice similarity coefficient. In MR images, the ZNet model’s ability to localize and auto-segment brain tumors is shown by the results and analysis of the DNN-derived tumor masks. Conclusion: In MR images, we can demonstrate the ability of deep learning algorithms and the new Znet framework to detect and segment tumors. In addition, pixel accuracy analysis may not be a valid evaluation tool for semantic segmentation in the case of class inconsistency in MR images segmentation. This is because the background is the most common style in ground truth photos. Consequently, a high degree of pixel accuracy can be misleading in certain computer vision applications. This paper presents a concrete example of AI applications in medical imaging that may be used as a tool for auto-segmentation of tumors in MR images.

Source link: https://doi.org/10.1109/JTEHM.2022.3176737


Case Report: Acquired Generalized Anhidrosis Caused by Brain Tumor: Review of the Literature

"Purpose" is the word that has been criticized in patients with brain tumors. Two patients have generalized anhidrosis caused by germinoma. And in winter, we also reviewed previous studies of generalized anhidrosis due to a brain tumor. Case Reports. a 12-year-old boy had regular heat shock-like episodes. Patient 2 was a 12-year-old girl with growth hormone deficiencies and generalized anhidrosis. After a germinoma treatment, all patients had germinoma and continued to need hormone replacement therapy. The hypothalamus was suspected in two patients with incomplete recovery of sweating, according to one patient with complete recovery, but there was no evidence of apparent hypothalamic involvement. One patient's sweating was not described, and catheterization in the hypothalamus may be relevant to incomplete sweating.

Source link: https://doi.org/10.3389/fendo.2022.877715

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions