A5021297349- Brain Tumor Detection and Classification
- Radiomics and Machine Learning in Medical Imaging
- Glioma Diagnosis and Treatment
- Advanced MRI Techniques and Applications
- Medical Image Segmentation Techniques
- Advanced Neural Network Applications
- Functional Brain Connectivity Studies
- Artificial Intelligence in Healthcare
- Imbalanced Data Classification Techniques
- EEG and Brain-Computer Interfaces
- Data Mining Algorithms and Applications
- Advanced Neuroimaging Techniques and Applications
- Machine Learning in Materials Science
UCSI University
2020-2025
MIT World Peace University
2022
ORCID
2021
Tumor-related epilepsy is a prevalent condition in patients with gliomas. Accurate prediction of crucial for early treatment. This study aimed to evaluate the novel application eXtreme Gradient Boost (XGBoost) machine learning (ML) algorithm into radiomics model predicting preoperative tumor-related (PTRE). Its performance was compared 4 conventional ML algorithms, including least absolute shrinkage and selection operator (LASSO), elastic net, random forest, support vector machine. used four...
A stroke is a medical condition that causes brain damage by rupturing blood vessels. It can also happen if the brain's supply and other nutrients are cut off. leading cause of death disability worldwide., according to World Health Organization (WHO). potentially fatal illness primarily affects adults over age 65. Doctors devote significant amount time effort predicting strokes. As result., primary goal study use various Machine Learning approaches predict likelihood occurring using hyper...
This study showed an alternative and non-invasive method for measuring brainwaves using Magnetic Resonance Imaging (MRI) with a gradient echo - planar imaging (GE-EPI) sequence. An attempt was made to measure the axonal magnetic fields of delta theta waves direct detection MRI. Time-varying produce current which may induce field according Biot-Savart law. The MR scanner can detect inhomogeneous caused by weak currents generated in subject that interact main field, $B_{o}$ , scanner. Fifteen...
Tumor-related epilepsy (TRE) refers to the condition in which primary brain tumors cause recurring seizures. A model that classifies as epileptogenic or non-epileptogenic could improve prognosis and treatment methods for TRE. This study aims identify MRI sequences machine learning algorithms (MLAs) be used build most accurate tumor classification model. T1W, T2W, T2W FLAIR T1W contrast-enhanced scans were acquired from 24 glioma patients, 8 with 16 without pre-operative epilepsy. total of 88...
20% of brain tumor patients present with seizures at the onset diagnosis, while a further 25-40% develop epileptic as progresses. Tumor-related epilepsy (TRE) is condition in which causes recurring, unprovoked seizures. The occurrence TRE differs between patients, along effectiveness treatment methods. Therefore, determining properties that correlate can help guide treatment. This article reviews MRI sequences and image post-processing algorithms study TRE. It focuses on caused by glioma...
Accurate segmentation is useful for diagnosing and treating brain tumors. However, manual can be difficult time-consuming, even experienced radiologists. In recent years, convolutional neural networks (CNN) have been deployed to perform this task automatically. CNNs are usually computationally expensive or require a long time train. This study aims identify the optimum hyperparameters constructing an efficient 3D U-Net model segmenting tumors from magnetic resonance images. has constructed...
The purpose of this study is to detect brainwave and physiological signals from MR images by applying Fourier transform Shannon's sampling theorem. Ten datasets were obtained 10 subjects using the gradient echo - planar imaging (GE-EPI) sequence with a 3.0 T Phillips scanner. 1500 on an axial plane consisting eyeballs acquired for each subject. MATLAB was used image processing. time frame being transformed produce frequency spectral. Then, bandpass filter applied out unwanted in order...
Tumor segmentation algorithms can aid in prognosis and treatment, are a better alternative to manual segmentation. This study combined thresholding, morphological operations k-means create new algorithm called 3D multimodal (3D-MKM) for segmenting tumors. used the fast spoiled gradient (FSPGR), T2 weighted spin echo (T2-FSE), fluid-attenuated inversion recovery (T2-FLAIR) contrast enhanced FSPGR (C-FSPGR) as input images. It adjusted histograms of each sequence highlight tumor regions, then...