A5020145471- Brain Tumor Detection and Classification
- Advanced Neural Network Applications
- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Cell Image Analysis Techniques
- Advanced X-ray and CT Imaging
- Medical Imaging Techniques and Applications
- Generative Adversarial Networks and Image Synthesis
- Privacy-Preserving Technologies in Data
- Domain Adaptation and Few-Shot Learning
- Advanced Image and Video Retrieval Techniques
- Advanced Image Processing Techniques
- Abdominal Trauma and Injuries
- Glioma Diagnosis and Treatment
- Pelvic and Acetabular Injuries
- Digital Imaging for Blood Diseases
- COVID-19 diagnosis using AI
- Colorectal Cancer Screening and Detection
- Image and Signal Denoising Methods
- Trauma and Emergency Care Studies
- Hepatocellular Carcinoma Treatment and Prognosis
- Image Retrieval and Classification Techniques
- Blind Source Separation Techniques
- Computational Drug Discovery Methods
Queen's University
2021-2025
Qazvin Islamic Azad University
2017-2025
Memorial Sloan Kettering Cancer Center
2025
The University of Texas MD Anderson Cancer Center
2025
Islamic Azad University, Tehran
2023
Shenzhen University Health Science Center
2020
Shenzhen University
2020
Iran University of Science and Technology
2009
Accurate brain tumour segmentation is critical for tasks such as surgical planning, diagnosis, and analysis, with magnetic resonance imaging (MRI) being the preferred modality due to its excellent visualisation of tissues. However, wide intensity range voxel values in MR scans often results significant overlap between density distributions different tissues, leading reduced contrast accuracy. This paper introduces a novel framework based on conditional generative adversarial networks (cGANs)...
Multi-detector contrast-enhanced abdominal computed tomography (CT) allows for the accurate detection and classification of traumatic splenic injuries, leading to improved patient management. Their effective use requires rapid study interpretation, which can be a challenge on busy emergency radiology services. A machine learning system has potential automate process, potentially faster clinical response. This aimed create such system.
Establishing the reproducibility of radiomic signatures is a critical step in path to clinical adoption quantitative imaging biomarkers; however, must also be meaningfully related an outcome importance value for per- sonalized medicine. In this study, we analyze both and prognostic features extracted from liver parenchyma largest metastases contrast enhanced CT scans patients with colorectal (CRLM). A prospective cohort 81 two major US cancer centers was used establish images reconstructed...
Abstract Background Chronic rhinosinusitis (CRS) is diagnosed with symptoms and objective endoscopy or computed tomography (CT). The Lund–Mackay score (LMS) often used to determine the radiologic severity of CRS make clinical decisions. This proof-of-concept study aimed develop an automated algorithm combining a convolutional neural network (CNN) for sinus segmentation post-processing compute LMS directly from CT scans. Results Radiology Information System was queried outpatient paranasal...
In this paper a brain tumor segmentation method is proposed which based on the Random Forest algorithm. The technique applied to magnetic resonance images and performance indices including Dice Similarity Coefficient (DSC) as well algorithm accuracy (ACC) are calculated that 98.38% 97.65%, respectively. obtained results show model can have good when compared with other methods. Besides, in mathematical modeling of provided.
"Just Accepted" papers have undergone full peer review and been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, proof before it is published its final version. Please note that during production of the copyedited article, errors may be discovered which could affect content. ©RSNA, 2023
Leukocytes are categorized into five groups in a peripheral blood microscopic image. A differential count of these various types cells is used to determine the presence an infection human body. Precise boundary leukocyte and cytoplasm necessary for extracting features them. In fact texture, color, size morphology nucleus make differences among different kinds leucocytes. Otsu method has been thresholding order detect precise nucleuses image Active contour employed find boundaries Cytoplasm...
Brain lesion segmentation is crucial for diagnosis, surgical planning, and analysis. Owing to the fact that pixel values of brain lesions in magnetic resonance (MR) scans are distributed over wide intensity range, there always a considerable overlap between class-conditional densities lesions. Hence, an accurate automatic still challenging task. We present novel architecture based on conditional generative adversarial networks (cGANs) improve contrast segmentation. To this end, we propose...
There are many different classes of leukocyte in peripheral blood image. Leukocyte count is used to determine the presence an infection human body. To be able observe and recognize kinds leukocyte, you must stain them. For this purpose, normally Giemsa used. two difficult issues image segmentation which common algorithms can not overcome Nucleus laid inside white cell darkest part cells. Since staining done by humans, intensity images slightly from each others. Neutrophils leukocytes have...
Magnetic resonance imaging (MRI) provides varying tissue contrast images of internal organs based on a strong magnetic field. Despite the non-invasive advantage MRI in frequent imaging, low MR target area make segmentation challenging problem. This paper demonstrates potential benefits image-to-image translation techniques to generate synthetic high (HTC) images. Notably, we adopt new cycle generative adversarial network (CycleGAN) with an attention mechanism increase within underlying...
One of the purposes from segmenting brain tissues is to separate damaged tissue in patient's brain. In fact, segmentation one essential steps detection and treatment abnormalities. This time-consuming task usually performed by clinical experts who are not errorless. The proposed method this paper automate tumor with aim making process more complete closer treatments. We propose a novel that combination neural networks active contours automatically segment gliomas MRI multi-modalities images....
Small liver lesions common to colorectal metastases (CRLMs) are challenging for convolutional neural network (CNN) segmentation models, especially when we have a wide range of slice thicknesses in the computed tomography (CT) scans. Slice thickness CT images may vary by clinical indication. For example, thinner slices used presurgical planning fine anatomic details small vessels required. While keeping effective radiation dose patients as low possible, various employed CRLMs due their...