A5040830959- Advanced Neural Network Applications
- Brain Tumor Detection and Classification
- Medical Image Segmentation Techniques
- AI in cancer detection
- Neonatal and fetal brain pathology
- Advanced Image and Video Retrieval Techniques
- Cell Image Analysis Techniques
- Lung Cancer Diagnosis and Treatment
- Advanced Radiotherapy Techniques
- Obstructive Sleep Apnea Research
- Face and Expression Recognition
- Non-Invasive Vital Sign Monitoring
- Sleep and Work-Related Fatigue
- Medical Imaging Techniques and Applications
Jadavpur University
2019-2024
Hippocampus (HC) segmentation plays a key role in diagnosis of predominant neuro-degenerative diseases like Alzheimer's, Parkinson's and common neurological disorders Epilepsy. In this paper, we propose solution to the 3D HC problem from MRI data using shape driven loss function attention Unet. particular, Histogram Oriented Gradients (HOG) based formulation is developed extract features. We suggest pooling technique as substitute histogram calculation for HOG. This address that not...
One of the most common sleep-related disorder is Obstructive Sleep Apnea Syndrome (OSAS). Increased upper airway resistance during sleep causes partial or full airflow interruptions. In stroke patients, severe OSAS increases risk mortality, neurological impairments, functional result after rehabilitation, and uncontrolled hypertension, making identification treatment crucial. Polysomnography best test. this work, it proposed to use PSG signals identify different sub-types OSAS. Although many...
Brain tumor segmentation plays a key role in diagnosis and surgical planning. In this paper, we propose solution to the 3D brain problem using deep learning graph cut from MRI data. particular, probability maps of voxel belong object (tumor) background classes UNet are used improve energy function cut. We derive new expressions for data term, region term weight factor balancing individual voxels our proposed model. validate performance model on publicly available BRATS 2018 dataset. Our...
In this paper, we address the problem of brain tumor classification from radiology and histopathology data. A coarse-to-fine approach is adopted using a combination deep features Graph Convolution Network (GCN). As first coarse step, use 3D CNN to detect Glioblastoma MRI images. order infer about Astrocytoma Oligodendroglioma, Whole Slide Images (WSI) are employed in second stage. During fine stage, 2D extracted at two different (global local) magnification levels. graph constructed with...