← Back to cohort

Mohib Hameed

FAST · 2019 · i19 - 1689

—

Phone

—

GitHub

—

Academic

Program

—

CGPA

—

Year

2019

Education

—

Address

—

DOB

—

Career

Current role

—

Target role

—

Skills

Python, PyTorch, VTK, Plotly-Dash, Google Cloud Platform

Verbatim text

The exact text the LLM saw on the page (or the booklet text from the old import). This is what powers semantic search.

VizBriTS (Visualization for Brain Tumor Segmentation)
A visualization web application for Brain Tumor Segmentation and 3D visualization of tumor and its sub regions (edema, enhancing, non-enhancing). We have trained the image segmentation neural model (U-Net 2D) from scratch on BRATS (Brain Tumor Segmentation Dataset 2020); the trained model then accepts raw MRI modalities (flair, t2 & t1ce) and generates tumor segmentation masks, highlighting pixels occupying the tumor and its sub regions. These 2D segmentation masks then move to the 3D rendering pipeline which generates a 3D spatial volume of the entire tumorous region inside the brain.
Features include:
• Interactive and Controllable 3D visualizer of the brain and tumorous part, with option to select required tumor regions to view in 3D.
• 2D visual animator and slicer for segmentation masks overlay-ed on raw MRI slices for two separate axis, namely axial and coronal views.
• Statistical visualization showing percentages of tumor and sub regions with respect to whole tumor and with respect to entire brain.
TEAM MEMBERS: TAIMUR MUHAMMAD KHAN 19-1659, MOHIB HAMEED 19-1689, MUHAMMAD USAMA SALEEM 19-1901
SUPERVISOR: DR. AHMAD RAZA SHAHID
CO-SUPERVISOR: DR. FAISAL CHEEMA
FAST NUCES, ISLAMABAD
VizBriTS VISUALIZATION FOR BRAIN TUMOR SEGMENTATION
OBJECTIVE: A web application for Brain Tumor segmentation and 3-D Visualization of tumor and its sub-regions.
WORKFLOW: Raw MRI scans -> Trained U-Net Architecture -> 3D Brain Tumor Render -> Tumor & sub-regions segmentation / 2D segmentation maps
TIMELINE: SEP-OCT/22 Data Processing and Deep Learning pipeline, NOV-DEC/22 Model Training and 2D Visualizations, JAN-MAR/23 3D visualization and Rendering Pipeline, APR-MAY/23 App Development and Hosting
TOOLS & TECHNOLOGIES: Python, PyTorch, VTK, Plotly, Dash, Google Cloud
Technology Used: Python, PyTorch, VTK, Plotly-Dash, Google Cloud Platform
Supervisor Name: Dr. Ahmad Raza Shahid
Group Members: Taimur Muhammad Khan (i19 - 1659), Mohib Hameed (i19 - 1689), Muhammad Usama Saleem (i19 - 1901)

AI enrichment

Mohib Hameed is a student who contributed to a university capstone project involving a web application for brain tumor segmentation and 3D visualization. The project utilized U-Net for image segmentation and integrated VTK and Dash for interactive 3D rendering and statistical analysis.

Skills (AI)

["Python", "PyTorch", "U-Net", "VTK", "Plotly-Dash", "Google Cloud Platform", "Deep Learning", "Image Segmentation", "3D Visualization"]

Status: ai_done

Provenance

Source file: FAST - School of Computing -Graduate Directory-2023.pdf
From job #14 page 511
Created: 1778170171