3D-visualNet

Combining 3D brain MRI and Tabular Data for 3D Tumor Segmentation , Tumor type , genomic analysis and prognosis prediction

Multi-Input 3D Brain Tumor Segmentation and radiogenomic Model

This repository contains code and architecture for a deep learning model that performs 3D brain tumor segmentation and radiogenomic analysis using multimodal inputs — 3D MRI volumes and tabular clinical data. The model also predicts five classification outputs related to patient-level diagnosis , prognosis and genomics .

---

🧠 Project Overview

• Goal:
  • Accurately segment brain tumors from 3D MRI scans.
  • Predict 5 classification outputs based on multimodal input.
• Inputs:
  • 3D MRI volumes (NIfTI format)
  • Tabular patient data (CSV)
• Outputs:
  • Binary 3D segmentation mask
  • Five classification heads ( tumor type,tumor grade, IDH status,MGMT status , relapse)

---