How ghanshyamdhamat/medgemma_impact_challenge works
UniRad3s is a Gradio radiology application for MRI patient review, lesion diagnosis and segmentation, analysis, and report generation. Its documented workflow runs from NIfTI upload and patient selection through interactive MedSAM2 prompts and slice propagation to measurement, parcellation, and clinical and patient reports. The README supplies the product-level workflow; SOURCE INDEX references support selected model relationships. The large UI and pipeline sources were not excerpted, so internal wiring beyond those references is represented only where documented.
Main entry points: app_unirad3s.py Gradio Application · sam2_base.py SAM2 Model Core · sam2_video_predictor.py Slice Propagation · medsgilip_finetuning.py Scan Diagnosis
Components 22
Radiology Workflow
- Gradio Application
unirad3s/app_unirad3s.py - Patient Records
unirad3s/app_unirad3s.py - Batch Pipeline
unirad3s/pipeline.py - Volume to Video
unirad3s/app_unirad3s.py - Interactive Prompts
unirad3s/app_unirad3s.py - Slice Propagation
sam2/sam2_video_predictor.py
AI Models
- Scan Diagnosis
unirad3s/medsgilip_finetuning.py
Analysis and Reporting
- Lesion Measurements
unirad3s/app_unirad3s.py - Parcellation
parcellations/parcellation_by_registration.py - NIfTI Mask Export
unirad3s/app_unirad3s.py - Report Generation
unirad3s/simplify_report.py
Segmentation Engine
- SAM2 Builder
sam2/build_sam.py - SAM2 Model Core
sam2/modeling/sam2_base.py - Prompt Encoder
sam2/modeling/sam/prompt_encoder.py - Mask Decoder
sam2/modeling/sam/mask_decoder.py - EfficientTAM Builder
efficient_track_anything/build_efficienttam.py - EfficientTAM Model
efficient_track_anything/modeling/efficienttam_base.py - EfficientTAM Video Tracker
efficient_track_anything/efficienttam_video_predictor.py
Other components
- Clinician
- NIfTI Scans
- MedSigLIP
- MedGemma
Connections 20
- Clinician → Gradio Application: uses
README.md - NIfTI Scans → Gradio Application: uploads
README.md - Gradio Application → Patient Records: organizes
README.md - Gradio Application → Scan Diagnosis: requests diagnosis
README.md - Scan Diagnosis → MedSigLIP: analyzes scans
README.md - Gradio Application → Volume to Video: prepares volume
README.md - Gradio Application → Interactive Prompts: accepts prompts
README.md - Interactive Prompts → Slice Propagation: starts propagation
README.md - Slice Propagation → SAM2 Builder: uses model
sam2/sam2_video_predictor.py - SAM2 Builder → SAM2 Model Core: builds
sam2/build_sam.py - SAM2 Model Core → Prompt Encoder: uses
sam2/modeling/sam2_base.py - SAM2 Model Core → Mask Decoder: uses
sam2/modeling/sam2_base.py - Gradio Application → Lesion Measurements: shows measurements
README.md - Gradio Application → Parcellation: requests grounding
README.md - Gradio Application → NIfTI Mask Export: exports masks
README.md - Gradio Application → Report Generation: requests reports
README.md - Report Generation → MedGemma: generates reports
README.md - Gradio Application → Batch Pipeline: offers batch workflow
README.md - EfficientTAM Video Tracker → EfficientTAM Builder: loads model
efficient_track_anything/efficienttam_video_predictor.py - EfficientTAM Builder → EfficientTAM Model: constructs
efficient_track_anything/build_efficienttam.py