PopularMajor Project
Artificial Intelligence
Face Recognition Attendance System
Camera-based attendance with liveness detection and analytics
- Python
- OpenCV
- FaceNet
- Flask
- +2
₹4,900₹5,999
4.8/301 reviews
Artificial Intelligence · Capstone
Liver Lesion Segmentation (Tumour, Cyst, Abscess)
U-Net segmentation of the liver and focal lesions in abdominal CT
5-6 weeks build
Overview
A two-stage pipeline that first segments the liver in CT slices, then segments lesions inside it with a U-Net. Liver and tumour masks come from the public LiTS challenge data; cyst and abscess classes are handled as an extension on additionally annotated scans. A slice viewer overlays masks and reports lesion volume.
What makes it stand up
- 01Cascaded liver-then-lesion U-Net trained with MONAI
- 02LiTS CT volumes with Hounsfield windowing and resampling
- 03Dice and IoU per class, with failure cases in the report
- 04Slice viewer with mask overlays and volume estimates
- 05Academic research prototype, not for clinical decision-making
Module breakdown
- 01
Volume loading
NIfTI reading, HU windowing and spacing normalisation.
- 02
Liver stage
U-Net that crops the region of interest for the next stage.
- 03
Lesion stage
Multi-class U-Net with Dice plus cross-entropy loss.
- 04
Post-processing
Connected components and hole filling on predicted masks.
- 05
Viewer
Web slice browser with overlays and per-lesion volumes.
After this, you will be able to
- Explain why a cascaded pipeline beats one model on small lesions
- Defend Dice loss over plain cross-entropy for segmentation
- Demo scrolling through a CT volume with predicted masks
Technology stack
- Python
- PyTorch
- MONAI
- OpenCV
- FastAPI
- React
- Level
- Capstone
- Domain
- Artificial Intelligence
- Build time
- 5-6 weeks
- Difficulty
You receive
- Complete, commented source code
- 60-90 page project report (IEEE format)
- Editable presentation deck
- Architecture, ER and UML diagrams
- Local setup and deployment guide
- Sample dataset or seed data
- Viva question bank with answers
- One 45-minute walkthrough call
- 30 days of doubt-clearing support