5.6 Code Labs / Practice Appendix

This page is the practice appendix for Chapter 5. Its job is different from the four mainline pages:

• mainline pages (5.1–5.4) explain concepts and decision logic;
• this appendix collects runnable entry points, dependencies, outputs, and implementation pointers;
• 5.5 remains optional frontier reading.

Where is the full implementation?

All complete scripts, training logic, demos, and generated outputs are under src/ch05/.

Script groups by topic

Mainline question	Local code area	Representative scripts
How should data be prepared?	preprocessing	clip_hu_values/, medical_image_resampling/, n4itk_bias_correction/, white_stripe_normalization/, detect_metal_artifacts/, visualize_bias_field/
Why does segmentation work?	segmentation	lung_segmentation_network/, medical_segmentation_augmentation/
How should we think about classification and detection?	classification	medical_image_classification/
When should enhancement or restoration be used?	augmentation / restoration	medical_image_augmentation/, clahe_enhancement/, plus the MRI bias-field tools

---

Recommended reading order inside src/ch05/

1. Start with src/ch05/README_EN.md for the chapter-wide experiment index.
2. Then open the subfolder README for the script you want to run.
3. Use the local output/ directory in each experiment to inspect generated images and reports.

---

Typical run pattern

Most Chapter 5 demos follow the same structure:

cd src/ch05/<experiment_name>
python main.py

Some experiments also provide a simplified entry or an extra test file:

python simple_augmentation.py
python test.py

---

Environment and dependencies

For Chapter 5 practice, dependencies are documented in:

• src/ch05/requirements.txt
• src/ch05/README_EN.md
• individual experiment READMEs when extra packages are needed

Typical packages include:

• numpy, matplotlib, scipy, scikit-image
• opencv-python
• torch, torchvision
• pydicom, nibabel, SimpleITK

---

Where are full outputs stored?

Each experiment keeps its own generated artifacts, usually in one of these folders:

• output/
• outputs/

Examples:

• src/ch05/lung_segmentation_network/output/
• src/ch05/medical_image_classification/output/
• src/ch05/medical_image_augmentation/output/
• src/ch05/clahe_enhancement/output/

These folders hold figures, reports, and demo visualizations that would be too heavy for the main tutorial pages.

---

How should the appendix be used?

Use this appendix when you want to:

• run the chapter code locally;
• inspect full implementation details;
• compare generated outputs;
• understand environment setup and script entry points.

Return to the mainline pages when you want to answer the conceptual questions of Chapter 5.

One-sentence summary

The mainline pages explain why; this appendix and src/ch05/ show how to run and inspect the full workflow.