Getting Started

This guide walks you through setting up the repository and running training and evaluation for breast cancer risk prediction models.

---

⚙️ Installation

Clone the repository and set up a Python environment:

git clone https://github.com/sot176/BreastCancerRiskBenchmark.git
cd BreastCancerRiskBenchmark

conda create -n bc_risk python=3.12
conda activate bc_risk

pip install -r requirements.txt

---

📁 Configuration

All experiments are controlled through scripts located in the scripts/ directory.

⚠️ Important:

Before running any script, update the paths inside the scripts to: - point to your dataset location
- specify output directories for logs and results

For dataset formatting and preprocessing, see the Datasets section.

---

🧠 Training

Each model can be trained using its corresponding script:

scripts/train_mirai.sh
scripts/train_vmra_mar.sh
scripts/train_oa_breacr.sh
scripts/train_imgfeatalign.sh
scripts/train_lmv_net.sh

---

📊 Evaluation

Evaluate trained models using:

scripts/test_mirai.sh
scripts/test_vmra_mar.sh
scripts/test_oa_breacr.sh
scripts/test_imgfeatalign.sh
scripts/test_lmv_net.sh

Evaluation scripts compute performance metrics such as AUC.

---

🧪 Workflow

Typical usage:

1. Prepare dataset (see Datasets)
2. Update paths in scripts
3. Train a model
4. Run evaluation

---

❗ Notes

• Ensure dataset formatting is correct before training
• Use consistent preprocessing across experiments
• GPU acceleration is recommended