🧬 RadComp: Radiobiology Calculator for Medical Physics

RadComp is a clinical decision support tool designed to streamline the conversion of physical doses into biologically equivalent doses ($BED$ and $EQD2$) and to perform risk analysis in complex re-irradiation scenarios, including SBRT/SRS.

🚀 Live App Access: [https://radcomp.streamlit.app/]

✨ Key Features

• Dual-Engine Modeling: Seamless switching between Standard LQ and Linear-Quadratic-Linear (LQL) models based on dose per fraction.
• Smart Clinical Alerts: Dynamic detection of biological validity thresholds ($d_T = 2\cdot\alpha/\beta$) specific to the selected tissue (e.g., Spinal Cord vs. Tumor), preventing model misuse.
• Clinical Database: Pre-configured $\alpha/\beta$ ratios and dose-volume constraints based on QUANTEC, HyTEC, and international peer-reviewed literature.
• Advanced Re-irradiation Module:
  • Time-based biological recovery modeling (12-24 months).
  • Spatial overlap penalty adjustment for high-dose regions.
  • Logic validation to prevent penalties on zero-dose structures.
  • Cumulative dose assessment with dynamic stacked charts.

🧮 Radiobiological Models

RadComp utilizes a hybrid approach to prevent the known overestimation of cell kill by the LQ model at high doses per fraction (SBRT/SRS).

1. Standard Linear-Quadratic (LQ) Model

Used for conventional fractionation where dose per fraction $d$ is within the "shoulder" of the survival curve.

$$BED = D \times \left(1 + \frac{d}{\alpha/\beta}\right)$$

2. High-Dose Correction (LQL Model)

For hypofractionated treatments (SBRT/SRS), RadComp implements the Linear-Quadratic-Linear (LQL) model proposed by Astrahan (2008). The model transitions from a quadratic curve to a straight line at a specific threshold dose $d_T$:

Validity Threshold: $$d_T = 2 \cdot (\alpha/\beta)$$

Calculation ($d > d_T$): $$BED_{LQL} = \frac{1}{\alpha} \left[ \alpha d_T + \beta d_T^2 + \gamma (d - d_T) \right] \times N$$

This correction is suggested automatically by the interface when the dose per fraction exceeds the specific biological threshold of the selected organ.

3. Equivalent Dose in 2 Gy (EQD2)

To normalize treatment schemes to a standard 2 Gy fractionation:

$$EQD2 = \frac{BED}{1 + \frac{2}{\alpha/\beta}}$$

🧪 Clinical Validation

Reliability is our priority. RadComp's calculation engine has been validated using test vectors compared against reference clinical cases:

Test Case	Reference Model	Expected Cumulative EQD2	Status
Spinal Cord Re-irrad	Nieder et al. (2006)	~56 Gy	✅ Validated
Lung Re-irrad	Central Toxicity Protocols	~69 Gy	✅ Validated

Note: Validation assumes a 50% recovery factor at 12 months and an overlap penalty applied to RT1 (Previous Course).

⚖️ License

This project is licensed under the MIT License. Feel free to use, modify, and collaborate. See the LICENSE file for details.

⚠️ Disclaimer

For Research and Educational Use Only. This tool is not a medical device and has not been cleared for clinical use by any regulatory authority. All calculations must be independently verified by a certified Medical Physicist or Radiation Oncologist. The author assumes no liability for clinical errors or misuse of this software.

✉️ Contact & Collaboration I am a Medical Physicist interested about the intersection of oncology and software development. I am open to feedback, collaborations, and professional opportunities.

LinkedIn: Luis Fernando Paredes https://www.linkedin.com/in/lfparedes1/ Email: luisfernandoparedes2@gmail.com

🛠️ Tech Stack

• Python 3.10+
• Streamlit (UI Framework)
• Plotly (Interactive Visualizations)
• NumPy/Pandas (Calculation Engine)

🚀 Installation & Local Run

To run this project locally, clone the repository and install the dependencies:

git clone [https://github.com/LuisParedesOcampo/RadComp.git](https://github.com/LuisParedesOcampo/RadComp.git)
cd RadComp
pip install -r requirements.txt
streamlit run main.py