Winston-Lutz Offset Analyzer (Python Script + Visual Overlay)

🔍 Quickly verify BB-to-field alignment in radiation therapy QA images.

This Python-based tool automates Winston-Lutz analysis by:

• Reading a DICOM image from EPID or portal imaging systems
• Detecting the ball bearing (BB) with subpixel precision using a 2D Gaussian fit
• Identifying the radiation field center via thresholded contour detection
• Calculating offset in millimeters using IEC 61217 U–V convention
• Generating annotated overlay images (BB, field center, field box)

✅ Output includes:

• dU (gun–target direction)
• dV (left–right direction)
• Total 2D offset
• Overlay PNG for easy review

⚙️ Requirements: Python 3.7+ with OpenCV, SciPy, NumPy, and pydicom

🧪 Tested with 1600 mm SID images, 0.251 mm pixel size (modifiable)

🎓 Ideal for medical physicists, dosimetrists, and QA engineers who want a clean, transparent alternative to commercial tools — or just want to understand what’s happening behind the scenes.