DXAnalyzer

DXAlyzer is a desktop tool for extracting, reviewing, validating, and exporting DXA measurement data from Hologic DXA PDF reports and OCR text exports.

The application runs locally, stores data in a local SQLite database, and prepares cohort-level CSV tables and SVG figures for research and publication workflows.

• Import DXA reports from .pdf, .txt, .text, and .ocr files.
• Extract patient identifiers, scan dates, BMD, BMC, T-score, Z-score, fat mass, lean mass, and total mass values.
• Review and edit extracted records before saving them.
• Validate measurements with QC status and notes.
• Compare repeated measurements for the same patient.
• Export full measurement data to CSV.
• Generate publication summary tables and subject-level change tables.
• Export cohort figures as SVG.
• Manage parser templates for different report formats.

For running from source:

• Python 3.11 or newer
• Windows, macOS, or Linux with Tkinter support

Python dependencies are listed in requirements.txt.

python -m pip install -r requirements.txt
python main.py

Application data is stored locally in:

~/DXAlyzer/dxa.db
~/DXAlyzer/templates/

The Windows executable is built with PyInstaller and published through GitHub Actions.

To download it:

1. Open the repository on GitHub.
2. Go to Releases.
3. Download DXAlyzer.exe from the latest release.

If there is no release yet, open Actions, run the Windows Release workflow manually, and download the windows-DXAlyzer artifact.

From a Windows command prompt in the project folder:

build.bat

Expected output:

dist\DXAlyzer.exe

More release and code-signing details are documented in WINDOWS_RELEASE.md.

1. Import PDF reports or OCR text exports.
2. Review extracted records before saving.
3. Check measurement quality in the QC tab.
4. Run cohort analysis.
5. Export CSV tables and SVG figures.

This software was developed as part of the project Research of Excellence on Digital Technologies and Wellbeing, CZ.02.01.01/00/22_008/0004583, co-funded by the European Union.

This project is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).