User-Friendly AI Software for Automated Quantitative CMR Reporting on Low-Cost, Energy-Efficient Devices
Project description
myopari
myopari is a napari plugin for cardiac MRI segmentation and report generation. It provides an interactive widget to:
- select a 2D or 3D image layer,
- run ONNX-based segmentation,
- compute basic volumetric metrics,
- optionally enrich the report with an LLM rewrite,
- save a markdown report.
Main Features
- Napari dock widget for interactive segmentation
- Built-in segmentation models:
TIRAMISU_ACDCTIRAMISU_EMIDEC
- Works with both 2D images and 3D volumes
- Optional myocardium-only segmentation mode
- Markdown report generation with:
- per-class volumes (mL)
- myocardium mass estimate (g)
- optional external patient info files (
.cfg,.txt,.md)
- Optional LLM-based report rewriting (via
llama-cpp-python)
Installation with pip
pip install myopari
Quick Start in napari
- Launch napari.
- Load a cardiac image (2D or 3D).
- Open the widget:
Plugins -> myopari -> myopari
- In the widget:
- Click
Select image layer - Choose
Edge device - Choose
Segmentation model - (Optional) Enable
Myocardium only - Click
Segment
- Click
- A labels layer is added with a name like:
segmentation_<input_layer_name>_<count>
Report Workflow
After segmentation:
- (Optional) Click
Choose patient info filesand pick.cfg,.txt, or.mdfiles. - (Optional) Enable
Use LLM for report. - Click
Create report. - Click
Save report to .mdto export.
Notes:
- The report includes per-label volumes in mL and myocardium mass estimate.
- If a logo asset exists in
Resources, it is embedded in the report and copied next to the saved markdown file.
Troubleshooting
- Segmentation runtime/provider issues:
- Check installed ONNX Runtime package (
onnxruntimevsonnxruntime-gpu) - Ensure CUDA and driver versions match your
onnxruntime-gpubuild
- Check installed ONNX Runtime package (
- LLM report generation fails:
- Install
llama-cpp-python - Ensure internet access for first model download
- Disable
Use LLM for reportto keep standard report generation
- Install
License
MIT. See LICENSE.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
myopari-0.1.3.tar.gz
(10.4 MB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
myopari-0.1.3-py3-none-any.whl
(10.4 MB
view details)
File details
Details for the file myopari-0.1.3.tar.gz.
File metadata
- Download URL: myopari-0.1.3.tar.gz
- Upload date:
- Size: 10.4 MB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.10.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
202c4b767995aead9eb1f732459d69584b73a6f61fb4f4f7bfbef222af8fbad2
|
|
| MD5 |
1cd414d9c08ed5885aa32ee26e2440c5
|
|
| BLAKE2b-256 |
7bc100ee2ee3b9b77037229095517edbdaee893058f6f82d9449ef8c2e50b551
|
File details
Details for the file myopari-0.1.3-py3-none-any.whl.
File metadata
- Download URL: myopari-0.1.3-py3-none-any.whl
- Upload date:
- Size: 10.4 MB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.10.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f5b0c5930fd2d4162850b920fca951335775b4532566eca6c603ac555f5f50e8
|
|
| MD5 |
f2a50aa486b0680d7e3447ac1a306aff
|
|
| BLAKE2b-256 |
ab9efd4e36e447625ff0c091b881e9f0034775ee5d8a221a6b4a896d5230a7c5
|