File size: 1,670 Bytes
4e8974f
 
 
 
 
 
271709e
4e8974f
057d45f
4e8974f
271709e
 
 
 
 
4e8974f
94e0e92
4e8974f
94e0e92
4e8974f
 
 
94e0e92
4e8974f
 
 
 
 
 
 
 
 
 
 
 
94e0e92
4e8974f
 
 
 
 
94e0e92
4e8974f
 
 
 
94e0e92
4e8974f
 
 
94e0e92
4e8974f
271709e
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
---
title: ZO-1 Network Analysis (RIS + TiJOR)
emoji: 🔬
colorFrom: purple
colorTo: blue
sdk: gradio
sdk_version: 5.44.1
app_file: app.py
python_version: 3.11
tags:
- gradio
- bioimage
- segmentation
- cellpose
- computer-vision
---

## ZO-1 Network Analysis Tool

Gradio app for AI-powered segmentation and quantification of ZO-1 tight junction networks using:
- RIS (Radial Integrity Score): concentric circle crossings per pixel length
- TiJOR (Rectangular method): rectangle perimeter crossings per pixel length

### How to use
1. Upload a ZO-1 image (TIFF, PNG, JPG). 16-bit TIFFs are supported.
2. In Segmentation:
   - Adjust Cell Diameter Estimate (px) if results are off
   - Scale Factor 1.0 = full size (lower if slow)
   - AI Contour Validation can improve contours but is slower
3. In Analysis:
   - Pick RIS (circles) or TiJOR (rectangles)
   - Tune parameters (radii/rectangles, min separation)
   - Optionally show contours/geometry/cross-sections
4. In Export:
   - Download CSV or Text report; files are named `<image>_RIS.csv/.txt` or `<image>_TiJOR.csv/.txt`

### Features
- Cellpose-based segmentation (GPU or CPU)
- Robust TIFF handling (8/16-bit)
- RIS and TiJOR analyses with visual overlays
- Exports with image-based filenames

### Parameters
- Segmentation: Cell Diameter (px), Scale Factor, AI Validation
- RIS: κ (packing factor), min/max radius (%), number of circles, min separation (px)
- TiJOR: initial/max rectangle size (%), steps, min cross-section distance (px)

### Notes
- Large images can be slow; reduce Scale Factor if needed
- GPU is recommended for faster segmentation

### Acknowledgements
- Built with Gradio and Cellpose