New Feature: Label cartilage “Healthy” or “Damaged.”

[ninanikitina]

Based on the observation of cartilage, we suggest assuming that cartilage is damaged if:

• There is a hole inside the cartilage. In the case of projection, we have two contours, one contour inside another contour
• Rough edges

[ninanikitina]

Timing: ~ 7.5 h

The task has a story size:

• I implemented the utils function based on the OpenCV library. These functions can be useful in finding other feature
• opencv-pyton==4.5.5.64 We need to specify a version of the OpenCV since version 4.6 has a lot of unfixed bugs.
• I took time to figure out how the camera position in Pyvista works to extract cartilage projection for further image analysis.

[ninanikitina]

For the second feature, I tried to compare the area of an inserted circle and the cartilage projected contour itself. I used this idea But, it did not work. The intermediate output is shown below. The first contour of damaged cartilage the second is healthy.

So I found another way that is more straightforward: compare the area of contour and convex hull contour