Add further features for spots and branch graph

[stefanhahmann]

User Story

A user may want to have additional features on the branch graph for some analysis.

Possible Features

Spot Features:

• Axes lenghts, i.e. minor axis, middle axis, major axis. The axes should be derivable from this, cf. Spot#getCovariance.
• Spot volume, spot radius already exists. The ellipsoids are shown in BigDataViewer. The axes should be derivable from this, cf. Spot#getCovariance.
• aspect ratios of ellipsoid axes
  • 3 ratios considering 2 axes each
  • further ratios considering all 3 axes, e.g. (b-c)/(a-c). a .. major axis, b .. middle axis, c .. minor axis
• Average distance to 5(?) nearest neighbours
• Texture params, i.e. histogram analysis within spot - how to express with a few numbers??? E.g. mean, median, variance of grey values. Perhaps frequency based indicators.
• Division orientations (a 3 dimensional vector)
• Moving average of intensity (e.g. average including the previous 3 and the next 3 spots)

BranchSpot Features:

• Average velocity. May be derived from displacement and duration. Needs to checked, if displacement equals the distance along the cell path.
• Measure for how direct the movement of the cell ("directness of movement") is, i.e. "euclidean distance between cell divisions" / "distance along cell path"
• Cell behaviour:
  • Growing cell / shrinking cell
    • Volume quotient between cell divisions
    • Average growth/shrinking rate
    • Absolute delta of growth/shrinking
• Average spot intensities
• Number of leaves of the BranchSpot
• Number of (sub-)divisions of the BranchSpot
• Time since last division
  • t.b.c.

BranchLink Features:

• Proportion of size/volume compared to the "sister"-cell (i.e. the other "daughter"-cell of the cell division)
• Division orientations for both "daughter"-cells.

Discussion:

• Which features may be useful?
• Are there further features?
• Do we need a plotter for the features - extend Grapher with Branch Features? By this params of cell life cycles could be analysed.

[tinevez]

Hello Stefan,
This looks like fantastic mew features.
The spot volume and aspect shape can indeed be computed from the decomposition of the covariance matrix.
As the question to what is useful, I think we need to go to biologists and publication to know. What do they use, what do they need. What do they put in their publications. That kind of things.

[stefanhahmann]

Moved this issue to here: mastodon-sc/mastodon-deep-lineage#1