Precedent Note ยท Updated June 2026
Microglia, fractal analysis, and subtle morphology
This page records a biology precedent for why morphology quantification must be multi-metric, acquisition-aware, and restrained about what any single descriptor can claim.
Why it matters for Fractalish: it supports morphology quantification, multi-metric descriptor extraction, and the warning that fractal dimension is useful but incomplete.
Key points
- Microglial form and function are linked, and subtle intermediate forms matter.
- Traditional categories are often too blunt to capture the actual variation.
- Box counting, lacunarity, and multifractal methods can quantify morphology that simple labels flatten.
- Fractal dimension is not unique or complete. It can support description without exhausting what the morphology means.
- Preprocessing, image acquisition quality, segmentation, and scale all matter to the trustworthiness of the descriptor output.
Tie to Fractalish
Fractalish does not need this paper to prove its framework. It does gain a serious methodological precedent: subtle morphology can require multiple descriptors, intermediate states matter, and acquisition or preprocessing can easily damage interpretability. That is exactly why AMCVA and HOLD belong in the stack.