Posts
2
Figures
20, 12 interactive
Topics
tissue engineering, machine learning, modelling, MLATE
Feed
RSS

Blog

Essays and explorable notes on machine learning for biomaterials and tissue engineering. Every chart and diagram says whether it shows data, a model, a schematic or an illustration with invented numbers.

2 posts · 20 figures, 12 interactive · RSS feed

Latest

· 10 min read

Computational tissue engineering and why the field is still experimental

Computers already help design scaffolds and bioreactors, yet tissue engineering still depends on making things and testing them. Why simulation has not replaced the experiment here, and what would have to change.

  • Computation is already good at narrowing the options, from bioreactor flow rates to which formulations to try first.
  • It cannot yet guarantee an outcome, because cell behaviour has no agreed governing laws and the inputs vary from batch to batch.
  • In my reading of the field, most models are tested on data like their training data; few are tested on new experiments in other labs.
  1. · 5 min read

    The idea behind MLATE

    Why I started turning the scaffold literature into data for machine learning, how the question changed over three versions, and what a tool like this can and cannot do.