Key Takeaways

  • Clay content is a first-order driver of flotation, filtration, and tailings behavior — getting it wrong in the block model has real downstream cost.
  • Corescan’s HCI and CSIRO’s HyLogger are the two dominant systems for measuring clay mineralogy from core at scale.
  • CSIRO’s new MyLogger tool applies trained neural networks to interpret HyLogger spectra directly into a geological log — a genuinely mature, purpose-built AI step, not a research prototype.
  • This is a real “Yes” for AI-relevant today, not an emerging/speculative category.

TL;DR

Scan your core with Corescan or HyLogger, run the spectra through a neural-net interpretation layer like CSIRO’s MyLogger to get first-pass clay/alteration mineralogy automatically, then feed that directly into your geomet block model instead of waiting on manual spectral interpretation.

How Do I Predict Clay Content With AI?

Clay content matters more than its line-item status on the poster suggests — it drives flotation reagent consumption, thickener/filter performance, and tailings rheology, which makes an accurate spatial clay model directly valuable rather than a nice-to-have. The data-acquisition side of this is well established: Corescan’s Hyperspectral Core Imager and CSIRO’s HyLogger system both scan core at high spatial resolution and detect the spectral signatures of specific clay minerals (kaolinite, montmorillonite, illite, and others) that are difficult to distinguish visually.

The genuinely new piece is interpretation speed. Hyperspectral mineralogy has historically required a specialist to interpret — a real bottleneck, since hyperspectral mineralogy isn’t part of standard geoscience training. CSIRO’s MyLogger, released in 2023 and funded through the MinEx CRC, closes that gap: it’s a trained neural network that converts raw HyLogger thermal-infrared spectra directly into a first-pass geological log, identifying major lithological boundaries within about a meter of manually logged contacts. That’s a purpose-built, already-deployed AI tool — not a research paper — which puts clay/geomet modelling ahead of most of the other “model” categories on this poster in terms of maturity.

The practical pipeline, then: scan with Corescan or HyLogger, run MyLogger (or an equivalent interpretation layer) to get automated clay/alteration classification, and feed that directly as point data into your geomet block model rather than routing through a manual interpretation step.

Try It With Geocluster

Comparing Corescan against HyLogger for your specific deposit type and clay minerals of concern — and understanding what MyLogger can and can’t do for you — is exactly the sourced, comparative research Geocluster is designed to run.