Hyperspectral Imaging

Can tissue be characterised from its reflectance spectra alone?

  1. hyperloom Install the suite
  2. tivis.r Read TIVITA recordings
  3. cuvis.r Read Cubert measurements
  4. hyperspectR Spectral cubes, preprocessing and analysis
  5. hyperspectaculR Spectral visualisation
The hyperloom suite: camera readers feed a common spectral cube, which hyperspectR analyses and hyperspectaculR visualises.

A hyperspectral camera records a reflectance spectrum at every pixel. CTTIR’s hyperspectral packages turn such recordings into tissue maps, and hyperloom installs them together as one suite.

  • tivis.r reads Diaspective Vision TIVITA recordings, their acquisition metadata and exported parameter images, in pure R.
  • cuvis.r connects to Cubert cameras and measurements through the Cubert SDK.
  • hyperspectR holds the spectral cube and does the analysis: preprocessing recipes, hemoglobin fits and band ratios, embeddings and classification with held-out evaluation.
  • hyperspectaculR draws spectral summaries and visual compositions of a cube.

Walkthrough Link to heading

Hyperspectral Tissue Characterization runs the suite end to end. It reads the TIVITA example recording, simulates tissue scenes with known classes, preprocesses them, fits hemoglobin, classifies whole held-out scenes and summarises the result. Every figure on that page comes from running the code shown.

Related spectral packages, such as region annotation (hyperannotatR) and modelling and validation (hypermodR), are listed under Spectral imaging on the Software page.

← Projects & Workflows