Home

LARDISS

LARDISS (the acronym comes from LARyngopathy DIagnosis Support System) is a computer tool supporting a non-invasive diagnosis of laryngopathies. Data for diagnosis are extracted from sound samples of the subjects.

LARDISS enables users to:

  • Load a patient's speech signal sample in the wave format.
  • Perform some basic preprocessing operations on a signal in time domain (cutting/selecting time windows, splitting a signal into consecutive time windows, more detailed visualization of selected time windows, etc.).
  • Calculate and visualize the Discrete-Time Fourier Transform (DTFT) - spectrum of a speech signal sample.
  • Classify a speech signal sample using the signal analysis in time domain.
  • Classify a speech signal sample using the signal analysis in frequency domain with respect to the knowledge base embodied in LARDISS.
  • Visualize a decision path for the loaded case - a patient's speech signal - in embodied decision trees.
  • Visualize decision rules included in embodied rule sets and used for making diagnosis for the loaded case - a patient's speech signal.

The listed functionality is implemented in the Java version of the tool. Most of the functions is also available in the LabVIEW version.






















LARDISS - LabVIEW version (Beta) (download)





















LARDISS - Java version (Beta) (download) - Requires JRE 7