U.S. Patents

6,850,062: Local Multi-scale Fourier Analysis for MRI

7,005,854: Synthetic Aperture MRI

U.S. Provisional Patents

10/430294: Distributed Vector-Processing of the S-Transform for Medical Applications

10/430293: Visualization of the S-Transform Data Using Principle Component Analysis

10/430204: Filtering Artifact from fMRI Data Using the Stockwell Transform

1/118,366: Image Texture Segmentation Using Polar S-Transform and Principal Component Analysis

60/697,965: Polarization Analysis and Polarization Filtering of 3-Component Signals Using the S-Transform

60/749,638: Local Dominant Wave-Vector Analysis of Seismic Data

60/787,699: Super Resolution Contextual Close-up Visualization of Volumetric Data

60/685,539: Method and System for Signal Processing Using a Sparse Approximation of the S-Transform

60/799,649: Processing of Seismic Data Using the S-Transform

Filing 173-12 US: Visualization of Volumetric Medical Imaging Data

Technology

Calgary Scientific invests deeply in research to formulate progressive new technologies, either in-house or in partnership with accomplished researchers and institutions. We commercialize these technologies to develop products and solutions that address industry-specific needs where ample return on investment is anticipated. We choose to concentrate on technologies that can be applied to multiple market segments to maximize the probability of success, while minimizing related research and development costs for industry-specific adaptations.

Our technologies are commercialized by a seasoned team that applies agile software development methodologies to rapidly advance the company’s product line.

Our current competitive strength is based on patented technologies related to signal and image processing, advanced visualization, analysis and user interface design that will help customers maximize the value of very large data sets.

The Stockwell Transform

Currently, a significant portion of Calgary Scientific’s core technology is based on a patent pending, proprietary version of the Stockwell Transform (S-Transform), a signal processing technique that allows for time-frequency spectral localization, or space-frequency in the case of images. It is similar to the short-time Fourier Transform (STFT), but with a Gaussian window having frequency-dependent width and height.

The S-Transform has an advantage over the STFT and wavelet transforms in that it provides multi-resolution analysis while retaining the absolute phase of each frequency. This has led to its application for the analysis of data from a variety of fields, such as the following:

Magnetic resonance (MR) images of patients with multiple sclerosis
Functional MR imaging data to correct for motion artifacts
Phonocardiogram data in cardiology for heart disorders
Electroencephalograms in neurology to study seizures and to assess other diseases such as Parkinson’s
Electrocardiograms in cardiology for non-invasive detection of heart arrhythmias
Seismic data to improve the interpretation of subsurface geological formations and their content (e.g. oil vs. water)
Ocean surface winds to assess relationship with tidal motion

Technical Papers

The following papers and articles are property of the respective authors.

“The S-Transform with Windows of Arbitrary and Varying Shape”, C.R. Pinnegar and L. Mansinha, Geophysics 68(1), 381-385 (2003)
“A Method of Time Analysis: The TT-Transform“, C.R. Pinnegar and L. Mansinha, Digital Signal Processing 13(4), 588-603 (2003)
“Application of the S-Transform to Pre-Stack Noise Attenuation Filtering”, C.R. Pinnegar and D. W. Eaton, Journal of Geophysical Research – Solid Earth 108(B9), no.2422 (2003)
“The Local Spectral Analysis for Non-Stationary Time Series: The S Transform for Noisy Signals”, C.R. Pinnegar and L. Mansinha, Fluctuation and Noise Letters, Vol. 3, No. 3 (2003) L357-L364
“The Bi-Gaussian S-Transform”, C.R. Pinnegar and L. Mansinha, SIAM Journal of Scientific Computing,Vol. 24, No. 5, pp. 1678–1692

S-transforms show details that are difficult to discern on a noisy earthquake seismogram.

When the S-transform is used as a time-frequency filter, polarity information that may not be easy to determine by visual inspection of the seismogram can also be obtained, and used to determine specific information on the fault and direction of slip on it.

[Learn more…]