Bergen Teknologioverføring

Invention

A New 2-D deconvolution technique for more advantageous ultrasound Imaging. The special image resolution of high-end ultrasound scanners has been substantially improved in the last decade. However, these improvements still did not yield favourable image resolution results and the problem of low spatial resolution remained the major limiting factor in the clinical usefulness of medical ultrasound B-mode images.

A new blind Bayesian deconvolution method offers a solution to the addressed problem. The blind Bayesian deconvolution method is a powerful ground-breaking technology that enables reliable, strong enhanced and accurate images in less time, right on your existing workstation. The easy to install, add-on software, helps clinicians to identify tissue interfaces with more confidence and even see structures that are hardly or not visible on unprocessed images. The blind Bayesian deconvolution excels everyday patient care with substantially improved image resolution which limits wrong-diagnostics thus facilitating effective communication with patients and families and between physicians, to accomplish more and effectively decide on treatment planning.

Applications

- in medical sonography (both humans and animals),
- in material sciences

Advantages

- Proven resolution improvements (especially for images of organs deep in the body)
- Makes vaguely defined images of organs visible after deconvolution
- Reduces patient waiting time, eliminates re-scans and streamline office workflow
- One-button-access – easy to use interface
- Easily addable to existing workstations. No additional investments necessary

Technology

The new ultrasound method is a new 2-D deconvolution technique for ultrasound imaging based on blind Bayesian deconvolution. The technique avoids limitations of the 2-D homomorphic deconvolution:  the assumption that the point-spread function and the tissue function lie in different 2-D cepstrum bands and the 2-D phase unwrapping problem.

The new approach is an alternative and improved way of data processing in harmonic imaging where advantages of the fundamental and second harmonic imaging are kept at the same time. The low number of iterations and the global character of the method make it possible to implement the algorithm in real time on ultrasound scanners. In addition, the deconvolution is applied to the complete radiofrequency image data.