NOVEL 3D WAVELET-BASED FILTER FOR VISUALIZING FEATURES IN NOISY BIOLOGICAL DATA

To date, no methods have been very successful in determining fine structural details of specific complex objects. Although imaging tools such as electron microscope tomography allow 1nm resolution in biological systems, there is generally a lot of associated background noise due to nonspecific staining. Methods of analyzing tomography images such as Fourier, correlation, and model fitting fail to filter through the low signal to noise ratios since they provide information on global frequency content, but no information on where particular frequencies occur.

UCSF investigators have developed a new wavelet-based image processing algorithm that can determine where particular frequencies occur, independent of how often they occur. The program is useful in determining the structure of an object that is buried in a cloud of nearly the same density material as the object. This filter was originally developed as a new approach to solving the chromosome structure problem. The investigators have verified the method works by analyzing the 3D tomographs of microtubules. However, investigators also found the filter to be potentially useful in a variety of other biological problems, including monitoring tissue changes and identifying the location of tissue lesions due to focused ultrasonic surgery.

In the latter application, the wavelet-based filter would allow clinicians to track lesions by processing the acoustically acquired data from the ultrasound unit in real time and provide continuous treatment feedback without the need of additional tools (such as Magnetic Resonance guidance for ultrasound ablation) during high intensity focused ultrasound surgeries. A self-contained ultrasound unit should significantly reduce the cost for surgical equipment.

ADVANTAGES:

• New, easy to use wavelet-based filter that adapts to the size of the potential structure.
• Can identify structures that occur rarely.
• Very fast method.
• Allows image processing in real time.

References: Moss, et al. (2005) Journal of Microscopy, Vol. 219, pp. 43-49.

The University has filed a patent application (U.S. Application No. 20050123216) for this technology covering the algorithm and its uses.

If you would like to receive further information about this technology and potential licensing opportunities, please contact:

Ha Nguyen, Ph.D.
Licensing Associate
phone (415) 353-4461
ngoc-ha.nguyen@ucsf.edu

Reference: OTM Case #2003-101