Method for Biological Tissue Analysis Using Block Based Local Binary Patterns

SUMMARY

Analysis of breast tissue images using a technique called LBP (Local Binary Pattern) is considered as a promising method for non-invasive detection of breast cancer cells. This technique calculates the statistical distribution of different patterns and uses them as feature vectors to represent textural differences. By removing the need for texture images to be highly repetitive, LBP enables improved tissue classification of OCT/OCM (Optical Coherence Tomography/Optical Coherence Microscopy) images. In this invention, the LBP method has been further refined into a process called BLBP (BlockLBP).

BLBP is able to provide a higher degree of classification accuracy by examining the intensity differences at a block level rather than only at the individual pixel level like LBP. It extracts the intensity differences among certain blocks of the neighborhood around a pixel. Then it organizes the blocks in two novel ways – spokes and rings. By integrating spoke BLBP features with ring BLBP features and then using a neural network classifier, BLBP is able to achieve very high classification accuracies.Competitive Advantage BLBP is superior to LBP and other classification/detection methods because it uses intensity differences between blocks of pixels in the neighborhood of the diseased area. This enables higher classification accuracy of texture images. It is also able to provide better results in both tissue classifications and determining if the structure is benign or not. Another aspect that differentiates BLBP is the ability to construct multi-scale features through the combination of BLBP features with different radius parameters while simultaneously being more robust to noise in the image and distinguishing between different types of human tissue. In this manner, BLBP is able to deliver a richer set of information than LBP by examining different blocks in the neighborhood.

Lehigh Tech ID #112014-01

Market Need/Opportunity

This invention will have broad applications in the breast and other cancer detection market space. Medical equipment manufacturers and diagnosis systems manufacturers that are involved in development and use of new tools and systems for early diagnosis and detection of breast cancer would be able to use this invention to enhance system performance and other capabilities.

Cancer/tumor profiling is an area that is expected to revolutionize cancer diagnostics and therapeutics. Industry research firm, Markets and Markets, indicates that the global cancer/tumor profiling market was valued at $13.30B in 2012. At the end of 2013, the market was estimated to be $14.99 billion and was poised to reach around $35 billion by 2018, growing at a CAGR of18.50% from 2013 to 2018.

Status

Lehigh University is looking for a partner for further development and commercialization of this technology through a license. The inventor is available to collaborate with interested companies.

App Type Country Serial No. Patent No. File Date Issued Date Expire Date
Provisional [PR] United States 62/083,391 11/24/2014    
For Information, Contact:
Thomas Meischeid
Interim Director
Lehigh University
tjm5@lehigh.edu
Inventors:
Chao Zhou
Xiaolei Huang
Sunhua Wan
Keywords: