Research Supervisor Connect

Object-based Volumetric Texture Feature Extraction for Biomedical Image Retrieval

Summary

Intelligent object-based characterization and extraction of pathologic volumetric texture patterns for biomedical image classification and retrieval.

Supervisors

Associate Professor Tom Weidong Cai, Professor David Feng.

Research location

Computer Science

Program type

N/A

Synopsis

Texture is a powerful discriminating visual feature which has been widely used in pattern recognition and computer vision for identifying visual patterns with properties of homogeneity that cannot result from the presence of only a single color or intensity, and presents almost everywhere in nature – the size of the image patch, the number of distinguishable grey-level primitives and the spatial relationships between these primitives, are all interrelated elements which characterize a texture pattern. Most of biomedical images acquired and represented in gray scale are often highly textured, and consequently, examination of biomedical images usually requires interpretation of organ / tissue / lesion appearance, i.e., the local intensity variations, based on different texture properties such as smoothness, coarseness, regularity, and homogeneity. Since texture acquires such distinguished importance, it is becoming one of the most commonly used characteristics in biomedical image classification and retrieval. This project aims to conduct the study of the characterization and extraction of intrinsic volumetric textures in multiple biomedical object levels such as anatomic organs, tissue regions, and focal lesion objects, with the design of proper statistical texture modeling approaches, to provide great discriminatory information which is of paramount importance to successful classification and retrieval of biomedical imaging data, and to offer intelligent diagnosis via comparing past and current biomedical images associated with pathologic conditions in case-based reasoning or evidence-based medicine.

Want to find out more?

Opportunity ID

The opportunity ID for this research opportunity is 312

Other opportunities with Associate Professor Tom Weidong Cai