User Tools

Site Tools


16.07 - An N-Point Statistics Framework for Predicting Tissue Traits in Biomedical Images

Project - Summary


To develop a general biomedical image informatics framework that is capable of predicting the traits of tissues captured in images of immunohistochemistry-stained (IHC) specimens. Our work on the framework has been on-going for a number of years. The CVDI Year 5 funding supported three separate sub-projects that contributed to the enhancement of the framework. They are:

  • Definition of an alternate color representation for the analysis of Hematoxylin and Eosin (H&E) images (a type of IHC image), which supports color normalization that effectively reduces inter-slide variability during downstream analysis.
  • Estimation of fine-scale histologic features at low magnification which allows for the analysis of IHC images at lower pixel resolutions; thus providing computational speed-ups that facilitate high-throughput analysis of while-slide IHC images.
  • Development of an approach for characterization spatial structures in biomedical images based on N-point statistics analysis. The features derived from the analysis may be used to predict tissue traits.

Project - Team

Team Member Role Email Phone Number Academic Site/IAB
David Breen PI Not available Not available Drexel University
Mark Zarella PI Not available Not available Drexel University
Chan Yeoh Student Not available Not available Drexel University
Matthew Quaschnick Student Not available Not available Drexel University
Dan Johnson Student Not available Not available Drexel University
Zahra Riahi Samani Visiting PhD Student Not available Not available Shahid Beheshti University (Iran)

Project - Impact and Uses/Benefits

We envision multiple impacts and uses for the technologies developed in the three sub-projects. They are:

  • SP1: Providing an improved tissue structure segmentation algorithms based on our alternative color space for H&E images.
  • SP1: Supporting color normalizataion that will alleviate the problem of staining variability, a major impediment to automated processing and anaylsis of H&E images.
  • SP2: The ability to estimate the histological composition of tissue in low-resolution images will facilitate the computational analysis of whole-slide images; thus enabling automated and high-throughput tissue analysis in pathology.
  • SP3: N-point statistics provides a method for characterizing the architecture of tumors; thus improving the automated analysis/grading of tissues that should lead to improved, more accurate and robust diagnosis of cancer.

Project - Deep Dive

Project - Documents

FilenameFilesizeLast modified
16.07_year_5_presentation.pdf388.6 KiB2019/08/22 13:21
16.07_year_5_final_report.pdf2.0 MiB2019/08/22 10:47
projects/year5/16.07.txt · Last modified: 2019/08/22 10:49 by sally.johnson