IEEE Computational Science & Engineering
Fast Radial Symmetry for Detecting Points of Interest
IEEE Transactions on Pattern Analysis and Machine Intelligence
Design of prefilters for discrete multiwavelet transforms
IEEE Transactions on Signal Processing
The application of multiwavelet filterbanks to image processing
IEEE Transactions on Image Processing
Hi-index | 0.00 |
A substantial shift is occurring in the field of histopathology towards the digital domain with the increasing adoption of digital microscopy and large image databases. There is a growing need for image analysis tools that can efficiently and objectively analyse this wealth of digital image data. This paper presents preliminary results on the development of a suite of such tools for the measurement of toxic effects in the liver. We present an automated procedure for themeasurement of one toxic effect, centrilobular hypertrophy, and present an evaluation of the components of this process. Centrilobular hypertrophy is a condition whereby liver cells in the region of central veins expand in response to a toxin. Our classification process has three stages. The first stage involves detecting the central veins using an interest point detection technique. In the second stage, the interest points are re-ranked to reduce the incidence of false positives. The third stage entails training a classifier to score the tissue in the regions of the putative central veins as hypertrophic or normal.