Related Experiment Video
Updated: Aug 3, 2026
12:03
Multi-photon Imaging of Tumor Cell Invasion in an Orthotopic Mouse Model of Oral Squamous Cell Carcinoma
Published on: July 25, 2011
Pathfinder: multiresolution region-based searching of pathology images using IRM
1Stanford Medical Informatics and Computer Science, Stanford University, Stanford, CA 94305, USA.
Proceedings. AMIA Symposium
|November 18, 2000
Summary
Pathfinder efficiently searches high-resolution pathology images using multiresolution, region-based techniques. This system demonstrates high accuracy and speed for large pathology image databases.
Area of Science:
- Digital Pathology
- Medical Image Analysis
- Computer Science
Background:
- Digitized pathology slides are rapidly increasing, posing challenges for image database retrieval.
- Current retrieval methods rely on text annotations, which are insufficient for high-resolution images.
- Searching pathology images by content is difficult due to their high resolution.
Purpose of the Study:
- To present Pathfinder, an efficient multiresolution region-based searching system.
- To address the challenges of retrieving information from high-resolution pathology image libraries.
- To improve the accuracy and speed of pathology image database retrieval.
Main Methods:
- Utilizing a multiresolution, region-based searching approach.
- Implementing wavelets for image processing.
- Employing the Integrated Region Matching (IRM) distance for similarity assessment.
Main Results:
- Pathfinder demonstrated high retrieval accuracy in experiments.
- The system achieved high-speed retrieval performance.
- Successful testing on a database of 70,000 pathology image fragments.
Conclusions:
- Pathfinder offers an efficient solution for searching high-resolution pathology image libraries.
- The system's performance is validated by high accuracy and speed.
- The algorithm is expandable for broader medical image database applications.
