Automated region of interest retrieval and classification using spectral analysis
2008

Automated Image Analysis in Pathology

Sample size: 73 publication Evidence: moderate

Author Information

Author(s): Myriam Oger, Philippe Belhomme, Jacques Klossa, Jean-Jacques Michels, Abderrahim Elmoataz

Primary Institution: F. Baclesse Cancer Centre

Hypothesis

Can spectral analysis improve automated region of interest retrieval and classification in pathology?

Conclusion

Spectral Analysis is effective for automating the diagnosis of cancers and sorting isolated cells.

Supporting Evidence

  • Spectral analysis was used to segment breast tumor tissue into stroma and epithelial zones.
  • The method allowed for the selection of representative patches of each histological type.
  • In blood smears, spectral analysis helped sort isolated blood cells into two classes.

Takeaway

This study shows that computers can help doctors find important parts of images of tumors and blood samples more easily.

Methodology

The study used spectral analysis for image segmentation and classification on whole slide images of breast tumors and blood smears.

Digital Object Identifier (DOI)

10.1186/1746-1596-3-S1-S17

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication