Reflexive algorithmic approach to clinical decision making: Breast cancer as a model
✍ Scribed by Douglas C. Aziz; Raj R. Barathur
- Book ID
- 102303287
- Publisher
- John Wiley and Sons
- Year
- 1993
- Tongue
- English
- Weight
- 91 KB
- Volume
- 53
- Category
- Article
- ISSN
- 0730-2312
No coin nor oath required. For personal study only.
✦ Synopsis
The ability to detect estrogen and progesterone receptors by immunocytochemical analysis in formalin-fixed, paraffin-embedded sections has clear advantages over other techniques, including the ability to assay small biopsy specimens, fine needle aspirate samples, and archival material. Twenty-two cases of breast carcinoma were evaluated for estrogen and progesterone receptors by immunocytochemical analysis and enzyme immunoassay. Using a true color-based image analysis system, histograms of area versus the optical density of the positive staining nuclei were generated. A binary decision algorithm was derived from these histogram parameters by the Classification and Regression Trees (CART) computer program. Estimates generated by the algorithm for image analysis/immunocytochemical analysis had a 90% concordance with the enzyme immunoassay values. We conclude that quantitative immunocytochemical results for estrogen and progesterone receptor content in formalinfixed, paraffin-embedded tissue can be generated using image analysis.