A measurement of nuclear roundness involving computer‐assisted image analysis of histologic specimens was used to predict the prognosis of 19 patients with Stage A2 prostate cancer. This technique accurately identified those tumors that clinically progressed without treatment and those that did not, in contrast to the Gleason grading system, which produced significant prognostic overlap. What separated most of these tumors was the quantification and degree of irregularity of a relatively small subpopulation of cancer nuclei among a majority of rounder nuclei. There are currently several practical limitations in the use of this method, but with improvements in technique and in methods of obtaining tissue, computerized image analysis of nuclear shape may become a valuable tool for analyzing prostate cancer in a quantitative manner, thereby providing valuable information relative to the prognosis of the individual patient.
|Original language||English (US)|
|Number of pages||6|
|State||Published - Oct 15 1984|
ASJC Scopus subject areas
- Cancer Research