Visually assessed nuclear grade has been found to be
prognostically important.  Now, computerbased analytical techniques
accurately measure size, shape, and texture features that constitute
nuclear grade.
	The cell samples used in this study were obtained by fine
needle aspiration (FNA) in the process of diagnosing a consecutive
series of 187 patients with invasive breast cancer.  Regions of FNA
preparations to be analyzed were converted by a video camera to
computer files that were displayed on a computer monitor.  Nuclei to
be analyzed were identified on the computer monitor and were roughly
outlined by an operator using a mouse.  Next, the computer generated a
"snake" that precisely enclosed each designated nucleus.  Ten nuclear
features are then calculated for each nucleus based on these snakes.
These results were analyzed statistically and by an inductive machine
learning technique that we call Recurrence Surface Approximation
(RSA).
	Both the statistical and the RSA machine learning analyses
demonstrate that computerderived nuclear features are prognostically
more important than are the classical prognostic features, tumor size
and lymph node status.
	The methods described in this paper provide the basis for
using computerized systems to determine prognosis from
computer-generated nuclear features.