Computer-based analytical techniques permit accurate
measurements of size, shape, and texture features.  This paper
describes the use of such analytical techniques to define
computergenerated nuclear features.  These features are then tested to
distinguish between benign and malignant breast cytology.
	The benign and malignant cell samples used in this study were
obtained by fine needle aspiration (FNA) from a consecutive series of
569 patients: 212 with cancer and 357 with fibrocystic breast masses.
Regions of FNA preparations to be analyzed were converted by a video
camera to computer files that were visualized 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
computergenerated a "snake" to precisely enclose each designated
nucleus.  The computer calculated ten nuclear features for each
nucleus.  The ability to correctly classify samples as benign or
malignant on the basis of these features was determined by inductive
machine learning and logistic regression.  Cross validation was used
to test the validity of the predicted diagnosis.  The logistic
regression cross validated classification accuracy was 96.2% and the
inductive machine learning cross validated classification accuracy was
97.5%.
	The accuracy in distinguishing benign from malignant cytology
based on the use of computergenerated nuclear features rivals that
attained by visual diagnosis.  The methods described in this paper
will provide the basis for computerized systems to diagnose breast
cytology.