Author/Editor		Zrimec, Tatjana; Sammut, Claude
Title		The potential for machine learning in medical image processing
Type		članek
Source		In: Kononenko I, Urbančič T, editors. CADAM-97. Zbornik Računalniška analiza medicinskih podatkov; 1997 nov 12; Bled. Ljubljana: Inštitut Jožef Stefan,
Publication year		1997
Volume		str. 166-77
Language		eng
Abstract		The problem addressed in this paper is how the background knowledge for a complex image processing task can be acquired. We argue why learning is necessary for processing medical images and what kind of learning is most suitable. Our aim is to use machine learning techniques to transform the experience of radiologists into knowledge that will refine object models and improve performance of medical image understanding. We discuss a method for retrieving and refining domain specific knowledge from images using interactive knowledge acquisition tools, such as ripple-down rules. We discuss also how a relational machine learning can be used for learning from images.
Descriptors		IMAGE PROCESSING, COMPUTER-ASSISTED ARTIFICIAL INTELLIGENCE DIAGNOSTIC IMAGING ANGIOGRAPHY