Browse Prior Art Database

MODEL-DIRECTED DETECTION OF RIBS IN CHEST RADIOGRAPHS

IP.com Disclosure Number: IPCOM000128605D
Original Publication Date: 1978-Dec-31
Included in the Prior Art Database: 2005-Sep-16
Document File: 11 page(s) / 40K

Publishing Venue

Software Patent Institute

Related People

Dana H. Ballard: AUTHOR [+3]

Abstract

The automatic description of the anatomy in medical images such as radiographs, ultrasound images, and computed tomography scans is an important topic of current research. Programs that generate descriptions of the anatomy from such image data, if successful, would provide the medical specialist with new perceptual modalities that could improve the accuracy of his or her decision making. In the recent past, we have attempted to lay the groundwork for a computational model of image processing which is specially designed for domains where there are many geometrical constraints? such as the medical and aerial image domains [Ballard, Brown, and Feldman, 19771. This paper extends these ideas by describing a detailed model that finds the ribs in chest radiographs. A general difficulty with many current medical image processing programs, and programs designed to find the ribcage in chest radiographs in particular [Toriwaki et al., 1973; Persoon, 1976; Wechsler and Sklansky, 1977; Kulick et al., 1976; and Li et al., 1977], is that they are relatively brittle. That is, they can be error prone, particularly in the face of unusual image data, and they have little facility for recognizing and recovering from mistakes. Furthermore, rib programs do not attempt to name the ribs that they find. Our model for the ribcage anatomy provides a context for the rib identification process which allows us to avoid or c " orrect mistakes in the rib-finding process and also to name the ribs that are found. It is especially important to us that the computational model expend as little effort as possible in identifying the ribs. To achieve this end, extensive use of the declarative part of the model is made to reduce the area of the image that must be examined. Also, the notion of incremental planning is used in the analysis process. The best procedure to find each rib is selected from a set of possible procedures, each of which embodies a different rib-detection algorithm. The appropriateness of these procedures varies depending on the state of the analysis process.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 10% of the total text.

Page 1 of 11

THIS DOCUMENT IS AN APPROXIMATE REPRESENTATION OF THE ORIGINAL.

MODEL-DIRECTED DETECTION OF RIBS IN CHEST RADIOGRAPHS

Dana H. Ballard Computer Science Department University of Rochester

TRll March We describe a computational model for the ribcage anatomy seen in a posterior- anterior chest radiograph. This model contains an executive program which examines a declarative structure and can invoke parts of a procedural structure. The declarative structure is in the form of a relational graph and is primarily used to represent gross geometrical relations between the ribs in the ribcage. The procedural structure contains low-level programs that are associated with parts of the' declarative structure and which can find detailed representations of the ribs in the image.

The executive operates incrementally with a cycle of three distinct phases. First, it develops a plan for the "best" rib to find next; secondly, it tries to find that rib; and thirdly, it evaluates its results, incorporating any new information into the declarative structure. Thus, it finds the ribs that stand out the most clearly in the image data first, and then brings this knowledge to bear on the more obscure ribs.

There are many advantages to this model. The ribs found are identified by name. The model is able to find the ribcage even if the ribs are partially obscured by disease processes or where they are faint, as in a high KVP film. The method of describing low-level programs allows experimentation with different kinds of algorithms without changing the high-level model structure, and the high-level declarative structure allows the identification and correction of some mistakes.

The research described in this document was supported by NIH Grant No. R23-HL-21253-01. Page 2

1. Introduction

The automatic description of the anatomy in medical images such as radiographs, ultrasound images, and computed tomography scans is an important topic of current research. Programs that generate descriptions of the anatomy from such image data, if successful, would provide the medical specialist with new perceptual modalities that could improve the accuracy of his or her decision making.

In the recent past, we have attempted to lay the groundwork for a computational model of image processing which is specially designed for domains where there are many geometrical constraints? such as the medical and aerial image domains [Ballard, Brown, and Feldman, 19771. This paper extends these ideas by describing a detailed model that finds the ribs in chest radiographs.

A general difficulty with many current medical image processing programs, and programs designed to find the ribcage in chest radiographs in particular [Toriwaki et al., 1973; Persoon,

1976; Wechsler and Sklansky, 1977; Kulick et al., 1976; and Li et al., 1977], is that they are relatively brittle. That is, they can be error prone, particularly in the face of unusual image data, and they have little facility for recognizing a...