BRAINDEX: An Interactive, Knowledge-Based System Supporting Brain Death Diagnosis

1990 ◽  
Vol 29 (03) ◽  
pp. 193-199 ◽  
Author(s):  
G. Schwarz ◽  
R. Grims ◽  
E. Rumpl ◽  
G. Rom ◽  
G. Pfurtscheller ◽  
...  

AbstractBRAINDEX (Brain-Death Expert System) is an interactive, knowledge-based expert system offering support to physicians in decision making concerning brain death. The physician is given the possibility of communicating in almost natural language and, therefore, in terms with which he is familiar. This updated version of the system is implemented on an IBM-PC/AT with the expert system shell PC-PLUS and consists of about 430 rules. The determination of brain death is realized with backward chaining and for the optional coma-scaling a forward-chaining mechanism is used.

Author(s):  
C. P. Huang ◽  
F. W. Liou ◽  
J. J. Malyamakkil ◽  
W. F. Lu

Abstract This paper presents an advisory conceptual design tool for mechanical transmission systems. Space consideration was taken into account during the design process. A prototype function tree was built in the form of knowledge-based system to transfer a designer’s idea into a set of mechanical components. An advisory expert system was also developed to help a designer in decision making. As an example, a packaging machine is designed using the developed system.


2018 ◽  
Vol 3 (1) ◽  
pp. 27
Author(s):  
Maura Widyaningsih

Computer field supports the existence of auxiliary program in medical development that is expert knowledge-based system, this system is one branch of Artifical Intellegence (AI). Expert systems are knowledge in learning about estimation or decision-making ability of an expert. Problem solving in the identification of a disease by using auxiliary program is needed a method and concept. Calculation techniques in computing systems are so important, given the level of need for information and the settlement of cases quickly.The results of the study are expert applications that assist in providing results of diagnosis of symptoms managed  the system, with inference using forward chaining, and reasioning with Dempster Shafer. Dempster Shafer's method is not monotonous in solving uncertainty problems, due to the addition or subtraction of new facts. Rule changes will occur, allowing the system to do the work of an expert.Data changes will occur both to diseases, symptoms, solutions and rules, allowing the system to do the work of an expert. The results of manual calculations with the system gives results in accordance with the application of Dempster Shafer method. Management of rules in the database facilitates the search for symptoms within the system.


1990 ◽  
Vol 4 (4) ◽  
pp. 315-331 ◽  
Author(s):  
B. Huber ◽  
J.P. Nyrop ◽  
W. Wolf ◽  
H. Reissig ◽  
A. Agnello ◽  
...  

1996 ◽  
Vol 118 (2) ◽  
pp. 121-126
Author(s):  
B. L. Josefson ◽  
L. O. Wikander ◽  
J. F. Hederstierna ◽  
F. K. Johansson

A fast and simple method for the determination of the residual deformation for a class of welding problems, ring-stiffened pipes, is proposed. The method can predictradial as well as angular distortion of the thin-walled pipe-ring-stiffener/flange assembly. The pipe and stiffener material is elasto-plastic. In particular, the accumulation of deformation in multipass welding is incorporated in the model. Each weld pass is treated separately. This facilitates the assessment of the influence of the sequence in which the weld passes are deposited on the residual deformation state. The method will be included in a conversational knowledge-based “expert” system for the production of a welded ring-stiffened pipe.


Author(s):  
CHUNG-MONG LEE ◽  
TING-CHUEN PONG ◽  
JAMES R. SLAGLE

The image correspondence problem has generally been considered the most difficult step in both stereo and temporal vision. Most existing approaches match area features or linear features extracted from an image pair. The approach described in this paper is novel in that it uses an expert system shell to develop an image correspondence knowledge-based system for the general image correspondence problem. The knowledge it uses consists of both physical properties and spatial relationships of the edges and regions in images for every edge or region matching. A computation network is used to represent this knowledge. It allows the computation of the likelihood of matching two edges or regions with logical and heuristic operators. Heuristics for determining the correspondences between image features and the problem of handling missing information will be discussed. The values of the individual matching results are used to direct the traversal and pruning of the global matching process. The problem of parallelizing the entire process will be discussed. Experimental results on real-world images show that all matching edges and regions have been identified correctly.


Sign in / Sign up

Export Citation Format

Share Document