A prototype expert system for proposing radiological examination for traumatic patients

Pao Yue-kong Library Electronic Theses Database

A prototype expert system for proposing radiological examination for traumatic patients

 

Author: Li, Kam-wah Kenneth
Title: A prototype expert system for proposing radiological examination for traumatic patients
Degree: M.Sc.
Year: 1996
Subject: Radiography, Medical -- Data processing
Expert systems (Computer science)
Hong Kong Polytechnic University -- Dissertations
Department: Dept. of Computing
Pages: vi, 107, [7] leaves : ill. ; 30 cm
Language: English
InnoPac Record: http://library.polyu.edu.hk/record=b1234973
URI: http://theses.lib.polyu.edu.hk/handle/200/4895
Abstract: Usually, we will immediately send people to hospital when they get injured. Patients who have either traumatic or non-traumatic injury will receive the first treatment from doctors at Accident and Emergency Department. There, doctors will give preliminary treatment to the injured patients including correction of cardiopulmonary impairment, immobilization of obvious fractures, controlling of active bleeding, and stabilization of haemodynamic status. After that, doctors may propose specialized radiological imaging examinations to investigate any hidden injuries. The examinations can indeed alter clinical performance quite successfully and rapidly. But the accuracy of proposing suitable radiological examination is very crucial and critical. Therefore, the task in the proposing exercise is difficult since it requires knowledge, experience and precise judgment to make the correct decision. Moreover, currently the situation is very problematic because of the serious limitations in human and financial resources, the shortage of experienced senior medical officers, the complexity of radiological examinations, the heavy workload and the need to prevent improper decision making which may lead to over-irradiation to patients. In view of this, a prototype expert system, the REES, to propose suitable radiological examination for traumatic patients has been developed in this project. In the development process, thorough requirement analysis through structured interviews with 12 senior medical officers in 4 main hospitals, the design of knowledge representation, the handling of uncertainty and the building of inference control have been catered for. The acquired knowledge was then refined, cross-checked with another 12 senior medical officers expert in radiological examination and was represented by objects. KAPPA-PC has been used as the development tool during the implementation stage to create object hierarchy. Totally, 4 base classes, 22 subclasses of objects, 4 main screens, 16 popup screens and 733 rules in the form of 13 sets of rules have been created. The main function of the REES is to propose suitable radiological examination. Apart from that, it can also compute the overall certainty factor in proposing the examination, the examination risk, the pregnancy risk and the allergy risk. The type of contrast medium and the maximum dosage are presented in the explanation screen. It can even tell the user what the examination is trying to indicate so that he can take some proactive precautions. The system is constructed in a user-friendly environment. All questions are presented in the form of multiple choices, buttons and sliders. User can just choose the correct options with mouse. The system will alert the user if he forgets to input information. Moreover, as patient data is very confidential, security control by means of password is incorporated in the system to prevent any illegal intruders. To ensure that the product is of an acceptable quality, the prototype has been validated and tested through (1) comparing results as recommended by doctors and the expert system which is run side-by-side with the doctor in Accident and Emergency at Tuen Mun Hospital and (2) testing by the potential users. The result was quite successful and encouraging as demonstrated by the positive comments received from users and the high correct indication rate. Out of the 30 cases tested, 20 cases i.e. 66.6% gave a correct recommendation. The user evaluation through questionnaire also showed that the prototype delivered acceptable result. Lastly, limitations in the using of this system prototype are also revealed. This prototype is at present more rudimentary, primitive and only suitable for proposing radiological examinations for patients with abdominal trauma. However, the development of the prototype cannot be seen as an end. The response from the questionnaire indicated that the system can encompass more functions including more explanatory facilities, on-line help and interactive user interfaces. It can also act as a mean to raise the awareness of the public in the applicability of expert system technology in enhancing efficiency and effectiveness of Medical and Health Care System in Hong Kong, to address the problem of loss of expertise and as a training tool to new and junior doctors.

Files in this item

Files Size Format
b12349732.pdf 4.006Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

     

Quick Search

Browse

More Information