Download or view PDF

Modelling an Integrated Fuzzy Based Decision Support System for an Anaesthetist during Pre-Operative Clinical Assessment

1Etuk Akaninyene U., 2Harmony Nwobodo-Nzeribe and 1Okoye Francis A.

1Department of Computer Engineering Enugu State University of Science and Technology (ESUT) Nigeria.

2Dir. Research and Innovation Enugu State University of Science and Technology (ESUT) Nigeria.

Corresponding Author Email: ibomatai1970@gmail.com

ABSTRACT

Pre-operative clinical assessment is a crucial responsibility for every anaesthetist before any major medical procedure. This assessment evaluates a patient’s suitability for anaesthesia and their overall physical condition to endure the stresses of surgery. This is especially vital given recent data linking many surgical deaths to improper anaesthesia administration, with a mortality rate of 20 deaths per 10,000 anaesthetics in developed countries. This paper proposes an integrated fuzzy-based decision support system to assist anaesthetists during pre-operative clinical assessments. The approach combines neural network algorithms to classify and normalize input data, which then feeds into a fuzzy logic system for multi-variable decision-making analysis. To evaluate this decision support system, patient anaesthesia assessment datasheets were sourced from Ibom Multi Specialist Hospital in Akwa Ibom State, along with an online dataset from electronic anaesthesia records (MetaVision, 1MDsoft) of adult patients undergoing surgery at a major academic medical center. The study focused on patients over 18 years old undergoing non-cardiac surgeries, taking into account factors such as urgency and type of anaesthesia. The next step involves a decision-making process that assesses patient suitability for surgical anaesthesia based on established rules and membership functions developed through the fuzzy inference system. The results from this integrated decision support system were found to be reliable and consistent with anaesthetists’ assessments. The system achieved an accuracy of 91.46%, as indicated by a ROC curve comparing sensitivity and specificity, demonstrating its effectiveness in evaluating patients for medical procedures. To enhance the training of future anaesthesiologists, the introduction of artificial intelligence into the curriculum is recommended, along with the application of evolutionary algorithms like genetic algorithms and programming. These strategies aim to significantly enrich the decision-making processes employed by anaesthetists.

Keywords: Fuzzy logic, neural network, anaesthesia, artificial intelligence

CITE AS: Etuk Akaninyene U., Harmony Nwobodo-Nzeribe and Okoye Francis A.(2024). Modelling an Integrated Fuzzy Based Decision Support System for an Anaesthetist during Pre-Operative Clinical Assessment. NEWPORT INTERNATIONAL   JOURNAL OF SCIENTIFIC AND   EXPERIMENTAL SCIENCES, 5(3):78-92. https://doi.org/10.59298/NIJSES/2024/53.789200