Automation of the anesthetic processnew computer-based solutions to deal with the current frontiers in the assessment, modeling and control of anesthesia

  1. González Cava, José Manuel
Dirixida por:
  1. Juan Albino Méndez Pérez Director
  2. José Luis Calvo Rolle Director

Universidade de defensa: Universidad de La Laguna

Fecha de defensa: 19 de novembro de 2020

Tribunal:
  1. Carlos Bordóns Alba Presidente/a
  2. Santiago Torres Álvarez Secretario
  3. Antonio Visioli Vogal
Departamento:
  1. Ingeniería Informática y de Sistemas

Tipo: Tese

Teseo: 640673 DIALNET lock_openRIULL editor

Resumo

The current trend in automating the anesthetic process focuses on developing a system for fully controlling the different variables involved in anesthesia. To this end, several challenges need to be addressed first. The main objective of this thesis is to propose new solutions that provide answers to the current problems in the field of assessing, modeling and controlling the anesthetic process. Undoubtedly, the main handicap to the development of a comprehensive proposal lies in the absence of a reliable measure of analgesia. This thesis proposes a novel fuzzy-logic-based scheme to evaluate the impact of including a new variable in a decision-making process. This scheme is validated by way of a preliminary analysis of the Analgesia Nociception Index (ANI) monitor on analgesic drug titration. Furthermore, the capacity of the ANI monitor to provide information to replicate the decisions of the experts in different clinical situations is studied. To this end, different artificial intelligence-based algorithms are used: specifically, the suitability of this index is evaluated against other variables commonly used in clinical practice. Regarding the modeling of anesthesia, this thesis presents an adaptive model that allows characterizing the pharmacological interaction effects between the hypnotic and analgesic drug on the depth of hypnosis. In addition, the proposed model takes into account both inter- and intra-patient variabilities observed in the response of the subjects. Finally, this work presents the synthesis of a robust optimal PID controller for regulating the depth of hypnosis by considering the effect of the uncertainties derived from the patient's pharmacological response. Moreover, a study is conducted on the limitations introduced when using a PID controller versus the development of higher order solutions under the same clinical and technical considerations.