Back to main page of IDEA

Project tasks

Task 1 - Project management and dissemination

This task aims at performing the management issues related with the project. This includes financial coordination, technical coordination and project dissemination. In order to ensure coordination among the different research groups involved, regular meetings with all the groups involved in the project will be organised at a convenient location to all partners. There will be meetings of two types: i) Meetings of the representatives of each partner, that will take place every months in order to review the project progress (both in technical and administrative/financial terms), analyse the difficulties, correct any deviations and make detailed plans for the near future. ii) Meetings of researchers to produce joint work or exchange results. These include missions to clinical tests.Furthermore, there will be exchange of information using electronic means (email and WWW) and more restricted work meetings. The dissemination of the project results will done: i) Through the publication of scientific papers; ii) By creating a WWW page; iii) By organising a final scientific workshop and a dissemination meeting for medical doctors. A plan of the major conferences to attend and of the journal papers to submit will be established and reviewed as the work progresses.

Results:Adequate management of the project, including the coordination between the partners to execute the different tasks on time and the financial execution according to the budget scheduled.


Task 2 - Specifications and risk assessment

In this task, a detailed critical literature overview of the state of the art in automatic anaesthesia will be performed. A list of the significant publications to the subject will be organised and used as basis for the review, in which the references corresponding to the main problems will be grouped. The main problems and methods referred in the literature will be identified. On the basis of this literature review and on the basis of discussions with the anaesthesiologists , both for human and veterinary anaesthesia, sets of specifications and risk factors (what are the risk factors and how may them be minimised) concerning automatic anaesthesia will be elaborated. The design of the system to develop will have to comply with these factors.

Results:Report with a detailed critical literature overview of the state of the art in automatic anaesthesia.


Task 3 - Control and supervision algorithm development

In this task, the development of the algorithms for control and supervision and the definition of the overall control block diagram will be made.

The adaptive and nonlinear algorithms will be such as to be incorporated in a robust and reliable control system for anaesthesia that allows to regulate the variables related with the degree of unconsciousness (in particular based on the state entropy index) and with the level of neuromuscular blockade. For this sake, adaptive control methods combining multiple model techniques with predictive techniques are going to be developed. Furthermore, algorithms for tackling interruptions of feedback measurements and that allow reconstruction of the missing signal, supposed unknown or disturbed with outliers are going to be incorporated in the control system. The algorithms developed will be gradually integrated in the automatic control system that is implemented in the clinical environment. Finally, the results obtained using the state entropy as an index of hypnosis are going to be compared with the ones obtained using BIS (proposed in the project “Automation in anaesthesia: monitoring, modelling and control”).

The algorithms of the supervisor perform the functions of a Fault Detection Monitor (FDM). This block verifies the correctness of the actions undertaken by the controller block, while being able to include the adequate high level information on the patient’s state to the practitioner provided by software developed in other projects. In particular, the FDM will detect malfunctions of sensors and actuators (e. g. blocking, outliers or unexpected behaviour), oscillatory and other poor performance behaviour of the feedback loop. The methods to apply are from the real of Fault Detection and Isolation for control systems.

The general methodology to follow is the one that is common to control systems development. On the basis of plant structural knowledge control algorithms will be selected and adapted to the class of problems at hand. Testing will be performed using MATLAB/SIMULINK or equivalent software on a detailed model of anaesthesia (obtained using results from other projects). The algorithms will be reviewed on the basis of the achieved results, the process of simulation/redesign being an iterative one. On parallel, a theoretical analysis (stability, performance) of the algorithms will be made in order to reveal limitations as well as to suggest new structures.

Results:Algorithms for control and supervision of automatic anaesthesia and definition of the block diagram of the global control system. The main variables to control are DoA and neuromuscular blockade.


Task 4 - Hardware assembly and installation

Specification, assembly and installation of the hardware platform of the integrated system. This platform will consist of an autonomous unit with two processing boards, one dedicated to control and the other to supervision. The unit will have to comply with the applicable standards for electrical equipment. The sampling rate will be of the order of a few seconds (typically 5 seconds for DoA and 20 seconds for neuromuscular blockade).

The processing board dedicated to the control will be based on a high-performance processor that supports high-precision floating-point computations. The supervision board will be based on a general purpose processor, able to gather data and information directly from the sensors and from the processing board dedicated to control. This last processing board can dynamically change processing parameters for the control, according to the gathered information.

Results:Hardware platform with 2 processors and interfaces to be used as a prototype platform.


Task 5 - System testing

In this task selected control algorithms and the overall system will be tested and validated in rats and, to a smaller extent, in human patients. For rats, the methodology is as follows: Male, 300 g wistar rats will be anaesthetised in the induction box with 4% isoflurane in oxygen. Once the rat is unconscious and has lost the righting reflex it will be place on the heating blanket with the isoflurane at 2.5-3% delivered via a face mask for surgical preparation. Electrodos will be located over the nasal bone, bregma and lambda. The electroencepahalograpy will be analyzed by a BIS and Entropy monitor.

In the first experiments animals will be anaesthetised with different infusion rates of propofol, in the second phase studies the brain responses will be recorded during induction and recovery. Induction protocols will use isoflurane (first studies) in an induction chamber or iv propofol (second studies). During maintenance, pulse rate, blood pressure and capnographic data will be collected using an electronic monitor following catheterisation of the femoral artery and endotracheal intubation. Animals will be intubated with a 14 gauge intravenous cannula. Temperature, respiratory rate, end-tidal carbon dioxide (CO2) and minimum inspired CO2 will be monitored continuously, with intermittent positive pressure ventilation carried out using a volume cycled rodent ventilator (Harvard Apparatus Ltd, Edenbridge, Kent, UK) with a ventilation rate of 45 to 50 breaths per minute with a tidal volume of 4 ml, and a fresh gas flow of 300ml/min. Animals will have a one or two 24G intravenous cannulae inserted into the lateral tail veins for separate administration of propofol and fentanyl.

Studies one and two described in the results will be followed by tests of developed controllers to maintain a stable level of anaesthesia in rats.

Sample size will be based o power and sample size calculation using PS Version 2.1.30 (http://www.mc.vanderbilt.edu/prevmed/ps/index.htm). This will reduce the number o animals. Research will be developed after ethical and institutional approved committee.

For human patients, comparisons will be made between the manual procedure and the automatic one. Again, research will be developed only after approval by the competent ethical and institutional committee.

Results:The results of this task will consist in the testing and validation of the partial algorithms and the overall system for automatic anaesthesia in rats and human patients. Only the algorithms passing simulation tests will be selected for actual clinical test. For rats, the expected results correspond to the following objectives: 1. The first objective is to evaluate the effects of different propofol infusions on the rats BIS and Entropy responses. The hypotheses that these responses show a dose dependent response similar to humans will be validated.
2. The second objective is to test developed controllers in a group of rats to maintain a stable level of anaesthesia during maintenance of anaesthesia.
3. The third objective is to follow induction and recovery of anaesthesia and describe the way anaesthesia is expressed in monitored EEG responses.
4. The fourth objective is to test developed controllers in a group of rats during all phase of anaesthesia.


Task 6 - Software development

The objective of this task is to develop the software needed to the autonomous integrated system. This software will run in the platform developed in task 5 and will use the control and supervision algorithms developed in tasks 3 and 4. It implies the translation of the algorithms from a development language (such as MATLAB) to a suitable language for implementing the autonomous system (such as C or C++). The software will use already available modules from previous projects, in particular for interfacing the sensors and actuators through USB ports, as much as possible. Although the sampling intervals are slow (as mentioned above, 5 seconds for DoA and 20 seconds for Neuromuscular Blockade), the use of Model Predictive Control, that implies the optimisation on line of a cost in the presence of constraints will require special cares in software requirements.

Results: Autonomous software package running on the hardware platform specified on task 4, that the embodies the control and supervision algorithms for automatic anaesthesia and the interface to the user.

Task 7 - Final report

Elaboration of the final report on the basis of the written documentation issued during the different tasks of the project.

Results: Final project report.