Decision support system for lung ventilator settings

Abstract

A ventilator system is capable of displaying complex information patterns in a GUI, thereby allowing a clinician to get subtract complex information from multiple parameters inputs.

Claims

1. A mechanical ventilation system for respiration aid of an associated patient, the system being adapted for providing decision support for choosing a ventilation strategy of said associated patient, the system comprising: a ventilator capable of mechanically ventilating said associated patient with air and/or one or more medical gases; a control, the ventilator being controllable by said control by operational connection thereto; the control configured to receive a first data indicative of parameters of inspired gas of said associated patient; the control further configured to receive a second data indicative of a respiratory feedback of said associated patient in expired gas; the control further configured to receive a third data indicative of one or more blood values of said associated patient, the control applying a set of preference functions to convert the first data, the third data, and the second data of the associated patient into corresponding scoring values; the system comprising a graphical user interface (GUI) with a multi-dimensional coordinate system, wherein multiple pairs of scoring values are arranged for being displayed in the coordinate system, wherein each pair of scoring values comprises a first scoring value and a second scoring value, the first and second scoring values have conflicting effects in response to over-ventilation and under-ventilation in the graphical user interface, wherein: the first scoring value is a translated variable related to the over-ventilation of the associated patient, the first scoring value being displayed at a first axis indicative of the over-ventilation in a first direction in the GUI, the first scoring value being displayed at a first distance from a first starting point corresponding to the first scoring value, and the second scoring value is a translated variable related to the under-ventilation of the associated patient, the second scoring value being displayed at a second axis indicative of the under-ventilation in a second direction in the GUI, the second scoring value being displayed at a second distance from a second starting point corresponding to the second scoring value, wherein the first scoring value and the second scoring value in each pair of scoring values represent opposite clinical preferences, and different pairs of scoring values represent clinical preferences associated with different physiological variables.

2. The system according to claim 1, wherein each pair of plotted scoring values have a common starting point in the multi-dimensional coordinate system, wherein the first starting point and the second starting point are the common starting point.

3. The system according to claim 1, wherein the first and second scoring values displayed closer to the first and second starting points, respectively, are indicative of lower associated patient risk than values plotted at greater distance from the first and second starting points, respectively.

4. The system according to claim 1, wherein the first and second scoring values of each pair are displayed in opposite directions.

5. The system according to claim 1, wherein the multi-dimensional coordinate system has an outer shape being a polygon, each corner in the polygon and the center of the polygon representing a scoring axis of the scoring values, along which scoring axis the scoring values are plotted.

6. The system according to claim 5, wherein the polygon has an even number of corners.

7. The system according to claim 5, wherein the GUI presents a value and/or indicator for an area formed by connecting the plotted/displayed first and second scoring values connected with lines forming the polygon.

8. The system according to claim 1, wherein the multi-dimensional coordinate system is a circle or circular shape, and wherein the first and second scoring values are plotted or displayed along a circle axis between a circumference of the circle or circular shape and a center of the circle or circular shape.

9. The system according to claim 1, wherein the first and second starting points for each pair begin at different points at a line in the multi-dimensional coordinate system dividing the coordinate system into two halves, one half indicative of the over-ventilation of the associated patient and the second half indicative of the under-ventilation of the associated patient.

10. The system according to claim 1, wherein the first and second scoring values represent patient risk values wherein all the patient risk values have been normalized to have comparable patient risk values.

11. The system according to claim 1, wherein the coordinate system is divided by a line through the center, dividing the coordinate system into two halves, one half indicative of the over-ventilation of the associated patient and the second half indicative of the under-ventilation of the associated patient.

12. The system according to claim 1, wherein the neighbouring displayed scoring values are connected with lines or curves forming an area in the GUI.

13. The system according to claim 1, wherein the system is arranged for presenting an overview over the first and second scoring values and/or areas determined at earlier time points.

14. The system according to claim 1, wherein decreasing areas over time is indicative of improved ventilation parameters and wherein increasing areas over time is indicative of suboptimal ventilation parameters.

15. The system according to claim 1, wherein the system is arranged to output the first and second scoring values and/or areas in the GUI based on suggested input ventilation parameters by a user.

16. The system according to claim 1, wherein the system is arranged to output an advice for a ventilation strategy of the associated patient and display modelled additional scoring values and/or areas based on said advice, the system using a physiological model to generate said advice.

17. The system according to claim 1, wherein the first and second scoring values have the conflicting effects with respect to: mechanical lung trauma vs. acidosis; oxygen toxicity vs. low oxygenation; stress vs. ventilator dependency; Volutrauma vs. atelectrauma and/or Alveolar derecruitment vs. adverse heamodynamic effects of high ventilator pressure.

18. The system according to claim 1, wherein the first and second scoring values have the conflicting effects with respect to: mechanical lung trauma vs. acidosis; oxygen toxicity vs. low oxygenation; and/or stress vs. ventilator dependency.

19. The system according to claim 1, wherein the system is arranged for receiving one or more therapeutic input parameters relating to the associated patient, wherein the system is arranged to recalibrate the preference functions based on said therapeutic input parameters, thereby recalibrating the scoring values generated by the system.

20. The system according to claim 19, wherein the system is adapted for changing the weight of each preference function based on the therapeutic input parameters.

21. The system according to claim 19, wherein the therapeutic input parameters are selected from a group consisting of information on head trauma, age, sex, clinical history, medication, and/or associated patient group.

22. The system according to claim 1, wherein a colouring of lines and/or areas are indicative of a total patient risk.

23. The system according to claim 1, wherein a colouring of lines and/or areas are indicative of extra patient risk in addition to risks presented on the axes.

24. A computer system for cooperating with and/or controlling an associated mechanical ventilation system for respiration aid of an associated patient, the associated mechanical ventilation system being of the type having: a ventilator capable of mechanically ventilating said associated patient with air and/or one or more medical gases; a first control, the ventilator being connectable to said first control by operational connection thereto; the first control configured to receive a first data indicative of parameters of inspired gas of said associated patient; the first control further configured to receive a second data indicative of a respiratory feedback of said associated patient in expired gas; the first control further configured to receive a third data indicative of one or more blood values of said associated patient, the computer system being adapted for providing decision support for choosing a ventilation strategy of said associated patient, the computer system comprising: a second control applying a set of preference functions to convert the first data, the third data, and the second data of the associated patient into corresponding scoring values; and a graphical user interface (GUI) with a multi-dimensional coordinate system, wherein multiple pairs of scoring values are arranged for being displayed in the coordinate system, wherein each pair of scoring values comprises a first scoring value and a second scoring value, the first and second scoring values have conflicting effects in response to over-ventilation and under-ventilation in the graphical user interface, wherein: the first scoring value is a translated variable related to the over-ventilation of the associated patient, the first scoring value being displayed at a first axis indicative of the over-ventilation in a first direction in the GUI, the first scoring value being displayed at a first distance from a first starting point corresponding to the first scoring value, and the second scoring value is a translated variable related to the under-ventilation of the associated patient, the second scoring value being displayed at a second axis indicative of the under-ventilation in a second direction in the GUI, the second scoring value being displayed at a second distance from a second starting point corresponding to the second scoring value, wherein the first scoring value and the second scoring value in each pair of scoring values represent opposite clinical preferences, and different pairs of scoring values represent clinical preferences associated with different physiological variables.

25. A non-transitory storage medium comprising a computer program product having instructions embodied thereon, the computer program product, when executed by a computing device or system, causes the computing device or system to: apply a set of preference functions to convert a first data indicative of parameters of inspired gas of a ventilated patient; a second data indicative of a respiratory feedback of the ventilated patient in expired gas; and a third data indicative of one or more blood values of the ventilated patient, into corresponding scoring values, display multiple pairs of scoring values in a multi-dimensional coordinate system, wherein each pair of scoring values comprises a first scoring value and a second scoring value, the first and second scoring values have conflicting effects in response to over-ventilation and under-ventilation in a graphical user interface, wherein: the first scoring value is a translated variable related to the over-ventilation of the ventilated patient, the first scoring value being displayed at a first axis indicative of the over-ventilation in a first direction in the multi-dimensional coordinate system, the first scoring value being displayed at a first distance from a first starting point corresponding to the first scoring value, and the second scoring value is a translated variable related to the under-ventilation of the ventilated patient, the second scoring value being displayed at a second axis indicative of the under-ventilation in a second direction in the multi-dimensional coordinate system, the second scoring value being displayed at a second distance from a second starting point corresponding to the second scoring value, wherein the first scoring value and the second scoring value in each pair of scoring values represent opposite clinical preferences, and different pairs of scoring values represent clinical preferences associated with different physiological variables.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The method according to the invention will now be described in more detail with regard to the accompanying figures. The figures show one way of implementing the present invention and is not to be construed as being limiting to other possible embodiments falling within the scope of the attached claim set.

(2) FIG. 1 is a schematic drawing of a mechanical ventilation system according to the present invention.

(3) FIG. 2 shows a preferred output format in the GUI in a schematic format.

(4) FIG. 3 shows a screenshot of a preferred output format in the GUI.

(5) FIG. 4 shows examples of output formats in the GUI.

(6) FIG. 5 shows the structure of a decision support system according to the invention, illustrating the components of the system (ovals) and the functionality (dashed lines).

(7) FIG. 6 shows examples of preference functions (CPF) where physiological variables and ventilator settings are translated into a unified scale (scoring values).

(8) FIG. 7 illustrates a set of mathematical model components of a decision support system (DSS) including the mathematical representation of a physiological model of respiratory control.

(9) FIG. 8 is a schematic flow chart of a method according to the invention.

(10) FIG. 9 shows examples of output formats in the GUI with each pair being displayed on a line. The areas of the columns and the areas under the graph represent single preference zones (areas) for each pair. The sum of these individual preference zones represents the overall preference zone for the specific patient.

(11) The present invention will now be described in more detail in the following.

DETAILED DESCRIPTION OF THE INVENTION

(12) As described above, the core aspect of the invention is the use of preference functions to calculate scoring values and corresponding preference zones/areas 15 from clinically measured variables allowing integration of relevant mechanical ventilation variables for a patient into a single presentation covering the contrasting preferences related to mechanical ventilation helping the clinician to minimise risk of over-ventilating and under-ventilating the patient.

(13) FIG. 1 is a schematic drawing of a mechanical ventilation system 10 for respiration aid of an associated patient 5, P, the system being adapted for displaying/plotting information of physiological parameters in a manner providing the clinician with information which is difficult to subtract from mere number values. The respiration aid may be fully controlled or supported.

(14) The system comprises ventilator means 11, VENT capable of mechanical ventilating said patient with air and/or one or more medical gases, e.g. oxygen and/or nitrogen. Conventional ventilator systems currently available may be modified or adapted for working in the context of the present invention.

(15) Furthermore, control means 12, CON is comprised in the system 10, the ventilator means 11 being controllable by said control means 12 by operational connection thereto, e.g. appropriate wirings and interfaces as it will be appreciated by the skilled person working with mechanical ventilation.

(16) Additionally, measurement means 11b, M_G2 are arranged for measuring the respiratory feedback of said patient in the expired gas 6b in response to the mechanical ventilation, e.g. respiratory frequency or fraction of expired carbon dioxide commonly abbreviated FECO.sub.2, cf. the list of some well-known abbreviations below. The measurement means are shown as forming part of the ventilator means 11, but could alternatively form an independent entity with respect to the ventilator means without significantly change the basic principle of the present invention. Similarly, the computer system according to the invention may work independently from a mechanical ventilator.

(17) Additionally, measurement means 11b, M_G2 are arranged for measuring parameters of the inspired gas 6a, the first means being capable of delivering first data D1 to said control means. It should be noted that the first data D1 may also include the ventilator settings (Vt_SET).

(18) The measurement means M_G (1 and 2) are capable of delivering the first data D1 and the second data D2 to the control means 12 CON by appropriate connection, by wire, wirelessly or by other suitably data connection.

(19) The control means 12 CON is also capable of operating the ventilation means by providing ventilatory assistance so that said patient 5 P is at least partly breathing spontaneously, and, when providing such ventilatory assistance, the control means being capable of changing one, or more, volume and/or pressure parameters Vt_SET of the ventilator means so as to detect changes in the respiratory feedback in general of the patient by the measurement means M_G (1 and 2).

(20) The control means is further being arranged for receiving third data D3, preferably obtainable from blood analysis of said patient performed by blood measurement means M_B 20, the third data being indicative of the respiratory feedback in the blood of said patient, e.g. pHa, PACO2, PA02 etc. Notice that the blood measurement means M_B 20 is not necessarily comprised in the ventilator system 10 according to the present invention. Rather, the system 10 is adapted for receiving second data D2 from such an entity or device as schematically indicated by the connecting arrow. It is however contemplated that a blood measurement means M_B could be comprised in the system 10 and integrated therein. In this embodiment, the mechanical ventilator system comprises at least the ventilator means VENT 11, the measurement means M_G (1 and 2) 11 (a and b), and the control means CON 12. The physiological model MOD is implemented on the control means, e.g. in an appropriate computing entity or device.

(21) In one variant of the invention, the third data D3 could be estimated or guessed values being indicative of the respiratory feedback in the blood of said patient, preferably based on the medical history and/or present condition of the said patient. Thus, values from previously (earlier same day or previous days) could form the basis of such estimated guess for third data D3.

(22) The control means is adapted for using both the first data D1 indicative of parameters of the inspired gas, the second data D2 indicative of changes of respiratory feedback in expired air 6b, and the third data D3 indicative of the respiratory feedback in the blood 7. By the use of preference functions the system translates D1, D2 and D3 into scoring values displayed in a coordinate system 14 in a graphical user interface (GUI) 13.

(23) The principle of this invention is further exemplified in FIGS. 2 and 3.

(24) FIG. 2A illustrates an example of a multidimensional coordinate system 14, in here displayed as a hexagon, with the scoring values S1-S6 being displayed at an axis from the center, or origin, O towards each corner of the hexagon. The scoring values being displayed in the upper half 14a is indicative of over-ventilation of the patient, whereas the scoring values being displayed in the lower half 14b is indicative of under-ventilation.

(25) It is to be understood that scoring values representative for current status S1, modelled/simulated status S1′ (based on user input) and advice S1″ based on a physiological model implemented in the system may all be displayed. The three different values may be displayed simultaneously or by selection of the user. The same can be the case for other scoring values.

(26) FIG. 2B shows an example of a coordinate system shaped as a square. Such square will then only display two scoring pairs, S1 vs. S4, and S2 vs. S3, respectively. FIG. 2C shows a circular coordinate system, where the scoring values are displayed on an axis going from the center towards a point on the circumference of the circle. In FIG. 2C, an area 15 constituted by the displayed scoring values is also displayed. Such an area may be indicative of the overall quality of the ventilator settings. Again such areas may be displayed for current status, modelled/simulated status (based on user input) and advice based on a physiological model implemented in the system. Thus, the system apply the physiological model (MOD) to generate said advice basing its advice on a model which minimizes the area of the polygon zones, without any of the scoring values exceeds predetermined threshold levels. Thus, the system may try to minimize the scoring values and thus also minimize the area/preference zone in the coordinate system.

(27) FIG. 3 is an actual screenshot of how data may be presented on a monitor, with actual scoring values presented. Areas 15 are also presented. To the right, the multi-dimensional coordinate system 14 is shown, i.e. a hexagon, the upwards direction in the figure being the direction representing over-ventilation OV and the downwards representing under-ventilation UV as in FIG. 2. To the left in FIG. 3, the current values, and simulated values of the respiratory volume in a single breath Vt and the fraction of inhaled oxygen FiO.sub.2, respectively, are shown. Additionally, the ‘Advice’ according to the present is displayed next to the simulated values.

(28) FIG. 4 shows other shapes (pentagon 141, octagon 142, heptagon 143, decagon 144, non-polygon 145) which may be implemented in a system according to the invention. The number of corners in the polygon may depend on the number of scoring pairs included in the system. Notice that for a polygons with an uneven number of corners, e.g. a pentagon, there will be one or more pairs of scorings values displayed, but at least one scoring value will be unpaired. Thus, in the pentagon shown one scoring value, e.g. S5, may be unpaired.

(29) FIG. 5 illustrates the conceptual model behind the system according to the invention. The core of the system is a set of physiological models describing pulmonary gas exchange, acid-base chemistry, lung mechanics etc. the system tunes these models to the individual patient such that they describe accurately current measurements, labelled “Outcome Variables” in the figure.

(30) The ovals illustrates components of the system, which includes ventilator settings (f, Vt, FiO2, LE-ratio, PEEP and PIP); model parameters (shunt, fA2, Vd, compliance, DPG, Hb, COHb, MetHb, temp, Q, VO2 and VCO2); physiological models and their variables (FetCO2, FetO2, SaO2, PaO2, PaCO2, pHa, SvO2, PvO2, PvCO2, and pHv); those variables selected as surrogate outcomes (PIP, SvO2, SaO2, pHv and FiO2); and functions describing clinical preference (barotrauma, hypoxia, acidosis-alkalosis, oxygen toxicity).

(31) Once tuned, the models are used by the system to simulate the effects of changing ventilator settings. These simulations are then used with a set of “Clinical preference functions” (CPF). Some of these functions are illustrated in FIG. 6 and describe clinical opinion as to the outcome variables. For example, an increased inspiratory volume will reduce an acidosis of the blood while detrimentally increasing lung pressure. Appropriate ventilator settings Vt_SET therefore imply a balance between the preferred value of pH weighted against the preferred value of lung pressure. A number of these balances exist, and the clinical preference functions quantitatively weight these, calculating a total score for the patient for any possible ventilation strategy. As the individual preference scores all range between 0 and 0.5 and the patients total score is a sum of these, then the patients score can range between 0 and 2, with 0 being the best condition and 2 the worst condition. The model simulations and preference functions are then used together in a process called “optimization” where the ventilator settings resulting in simulations giving the preferred patient score, i.e. the lowest, are found. These are then said to be the optimal ventilator settings and are a target advice. If the target advice is a substantial distance from current ventilator settings then “advice steps” may be generated, these steps represent a clinically reasonable step toward the target advice without overly aggressive modifications in ventilator settings.

(32) FIG. 7 illustrates the set of mathematical model components of a decision support system (DSS) including the mathematical representation in the form of physiological model of respiratory control that may be applied in the context of the present invention.

(33) The DSS includes models of: pulmonary gas exchange (A); acid-base status and oxygenation of blood (B); acid-base status of CSF (C); cardiac output, and arterial and mixed venous pools (D); interstitial fluid and tissue buffering, and metabolism (E); and chemoreflex model of respiratory control (F).

(34) FIG. 7A illustrates the structure of the model of ventilation and pulmonary gas exchange. FIG. 7B illustrates the structure of the model of oxygenation and acid-base status in the blood. FIG. 7C illustrates Duffin's model of CSF with appropriate model constants [3, 4]. This model includes mass-action equations describing water, phosphate and albumin dissociation plus the formation of bicarbonate and carbonate, and an equation representing electrical neutrality (equations 101-106). In addition, equation (107) is used to describe the equilibration of PCO.sub.2 with arterial blood across the blood-brain barrier. Equation (108) is a modification to Duffin's model which allows calibration of the CSF to conditions where blood bicarbonate, and hence buffer base (BB) or strong ion difference (SID) are modified, such as metabolic acidosis where blood bicarbonate is reduced, or chronic lung disease where blood bicarbonate is increased.

(35) The model illustrated in FIG. 7 includes compartments representing CO.sub.2 transport and storage including the arterial and venous compartments, and circulation represented as cardiac output (Q) (FIG. 7D).

(36) FIG. 7E illustrates the model of interstitial fluid and tissue buffering, and metabolism included in the system. This includes oxygen consumption (VO.sub.2) and carbon dioxide production (VCO.sub.2).

(37) FIG. 7F illustrates the model of respiratory control of Duffin, i.e. equations 109-112. Alveolar ventilation is modeled as a peripheral and central chemoreflex response to arterial and cerebrospinal fluid (CSF) hydrogen ion concentration ([H.sup.+.sub.a] and [H.sup.+.sub.csf]) plus wakefulness drive. Equation (109) describes the peripheral drive (Dp) as a linear function of the difference between [H.sup.+.sub.a] and the peripheral threshold (Tp). The slope of this function (Sp) represents the sensitivity of the peripheral chemoreceptors.

(38) Equation (111) describes central drive (Dc) as a linear function of the difference between [H.sup.+.sub.csf] and the central threshold (Tc). The slope of this function (Sc) represents the sensitivity of central chemoreceptors. Equation (112) describes the alveolar ventilation as the sum of the two chemoreflex drives and the wakefulness drive (Dw). Equation (113) describes the minute ventilation as alveolar ventilation plus ventilation of the dead space, that is equal to the product of tidal volume (Vt) and respiratory frequency (f).

(39) The model described above can be used to simulate respiratory control. The model enables simulation of the control of alveolar ventilation taking into account pulmonary gas exchange, blood and CSF acid-base status, circulation, tissue and interstitial buffering, and metabolism.

(40) FIG. 8 is a schematic flow chart of a method according to the invention. The invention thus relates to a method for operating a mechanical ventilation system for respiration aid of an associated patient 5, P, the system being adapted for providing decision support for choosing a ventilation strategy of said patient, the method comprising: Step 1 providing ventilator means 11, VENT capable of mechanical ventilating said patient with air and/or one or more medical gases; Step 2 providing control means 12, CON, the ventilator means being controllable by said control means by operational connection thereto Vt_SET; Step 3 providing first measurement means 11a, M_G1 arranged for measuring parameters of the inspired gas 6a, the first means being capable of delivering first data D1 to said control means; optionally, second measurement means 11b, M_G2 arranged for measuring the respiratory feedback of said patient in expired gas 6b, the second means being capable of delivering second data D2 to said control means; Step 4 providing third measurement means 20, M_B, arranged for measuring one or more blood values of said patient, the third means being capable of delivering third data D3 to said control means,
the control means applying a set of preference functions CPF to convert the first data D1, the third data D3, and optionally the second data D2, of the patient into corresponding scoring values, S1, . . . , S6;
the system comprising a graphical user interface (GUI) 13 with a multi-dimensional coordinate system 14, wherein at least one pair of scoring values is arranged for being displayed in the coordinate system, wherein each pair of scoring values providing Step 5 a first scoring value, S1, S2, and S6, being a translated variable related to over-ventilation of the patient, the first scoring value being displayed at an axis indicative of over-ventilation in a first direction OV in the GUI, the first scoring value being displayed at a distance from a first starting point O corresponding to the first scoring value, and Step 6 a second scoring value, S3, S4, and S5, being a translated variable related to under-ventilation of the patient, the second scoring value being displayed at an axis being indicative of under-ventilation in a second direction UV in the GUI, the second scoring value being displayed at a distance from a second starting point O corresponding to the second scoring value.

(41) FIG. 9 is illustrations showing alternative ways of displaying the scoring values in the GUI where the values indicative of over-ventilation and under-ventilation are presented on a common horizontal, or vertical axis. Notice that the pairs of scoring values, S1 vs. S4, etc., are then represented at separate centers O1, O2, O3, and O4, in the coordinate system of example A. Similarly, in the coordinate system of examples B, C and D, there are separate centers for the pairs of scoring values. The axis (indicated by the dotted lines in examples A, B, C and D) divides the scoring values into values representative of over- and under-ventilation. It should be noted that embodiments and features described in the context of one of the aspects of the present invention also apply to the other aspects of the invention.

(42) All patent and non-patent references cited in the present application, are hereby incorporated by reference in their entirety.

Glossary

(43) CSF Cerebral spinal fluid Vt Respiratory volume in a single breath, tidal volume Vt_SET Respiratory volume settings for mechanical ventilation, tidal volume FECO.sub.2 Fraction of carbon dioxide in expired gas. FE′CO.sub.2 Fraction of carbon dioxide in expired gas at the end of expiration. PECO.sub.2 Partial pressure of carbon dioxide in expired gas. PE′CO.sub.2 Partial pressure of carbon dioxide in expired gas at the end of expiration. RR respiratory frequency (RR) or, equivalently, duration of breath (including duration of inspiratory or expiratory phase) pHa Arterial blood pH PaCO2 Pressure of carbon dioxide level, SaO2 Oxygen saturation of arterial blood PpO2 Pressure of oxygen in arterial blood

REFERENCES

(44) 1. The Acute Respiratory Distress Syndrome (ARDS) Network (2000) Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N. Engl. J Med. 342:1301-1308. 2. Arnstein F. (1997) Catalogue of human error. Br. J. Anaesth. 79:645-656. 3. Wysocki M, Brunner J X. (2007). Closed-Loop Ventilation: An emerging standard of care? Crit. Care Clin. 23:223-240. 4. Arnstein F. (1997) Catalogue of human error. Br. J. Anaesth. 79:645-656. 5. Allerød C, Rees S E, Rasmussen B S, Karbing D S, Kjrgaard S, Thorgaard P, Andreassen S. (2008). A decision support system for suggesting ventilator settings: Retrospective evaluation in cardiac surgery patients ventilated in the ICU. Comput. Meth. Prog. Biomed. 92:205-212.