Patent Application
20240339210
The present invention relates to a computer system and a corresponding method for medical monitoring of a patient, wherein mathematical models are used to provide suggestions for the causes of changes in measured values from the patient or even identify worsening of the patient in the presence or absence of changes in the measured values from the patient.
Patients with severe acute illness, often requiring mechanical support for ventilation, usually reside in the intensive care unit (ICU) of the hospital. Here they are intensively monitored using a range of machines which measure physiological variables related to vital functions. These include measurement of the oxygenation and acid-base status of the blood, circulation, metabolism, respiratory function etc.
These measurements are interpreted by the clinician who then assesses the state of the patient and determines whether a change in therapeutic intervention is necessary. The measurement instruments can, in some instances, support the clinician in this interpretation through the use of alarms. For example, reduction in arterial oxygenation beyond a certain level e.g. 90% which is usually monitored by a pulse oximeter finger clip, can cause an alarm. The same is true of a reduction in blood pressure in the measurement of circulation, or an increase in inspiratory pressure in the measurement of pulmonary function and numerous similar alarms.
Common amongst them is that these alarms are based upon the variation in a single measured variable characterizing a single physiological system and often from the single device on which the measurement was taken.
A further type of alarm present in the ICU is that due to sensor detachment or equipment failure. Patients can often be restless and these alarms are therefore necessary and common in, for example, the situation where a pulse oximeter detaches from the patients finger. However, the detachment itself of such a sensor can be difficult to distinguishing from a change in physiological state of the patient.
US patent application 2007/0000494 (invented by Michael J. Banner et al., assigned to University of Florida Research Foundation) discloses a ventilator monitoring system and a method of using same. The invention described and shown in the specification and drawings include a system and method for monitoring the ventilation support provided by a ventilator that is supplying a breathing gas to a patient via a breathing circuit that is in fluid communication with the lungs of the patient. To solve the problem of appropriate adjustments of the ventilation support and advice to clinician in that respect, a neural network adapted for learning from earlier patient data is further described. After so-called training of the neural network implemented on a processor, it may be applied for recommendations of ventilation support settings to a clinician being faced with the task of adjusting ventilation settings. However, the described system also fails to provide more intelligent information behind any deviations or abnormalities observed during mechanical ventilation of a patient.
Hence, an improved monitoring method would be advantageous, and in particular a more efficient and/or reliable method would be advantageous.
It is a further object of the present invention to provide an alternative to the prior art. In particular, it may be seen as an object of the present invention to provide a computer system and a corresponding method that solves the above mentioned problems of the prior art with monitoring of patients, particular patients in intensive care units (ICUs).
The prior art alarm strategies, based upon single measurements or sensor detachment, are intended to aid in monitoring the patient. However, these strategies may have drawbacks.
Firstly, alarms based on single measurement variables are a poor help in aiding the clinician in understanding the deeper cause of the problem. For example, a decrease in oxygenation of arterial blood can be due to numerous causes. The patient may have become active or developed a fever, in which case metabolic demand may have increased. The patient's lung function may have deteriorated such that oxygen transports less freely from the lungs to blood, or it may just be the case that the clinician has inadvertently reduced inspired oxygen to a level where arterial values are compromised. Measurements and alarms indicating reduced arterial oxygenation are therefore useful but are not the complete picture.
Secondly, current monitoring and alarm strategies do not direct the clinician as to the requirement for further measurement. For example, an increase in carbon dioxide levels in expiratory gas, can indicate increased metabolic production of CO2, a reduced respiration, or a change in the acid-base status of the blood.
Furthermore, clinicians may get stressed by the vast number of settings, physiological parameters, screens, and the relationship between these values and their impact on therapeutic decisions in relation to conflicting goals. Stress and lack of overview can lead to errors in an ICU [1].
Thus, the above described object and several other objects are intended to be obtained in a first aspect of the invention by providing a computer system for medical monitoring of a patient using one or more physiological models implemented on the computer system, the system being arranged for:
Thus, the computer system of the invention provides an indication for the underlying cause for a change in a measured value from a patient. Furthermore, the computer system is also able to provide an indication of underlying physiological changes of the patient in a case where the data values are still inside threshold ranges. This may allow the clinician to take action even before measured values have changed.
In the context of the present invention, the physiological parameters being the result of the modelling may be obtained by fitting the models to the said data. The fitting may be performed numerically and/or analytically. The fitting may be performed continuously, regularly, or at user-defined intervals, possibly in a combination of these time regimes of fitting.
It is to be understood that the indication of the one or more organ systems, and/or one or more physiological systems and/or one or more non-physiological systems causing said deviation may be one possible cause, or a plurality of possible causes for the deviation. It is further to be understood that computer system may be arranged for further weighting a probability of a certain cause before outputting the possible cause to a user, e.g. a clinician. In some cases, the actual probability for a cause is unknown (but above zero), and the possible causes may then all be outputted and listed to a user. Thus, if below a certain probability the cause may be not be outputted to a user. Alternatively or additionally, the cause may be prioritized according their associated probabilities.
It may be mentioned that—in the context of the present invention—the said indications are intended for assisting or guiding e.g. clinician in making decisions of a therapeutic and/or a diagnostic character. Thus, the present invention is not designed to perform an actual diagnosis, merely providing intelligent information, i.e. indications that may assist the clinician in performing the subsequent step of making the intellectual exercise of providing a diagnosis of the patient state, the diagnosis may then be followed be an action of therapeutic character, if needed.
In the context of the present invention, the patient may be a mammal, such as a human, a monkey, a horse, a rodent, etc.
As mentioned above, a change in a data signal to a level above or below a predetermined threshold level may have different causes, making it difficult for the clinician to make the optimal choice in relation to treatment. Table 1 below list a number of non-exclusive examples of both physiological and non-physiological causes for a change in a data signal.
Thus, in an embodiment of the invention the observed deviation is a SpO2 level below a threshold level and
In another embodiment of the invention the observed deviation is a FetCO2 level above a threshold level and
In yet another embodiment of the invention the observed deviation is a mean arterial blood pressure below a threshold level and
In yet an embodiment of the invention the observed deviation is a positive inspiratory pressure above a threshold level
The indication may be based on different values. Thus, in an embodiment the indication of one or more organ systems, one or more physiological systems and/or non-physiological systems causing said deviation is based on
The computer system of the invention may employ further data input. Thus, in yet an embodiment the computer system is further arranged for
The term “organ systems” covers different organs of the patient. In an embodiment the one or more organ systems are chosen from the group consisting of lungs, blood, heart, kidney and tissues.
The term “physiological systems” covers different physiological systems of the patient. Thus, in a further embodiment the one or more physiological systems are chosen from the non-exhaustive group consisting of metabolism, circulation, respiration and renal function and used in a modelling of the patient, i.e. physiological modelling over time It is to be understood that the modelling over time may particularly include the possibility of predicting or estimating one, or more, parameters at a later time.
In the present context the term “physiological model” should be understood as the broadest sense of applied physiological models, the term physiological model here covering any means for linking clinically measurable values mathematically and for the individual patient, that is, that any parameters that needs tuning for the model to describe the individual patient's physiology should be possible to estimate from clinical measurements. In its simplest form such a physiological model can be a linear model of the equation y=a*x with a being the parameter of the model and y and x being clinical measureable values such as the link between pressure (x) and volume (y) with lung compliance (a) linking the two (Volume=Pressure*Compliance). The more complex end of the range include a model encompassing all chemical reaction equations of the buffer systems of human blood with parameters such as rate constants which may be assumed equal for all individuals but with for example Base Excess that should be tuned to the individual patient [2]. The invention may require physiological models of lung mechanics, gas exchange, blood acid-base status, respiratory drive, heart function, kidney function, and/or blood circulation in general.
A physiological model of lung mechanics links flows, volumes and or pressures which may be measured at the mouth, in the oesophagus, at the ventilator, in the ventilator tube, in the airways etc. More complex physiological models may describe the resistances of the respiratory system and the individual elastic characteristics of the alveoli, the lung tissue, the diaphragm, the surfactant, the chest wall, the airways, etc. In control modes, identification of the overall characteristics of the respiratory system such as compliance is straightforward and is routinely measured by the ventilator. In support mode, the volumes inspired and expired by the patient not only depend on lung mechanics and ventilator settings, but also patient respiratory work. Simple models of lung mechanics in support ventilation may therefore describe an effective compliance, which is the combined pressure volume relationship of ventilator and patient work.
A physiological model of gas exchange links measurements of contents of inspired and expired gases to measured contents of these gases in blood (arterial, venous, mixed venous or capillary blood). Often this is mathematically implemented as a set of equations describing the exchange of one, or more, respiratory gases (oxygen, carbon dioxide, nitrogen) or inert gases added to the inspired gas (such as xenon, krypton, nitric oxide, methane etc.) across the alveolar membrane to capillary blood in one or more compartments and the mixing of the blood leaving these compartments with shunted venous blood to constitute the arterial blood, the latter which is often sampled and analysed in the clinical settings. Venous or mixed venous blood may also be sampled in the clinical setting and can also be described by a model by including equations with patient metabolism and blood flow in the model. Estimation of model parameters for gas exchange models, in particular for models with one or more of the aforementioned compartments, requires either invasive measurements such as a mixed venous blood sample from the pulmonary artery or that the gas contents of inspired gas is modified in one or more steps with the patient response to these changes measured continuously and/or at steady state this making it possible to separate the contributions of one or more mechanisms for gas exchange problems these in the model represented by parameters such as intrapulmonary shunt, ventilation/perfusion ratio etc. Gas exchange models often encompass a model of blood acid base status and the binding and transport of respiratory gases in blood.
A model of blood acid-base status and/or oxygen and/or carbon dioxide transport describes either empirically or by reaction equations some or all of the chemical reactions occurring in blood for chemical buffering and binding of gases to haemoglobin. The models may be in so-called steady state describing the reaction equations at equilibrium or dynamical describing the reactions over time. As such, the models link values of blood acid base that can be measured routinely at the bedside (such as oxygen and carbon dioxide pressures and pH) to values not easily obtained (such as base excess, concentrations of buffers, concentrations of respiratory gases etc.).
A physiological parameter should in this context be understood as a value describing the characteristics of a physiological system and can be assumed constant for changes in some clinical variables and/or ventilator settings. For example, pulmonary shunt describes the fraction of blood flow to the lungs not reaching ventilated alveoli representing the most important gas exchange problem in mechanically ventilated patients. Pulmonary shunt is known to remain constant for variation in a number of clinical variables and ventilator settings such as respiratory frequency and tidal volume.
Below are listed, in an illustrative, non-limiting and non-exhaustively manner, some physiological parameters suitable for application in the present invention:
Lung mechanics parameters include lung compliance, respiratory resistances, respiratory system elastance, effective shunt, compliances and resistances of individual components of the respiratory system (alveoli, airways, chest wall, diaphragm, surfactant).
Gas exchange parameters include but is not limited to pulmonary shunt, venous admixture, effective shunt, Ventilation/perfusion (V/Q) ratio, degree of low V/Q, degree of high V/Q, ΔPO2, ΔPCO2, Arterial-end tidal O2 difference, Arterial-end tidal CO2 difference, anatomical dead space, alveolar dead space, physiological dead space, end-expiratory lung volume, functional residual capacity etc.
Blood acid-base parameters include but is not limited to base excess (BE), strong iron difference (SID), haemoglobin concentration, blood volume, rate constants for chemical reactions etc. This includes examples of third data (D3) according to the present invention.
The computer system of the invention may also comprise certain simulations. Thus, in another embodiment the computer system is arranged for, optionally when no physiological parameters have a deviation compared to a related physiological threshold: —performing a simulation of the patient state by keeping constant one or more, possible every, physiological parameters from the modelling and simulating the time development of the patient state. In yet an embodiment said simulation comprises a tuning process of one, or more, model parameters at regularly intervals, possible continuously at least with respect to the available computational resources or frequency of data measurements. In yet a further embodiment said simulation of the patient state over time is applied for comparing one, or more, physiological parameters and/or simulated physiological values corresponding to obtainable data with actual data to assess if the said simulation can adequately describe the patient state now or at a later time.
Besides providing an indication of an underlying cause for the change in measured values or physiological parameters, the computer system may also allow for providing an indication of further tests from which values may be used in the same models or new models to improve the suggested indications. Thus, in an embodiment the computer system being arranged for indicating one or more further patient measurements which can be performed for obtaining further data values which may improve the modelling, the simulation and/or the provided indication. Such further test may be tests which are usually not constantly measured/obtained during a standard monitoring. Thus, both the type of measurements for additional data and/or timing/frequency of such data may be suggested by the computer system. In yet an embodiment the further tests are selected from the non-exhaustive group consisting of arterial or venous blood gas, blood electrolyte values, blood glucose or ketone concentration, cardiac output measurement or other invasive measurements describing blood, circulation, metabolism, respiration or renal function.
Situations may arise where the computer system or the method (for some reason) is not able to describe the patient state. Thus, in an embodiment, if said simulation cannot describe the patient state now or at later time, the computer system being arranged for indicating one or more possible causes therefore. Possible causes could be errors in the previously measured values such that for example measurement of arterial oxygenation is not consistent with measurement of inspired oxygenation or metabolic demand. In a situation such as this one of these measurements may be in error.
As previously mentioned the method or the computer system may be able to give an indication in a change in a physiological parameter without any significant change has been measured in the data values. Thus, in an embodiment no data values lies outside a threshold range, while a change in one or more physiological parameters (organ systems, and/or one or more physiological systems) are indicated. This may be specifically advantageous since this provides the clinician with a warning of a specific failure which would not otherwise be indicated. Thus, in a similar embodiment no data values deviates significantly from previous provided data values from said patient. Thus, the computer system may be working even in the absence of changes in data values.
The number of organ systems and/or physiological systems that the computer system or the method of the invention is able to provide indication of may vary. Thus, in an embodiment the one or more physiological models being capable of modelling the patient over time by outputting a plurality of corresponding physiological parameters derivable from said first, said second and said third data describing the functioning of 2-10 organ systems and/or physiological systems simultaneous, such as 3-10, such as 5-10, such as 2-8, such as 4-8, or such as 4-6. The number organ systems and/or physiological systems that the computer system of the invention is able to provide indication of depends of the number of parameters included in the one or more physiological models or the number of physiological models the system comprises. Again,
The invention is particularly, but not exclusively, advantageous for obtaining that monitoring and alarms may be based upon changes in the patient's physiological state rather than symptomatic changes described by direct measurements as hitherto. Furthermore, simulations performed by the models may be used to identify when the patient is poorly understood and to direct the clinician as to the appropriate further measurements.
In a second aspect, the present invention relates to a method for medical monitoring of a patient using one or more physiological models implemented on a computer system, the method comprising:
In a third aspect, the present invention relates to a computer program product being adapted to enable a computer system comprising at least one computer having data storage means in connection therewith to implement a method according to the second aspect.
The second and/or the third aspect may be adapted for implementation on an entity selected from the group consisting of a mechanical ventilation monitoring system, an intensive care monitor system, a portable computing device, a handheld mobile device, such as a tablet, and an electronic patient journal system, etc., as it will readily be the skilled person once the teaching and general principle of the invention is comprehended. In separate aspects, the invention may also relate to a mechanical ventilation monitoring system, an intensive care monitor system, a portable computing device or similar as such for implementing the invention.
The third aspect of the invention is particularly, but not exclusively, advantageous in that the present invention may be accomplished by a computer program product enabling a computer system to carry out the operations of the first aspect of the invention when down—or uploaded into the computer system. Such a computer program product may be provided on any kind of computer readable medium, or through a network.
The individual aspects of the present invention may each be combined with any of the other aspects. These and other aspects of the invention will be apparent from the following description with reference to the described embodiments.
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.
This invention is a method/computer system for providing intelligent monitoring, and in doing so for providing intelligent alarms and directed measurements. Key to this invention is the application of physiological models to integrate the available measurements, and estimate parameters which describe the underlying causality. Simulations using these models can then be used to identify when patients underlying physiology changes, i.e. when measurements no longer match simulations, and as such to direct measurement strategy.
The situation may also arise where the input data values D1, D2, D3 do not lie outside any threshold ranges, but that after the input values have been used in one or more physiologic models one (or more) of the output physiological parameters lies outside (or deviates significantly, DEV_P) their specific threshold range. In this situation the system may also set of an alarm 3, e.g. a visible alarm, an audio alarm, and/or a tactile alarm, etc., and providing an indication IND of an underlying problem not detectable by only monitoring the data values. This situation is displayed in
An embodiment of the method of the invention is illustrated in
For the model presented in
For a patient presenting at the ICU it is possible to measure continuously respiratory gasses (oxygen and carbon dioxide), respiratory flows and pressures, arterial oxygenation, blood pressures, cardiac output and tissue production of CO2 and consumption of O2. In addition, it is usual to measure arterial blood acid-base and oxygenation status periodically. These measurements can be used to identify the parameters of the models using model fitting techniques. By doing this periodically over time (step 2,
In contrast,
Thus, in
When models have been fitted to clinical data they can then be used to perform simulations (step 5,
The next step is in directing the clinician to which measurements (step 7,
The invention can be implemented by means of hardware, software, firmware or any combination of these. The invention or some of the features thereof can also be implemented as software running on one or more data processors and/or digital signal processors.
The individual elements of an embodiment of the invention may be physically, functionally and logically implemented in any suitable way such as in a single unit, in a plurality of units or as part of separate functional units. The invention may be implemented in a single unit, or be both physically and functionally distributed between different units and processors.
Below are listed some embodiments of the invention:
1. A method for medical monitoring of a patient (P) using one or more physiological models (MOD1, MOD2) implemented on a computer system, the method comprising:
i. performing a mathematical physiological modelling of the patient using said first (D1), second (D2) and third (D3) data as a function of time, wherein the one or more physiological models (MOD1, MOD2) being capable of modelling the patient over time by outputting a plurality of corresponding physiological parameters derivable from said first, said second and said third data describing the functioning of at least two organ systems and/or physiological systems simultaneous,
Although the present invention has been described in connection with the specified embodiments, it should not be construed as being in any way limited to the presented examples. The scope of the present invention is to be interpreted in the light of the accompanying claim set. In the context of the claims, the terms “comprising” or “comprises” do not exclude other possible elements or steps. Also, the mentioning of references such as “a” or “an” etc. should not be construed as excluding a plurality. The use of reference signs in the claims with respect to elements indicated in the figures shall also not be construed as limiting the scope of the invention. Furthermore, individual features mentioned in different claims, may possibly be advantageously combined, and the mentioning of these features in different claims does not exclude that a combination of features is not possible and advantageous.
| Number | Date | Country | Kind |
|---|---|---|---|
| PA 2013 70748 | Dec 2013 | DK | national |
This application is a Continuation of U.S. application Ser. No. 15/101,497, filed Jun. 3, 2016, which is the U.S. national stage of PCT/DK2014/050414 filed Dec. 5, 2014, which claims priority of patent application DK PA 2013 70748 filed Dec. 5, 2013. The entire content of each application is incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 16872005 | May 2020 | US |
| Child | 18629221 | US | |
| Parent | 15101497 | Jun 2016 | US |
| Child | 16872005 | US |
