Jörn Kretschmer

Dynamically generated models for medical decision support systems

Doctors applying mechanical ventilation need to find the best balance between benefit and risk for the patient. Mathematical models simulating patients reactions to alterations in the ventilation regime may be employed. A framework is introduced that is able to dynamically combine mathematical models from different model families to form a complex interacting model system. Each of these families consists of submodels differing in complexity of dynamics formulation or anatomical/geometrical resolution. The interaction of model systems reveals qualitatively varying results depending on the complexity of the involved models. Realistic overlaying of respiratory and cardiovascular rhythms can be detected in blood gas concentrations.

Positioning of electrode plane systematically influences EIT imaging.

Up to now, the impact of electrode positioning on electrical impedance tomography (EIT) had not been systematically analyzed due to the lack of a reference method. The aim of the study was to determine the impact of electrode positioning on EIT imaging in spontaneously breathing subjects at different ventilation levels with our novel lung function measurement setup combining EIT and body plethysmography. EIT measurements were conducted in three transverse planes between the 3rd and 4th intercostal space (ICS), at the 5th ICS and between the 6th and 7th ICS (named as cranial, middle and caudal) on 12 healthy subjects. Pulmonary function tests were performed simultaneously by body plethysmography to determine functional residual capacity (FRC), vital capacity (VC), tidal volume (VT), expiratory reserve volume (ERV), and inspiratory reserve volume (IRV). Ratios of impedance changes and body plethysmographic volumes were calculated for every thorax plane (ΔIERV/ERV, ΔIVT/VT and ΔIIRV/IRV). In all measurements of a subject, FRC values and VC values differed ≤5%, which confirmed that subjects were breathing at comparable end-expiratory levels and with similar efforts. In the cranial thorax plane the normalized ΔIERV/ERV ratio in all subjects was significantly higher than the normalized ΔIIRV/IRV ratio whereas the opposite was found in the caudal chest plane. No significant difference between the two normalized ratios was found in the middle thoracic plane. Depending on electrode positioning, impedance to volume ratios may either increase or decrease in the same lung condition, which may lead to opposite clinical decisions.

A simple gas exchange model predicting arterial oxygen content for various FiO 2 levels

The application of mechanical ventilation is a life-saving routine therapy that allows the patient to overcome the physiological impact of surgeries, trauma or critical illness by ensuring vital oxygenation and carbon dioxide removal. Above a certain level of minute ventilation (usually set to ensure acceptable carbon dioxide removal and oxygenation) oxygenation is only marginally affected by a further increase in minute ventilation. Thus, oxygenation is predominantly influenced by inspiratory oxygen fraction (FiO2) Usually, finding the appropriate setting is a trial-and-error procedure, as the clinician is unaware of the exact value that needs to be set in order to reach the desired arterial oxygen partial pressures (PaO2) in the patient. Mathematical models of physiological processes in the human body may be used to predict patient reactions towards alterations in the therapy regime. These predictions can be exploited by Medical Decision Support Systems to find optimal therapy settings. A simple mathematical model is presented, that allows calculation of a patients shunt fraction, i.e. the percentage of blood that is not participating in lung gas exchange. On this basis, it predicts PaO2 at various FiO2-levels and thus allows reaching desired PaO2 in just one step. Due to its simple design it does not require complicated - and possibly error-prone - parameter identification procedures, thus allowing its application at the bedside. Retrospective analysis of oxygenation data from a patient data management system showed that the presented model predicted PaO2 with less than 10% deviation in 23 out of 29 measurements, proving the practical applicability of the presented model approach.

Efficient computation of interacting model systems

Physiological processes in the human body can be predicted by mathematical models. Medical Decision Support Systems (MDSS) might exploit these predictions when optimizing therapy settings. In critically ill patients depending on mechanical ventilation, these predictions should also consider other organ systems of the human body. In a previously presented framework we combine elements of three model families: respiratory mechanics, cardiovascular dynamics and gas exchange. Computing combinations of moderately complex submodels showed to be computationally costly thus limiting the applicability of those model combinations in an MDSS. A decoupled computing approach was therefore developed, which enables individual evaluation of every submodel. Direct model interaction is not possible in separate calculations. Therefore, interface signals need to be substituted by estimates. These estimates are iteratively improved by increasing model detail in every iteration exploiting the hierarchical structure of the implemented model families. Simulation error converged to a minimum after three iterations. Maximum simulation error showed to be 1.44% compared to the original common coupled computing approach. Simulation error was found to be below measurement noise generally found in clinical data. Simulation time was reduced by factor 34 using one iteration and factor 13 using three iterations. Following the proposed calculation scheme moderately complex model combinations seem to be applicable for model based decision support.

Kinect Based Physiotherapy System for Home Use

Abstract In physiotherapy, rehabilitation outcome is majorly dependent on the patient continuing exercises at home. To support a continuous and correct execution of exercises composed by the physiotherapist it is important that the patient stays motivated. With the emergence of game consoles such as Nintendo Wii, Sony PlayStation or Microsoft Xbox360 that employ special controllers or camera based motion recognition as means of user input those technologies have also been found to be interesting for other real-life applications. We present a concept to employ the Microsoft Kinect system as means to support patients during physiotherapy exercises at home. The system is intended to allow a physiotherapist to compose an individual set of exercises and to control the correct execution of those exercises through tracking the patient’s motions.

Hierarchical individualization of a recruitment model with a viscoelastic component for ARDS patients

Patient-specific mathematical models of respiratory mechanics enable substantial insight into patient state and pulmonary dynamics that are not directly measurable. Thus they offer potential e.g. to predict the outcome of ventilator settings for Acute Respiratory Distress Syndrome (ARDS) patients. In this work, an existing static recruitment model is extended by viscoelastic components allowing model simulations in various ventilation scenarios. A hierarchical approach is used to identify the model with measured data of 12 ARDS patients under static and dynamic conditions. Identified parameter values were physiologically plausible and reproduced the measured pressure responses with a median Coefficient of Determination (CD) of 0.972 in the dynamic and 0.992 in the static maneuver. Overall, the model presented incorporates physiological mechanisms, captures ARDS dynamics and viscoelastic tissue properties and is valid under various ventilation patterns.

A Family of Physiological Models to Simulate Human Gas Exchange.

Mathematical models are a widely accepted tool to simulate physiological processes in the human body and to predict patient response towards changes in the therapy regime. These results might be exploited for medical decision support with the goal of finding optimal settings for an individual patient. To allow for an optimal reproduction of patient physiology in all possible situations, rather complex model descriptions are necessary. However, these all-embracing models require an extensive amount of measurements for robust parameter identification and might be computationally costly. Therefore, it would be beneficial to provide multiple model versions each differing in simulation focus and complexity which the decision support system can choose from. We are thus presenting a hierarchically ordered family of gas exchange models and we are showing examples of applications for some of these.

The Influence of Airway Resistance in the Dynamic Elastance Model

The selection of optimal positive end-expiratory pressure (PEEP) levels during ventilation therapy of patients with ARDS (acute respiratory distress syndrome) remains a problem for clinicians. A particular mooted strategy states that minimizing the energy transferred to the lung during mechanical ventilation could potentially be used to determine the optimal, patient-specific PEEP levels. The dynamic elastance model of pulmonary mechanics could potentially be used to minimize the energy by localization of the patients’ minimum dynamic elastance range.

Effects of Different Models and Different Respiratory Manoeuvres in Respiratory Mechanics Estimation

The aim of mechanical ventilation (MV) is to provide sufficient breathing support for patients with respiratory failure in the intensive care unit (ICU). However, applying inappropriate ventilation parameters can result in ventilator induced lung injury. To prevent this, respiratory mechanics such as elastance and resistance can be estimated at the bedside to help guide MV parameters using respiratory mechanics models. Different models or methods provide different information and each have their own advantages and disadvantages. In this study, respiratory mechanics of 9 respiratory failure patients were estimated using the simple first order model (FOM) and viscoelastic model (VEM). These patients undergo different respiratory manoeuvres and their estimated respiratory mechanics using these models are studied and compared with a standard clinical method in estimating respiratory mechanics. The results showed that both models were able to capture patient-specific mechanics and responses. The VEM was able to provide higher correlation to the standard clinical method compared to FOM.

A Modular Patient Simulator for Evaluation of Decision Support Algorithms in Mechanically Ventilated Patients

Mechanical ventilation is a life-saving intervention, which, despite being routinely used in ICUs, poses the risk of causing further damage to the lung tissue if the ventilator is set inappropriately. Medical decision support systems may help in optimizing ventilator settings according to therapy goals given by the clinician. Before using the decision support algorithms in commercially available systems, extensive tests are necessary to ensure patient safety and correct decision making. Model-based patient simulators can assist in evaluating such decision support systems by creating different clinical scenarios. We propose a new Java based patient simulator that implements various models of respiratory mechanics, gas exchange and cardiovascular dynamics to form a complex patient model. The implemented models interact with one another to allow simulation of the ventilators influence on various physiological processes. Model simulations are running in real-time and simulation results can be extracted via multiple interfaces. Each of the implemented models has been validated to exhibit physiologically correct behavior. Results of the combined model system also showed to be physiologically plausible.