J.J. Ross

A hybrid hierarchical decision support system for cardiac surgical intensive care patients. Part II. Clinical implementation and evaluation

OBJECTIVEnPatients emerging from cardiac surgery can display varying degrees of cardiovascular instability arising from potentially complex, multi-factorial and interlinked causes. Stabilization and control of the cardiovascular system are currently managed by healthcare experts using experiential knowledge, and, in some centers, manually inputted decision pathway algorithms. This paper describes a clinical trial undertaken to determine the basic functioning of a clinical decision support system (CDSS) designed and constructed by the authors to facilitate the control of the major cardiovascular components in the early post-operative phase. Part II follows Part Is description of the software and simulation testing of the CDSS, and describes the hardware setup of a patient monitoring and CDSS. The system is evaluated on three post-cardiac surgery intensive care patients whom had all undergone cardio-pulmonary bypass.nnnMETHODSnThe study was approved by the Sheffield Teaching Hospitals National Health Service (NHS) Foundation Trust Research Ethics Committee and conducted at the North Trent cardio-thoracic surgical unit and cardiac intensive care unit (CICU), Northern General Hospital, Sheffield (UK). Patients considered as very likely to require active intervention to support the cardiovascular function following routine cardiac surgery were recruited during pre-operative surgical and anesthetic assessment, giving written informed consent when admitted for their operation. These patients underwent routine induction and maintenance of anesthesia by a non-study consultant anesthetist and the operation performed. There were no restrictions placed on the types of invasive monitoring used, on the use of trans-oesophageal echocardiography, drug selection, or the anesthetic agents selected by the clinicians performing the operations. All patients had full, routine invasive and non-invasive monitoring applied, including electrocardiography, central venous and peripheral arterial catheterisation, urine outputs and central temperature. After chest closure the patients were transferred to the CICU, sedated and ventilated, and the study commenced by the study anesthetist (1st author). The patients were in a clinically stable condition when admitted to the unit, and were attended by the treating clinicians until the handover to the study anesthetist occurred. The LiDCOplus (lithium dilution cardiac output) monitor (LiDCO Limited, Flowers Building, Granta Park, Cambridge CB1 6GU, United Kingdom) was calibrated after attachment to the patients arterial line, and the patients beat-to-beat hemodynamic data transferred to the host laptop computer. The CDSS graphical interface displays the patients clinical details and specific cardiovascular data and prompts the anesthetist to input the target ranges for each parameter, and select a suitable advisor frequency. This is the frequency with which the therapeutic advice is displayed on screen with an audible prompt for a control inputs from the anesthetist. In each case this was selected to be 30s. When the study anesthetist agreed with the CDSS advice (administration of fluid, commencing a drug, altering the drug infusion rate) the syringe motif on the Advisor Infusion Rates panel of the graphical interface was clicked on and the infusion rate immediately and manually inputted to Graseby 3400 pumps. If any disagreement between the anesthetist and the computers advice arose, the syringe motif on the Expert Infusion Rates panel of the preferred drug was clicked on and the experts therapeutic decision (e.g. infusion rate) was entered in the corresponding data field and then applied to the pump. During all trials, data was stored for off-line analysis.nnnRESULTSnThe CDSS successfully selected suitable drug therapies for each case and advised reasonable and appropriate infusion rates such that the study anesthetist did not have to override the suggested CDSS instructions and infusion rates. Under differing clinical conditions the system was able to maintain clinically appropriate and stable control of the cardiovascular system (CVS), with good profiles under noisy physiological measurements, and was readily able to regain control following transient deterioration of the patient hemodynamic parameters (coughing, or during blood sampling).

A hybrid hierarchical decision support system for cardiac surgical intensive care patients. Part I: Physiological modelling and decision support system design

OBJECTIVEnTo develop a clinical decision support system (CDSS) that models the different levels of the clinicians decision-making strategies when controlling post cardiac surgery patients weaned from cardio pulmonary bypass.nnnMETHODSnA clinical trial was conducted to define and elucidate an expert anesthetists decision pathway utilised in controlling this patient population. This data and derived knowledge were used to elicit a decision-making model. The structural framework of the decision-making model is hierarchical, clearly defined, and dynamic. The decision levels are linked to five important components of the cardiovascular physiology in turn, i.e. the systolic blood pressure (SBP), central venous pressure (CVP), systemic vascular resistance (SVR), cardiac output (CO), and heart rate (HR). Progress down the hierarchy is dependent upon the normalisation of each physiological parameter to a value pre-selected by the clinician via fluid, chronotropes or inotropes. Since interventions at each and every level cause changes and disturbances in the other components, the proposed decision support model continuously refers back decision outcomes back to the SBP which is considered to be the overriding supervisory safety component in this hierarchical decision structure. The decision model was then translated into a computerised decision support system prototype and comprehensively tested on a physiological model of the human cardiovascular system. This model was able to reproduce conditions experienced by post-operative cardiac surgery patients including hypertension, hypovolemia, vasodilation and the systemic inflammatory response syndrome (SIRS).nnnRESULTSnIn all the simulated patients scenarios considered the CDSS was able to initiate similar therapeutic interventions to that of the expert, and as a result, was also able to control the hemodynamic parameters to the prescribed target values.

A portable self‐learning fuzzy logic control system for muscle relaxation

We have assessed the practicality and performance of the Vital Signs Paragraph neuromuscular blockade monitor as part of a ‘self‐learning’ fuzzy logic control feedback system used to administer atracurium to a required depth of neuromuscular blockade. Fifteen patients undergoing surgery expected to last longer than 90u2003min entered the study. A Vital Signs Paragraph was used to measure the degree of neuromuscular blockade and control it such that the first twitch of the train‐of‐four was kept at 10% of its baseline value. The controller instructed a Graseby Medical 3400 infusion pump to administer an atracurium infusion to maintain this level of blockade. Five patients (33%) were withdrawn from the study due to inadequate piezo‐electric sensor function. In the remaining 10 patients, the system achieved stable control of neuromuscular blockade with a mean (range) error for the first twitch of the train‐of‐four of −0.45u2003(−1.06 to 0.13)%. The mean atracurium infusion rate ranged from 0.13 to 0.67u2003mg.kg−1.h−1. These results compare reasonably well with previous results using the Datex Relaxograph, whilst the system itself was portable and easy to use. However, the reliability of the system was limited due to variability in the sensitivity of piezoelectric sensors.

TEE for Estimating Pleural Effusion Volumes

We read with interest the article by Howard et al1 describing the use of transesophageal echocardiography to estimate the volume of chronic pleural effusions caused by heart failure. The maximal cross-sectional area (CSAmax) in centimeters squared of the pleural effusion seen on transesophageal echocardiography was recorded. CSAmax was then compared with the actual volume (V) of fluid in milliliters subsequently drained at surgery. Based on 28 data points obtained from 23 patients, the authors conclude that the volume of fluid can be estimated from the following formula: V 4.5 · CSAmax. We have published a similar study in which we measured effusions created after the pleura were surgically opened.2 Our study yielded 111 data points. In addition to CSAmax, we measured the axial length (AL) using the centimeter marks on the transesophageal echocardiographic probe to determine the proximal and distal extent of the effusions relative to the patient’s teeth. We evaluated the formula CSAmax · AL proposed by Swenson and Bull3 and found this to provide a easonably accurate measure of pleural fluid volume for both ightand left-sided effusions. Although recognizing the otential limitations of our data in the context of chronic ffusions, we also did find a good level of agreement with a omputed tomography– based formula derived from chronic ffusion patients.4 We have reanalyzed our data using Howard et al’s formula. These results are presented in Figure 1 and suggest that V 4.5 · CSAmax significantly overestimates the volume of larger effusions, particularly right-sided effusions. We note that Howard et al’s study contained only 3 data points for larger effusions (volume 1,200 mL). In addition, we would question their failure to measure AL as part of

Self-learning Fuzzy Control of Atracurium-induced Neuromuscular Block During Surgery

Abstract We have investigated the clinical application of a self-learning fuzzy controller to neuromuscular block during surgery using the infused drug atracurium. Our intelligent fuzzy controller commences with a completely blank 5 by 5 rule base and performs on-line learning of fuzzy control rules for each patient during the course of surgical procedures. Computer simulation studies were used to determine appropriate controller scaling factors and to perform off-line validation of the physical control system. The first known clinical application of this intelligent controller shows superior control performance over our previously reported non-adaptive fuzzy controller

PHYSIOLOGICAL MODELLING AND ANALYSIS OF THE PULMONARY MICROCIRCULATION IN SEPTIC PATIENTS

Abstract A physiological model integrating a pulsatile cardiovascular system model with a model of the pulmonary capillary fluid exchange based on the Starlings equations is proposed in order to analyse the micro-circulatory physiological alterations which occur during the evolution of sepsis. Sepsis-induced acute lung injury is typically characterized by an increased microvascular permeability and interstitial edema. Pulmonary edema occurs because the capillary filtration capacity increases and the reflection coefficient to proteins decreases causing fluid leak across the capillary barrier. Patients with severe sepsis may develop respiratory failure and hence requiring ventilatory support. These signs represent the clinical expression of the acute respiratory distress syndrome (ARDS) and are associated with high mortality in medical intensive care units (ICU). The proposed model is used to simulate the cardiovascular hemodynamics under these complex pathophysiological conditions.

Suggestive Therapeutic Pathways Using Hyper-Heuristics

Abstract Therapeutic decision support can be used to promptly assist clinical decision making process. This paper presents a new approach to interpreting multiple data streams in intensive care environments, the resulting model can be used to correct and maintain patients’ health whilst treating underlying illnesses. Rather than simply directing which treatments to be applied, multiple suggestive treatment pathways can be provided allowing several “what-if” scenarios to choose from. Hyper-heuristics are used to guide the treatments and therapeutic pathways selection. Algorithmic validation is made using a human cardiovascular system model parameterised with various post surgery conditions.

A Hierarchical Self-Organising Fuzzy Logic-Based On-Line Advisor for the Management of Cardiac Septic Patients

Abstract An on-line intelligent advisory system (IAS) for the management of septic patients emerging from cardiopulmonary bypass (CPB) has been developed and evaluated on a comprehensive physiological model which incorporates sepsis-induced capillary leak and the hemodynamic response to Intensive Care Unit (ICU)-like therapeutic interventions. The IAS is composed of the experts decision-making model which performs diagnosis and designates the appropriate therapeutic actions based on the patients hemodynamic status and a multi-inputs multi-outputs self-organising fuzzy logic controller (SOFLC) for adjusting the infusion rates of the selected drugs to maintain the hemodynamics variables at the specified target values. Simulated patient scenarios reproducing post-CPB hemodynamic abnormalities were developed with the expert anesthetist to test the IAS ability to advice for the appropriate therapeutic interventions whilst controlling individual drugs infusion rates to maintain the hemodynamic parameters at the prescribed targets values.

Online qualitative abstraction of cardiovascular hemodynamics for post cardiac surgery decision support

This paper addresses methods for online temporal abstraction of hemodynamic data for the purpose of being integrated in a computer-based decision support system in relation to post cardiac surgery patients. Temporal abstraction mechanisms provide a means of pre-processing and converting the monitored numerical data into a machine useable qualitative form. The temporal abstraction algorithm described in this paper is based on fuzzy clustering concepts. Synthetic signals which mimic the physiological signals of interest and real hemodynamic data recorded from clinical studies were used to evaluate these algorithms. Furthermore a cardiac output monitoring set-up was used to test the effectiveness of the algorithms under conditions similar to those existing in a real-time clinical environment.

Total Intravenous Anaesthesia with an Integrated Computer Control System

Abstract An integrated multivariable computer control system is currently under development to assist anaesthetists in the administration of anaesthetic, analgesic and muscle relaxant infusions for total intravenous anaesthesia This multivariable system is the integration of several single variable systems developed by the research team. A serimultiplexer, vas developed by the research team to allow communication of multiple infusion and monitoring devices via a single computer serial port. Initially the system was arranged with pharmacokinetic model-driven systems for the anaesthetic and analgesic agents and a Relaxograph (Datex) monitor was used for fuzzy closed-loop control of neuromuscular block (NMB).