Search PubMed⌕ Search

Biomedical subjects

D F Sittig

Publications and source records attributed to D F Sittig.

At least 19 recordsLinked to original sources
PubMed1992

Fuzzy classification of hemodynamic trends and artifacts: experiments with the heart rate.

Fuzzy set theory allows one to map inexact data, concepts, and events to fuzzy sets via user-defined membership functions. This paper describes a method for (1) robustly estimating the mean and slope of an arbitrary number of data points, (2) developing a set of fuzzy membership functions to classify various properties of heart rate trends, and (3) finding the longest consecutive sequence of heart rate data that fit a particular fuzzy membership function. Preliminary results indicate that fuzzy set theory has significant potential in the development of a clinically robust method for classifying heart rate data, trends, and artifacts.

Data Interpretation, Statistical↗
PubMed1992

A parallel implementation of a multi-state Kalman filtering algorithm to detect ECG arrhythmias.

Detecting arrhythmias from the electrocardiogram (ECG) is of great importance for the continued development of intelligent cardiovascular monitors (ICM). An ICM's main goal is to present to the clinician a 'high-level' analysis of the patient's condition (e.g., the patient is slightly hypovolemic) based upon 'low-level' physiologic signals (e.g., blood pressure, heart rate, etc.). This paper reports on a parallel implementation of a multi-state Kalman filtering algorithm, within a prototype ICM, to help detect ECG arrhythmias. Preliminary test results show that the parallel, multi-state implementation performed exactly as the original sequential version. Several different rhythm disturbances were correctly identified after 3-5 beats. We conclude that our parallel implementation of the multi-state Kalman filter provides a faster and still reliable means of accurately detecting ECG arrhythmias in real-time.

Algorithms↗
PubMed1992

A parallel implementation of the backward error propagation neural network training algorithm: experiments in event identification.

An artificial neural-network-based (ANN) event detection and alarm generation system has been developed to aid clinicians in the identification of critical events commonly occurring in the anesthesia breathing circuit. To detect breathing circuit problems, the system monitored CO2 gas concentration, gas flow, and airway pressure. Various parameters were extracted from each of these input waveforms and fed into an artificial neural network. To develop truly robust ANNs, investigators are required to train their networks on large training data sets, requiring enormous computing power. We implemented a parallel version of the backward error propagation neural network training algorithm in the widely portable parallel programming language C-Linda. A maximum speedup of 4.06 was obtained with six processors. This speedup represents a reduction in total run-time from 6.4 to 1.5 h. By reducing the total run time of the computation through parallelism, we were able to optimize many of the neural network's initial parameters. We conclude that use of the master-worker model of parallel computation is an excellent method for speeding up the backward error propagation neural network training algorithm.

Algorithms↗
PubMed1992

BIO-SPEAD: a parallel computing environment to accelerate development of biologic signal processing algorithms.

We have created BIO-SPEAD (pronounced speed), a BIOlogical Signal Processing Environment for Algorithm Development. BIO-SPEAD is designed to accelerate development of complex algorithms which integrate information derived from single or multiple physiologic waveforms. BIO-SPEAD currently performs all of the basic analyses of several arterial blood pressure waveforms, and allows the user to utilize the results of those low-level analyses for development of more complex algorithms. We utilized a parallel programming architecture called the Process Trellis which keeps the different tasks, or processes, within BIO-SPEAD independent of each other. Additionally, we have developed a graphics interface to enable the user to visualize the waveform under analysis, the low-level system analysis, and the internal workings of the algorithm under development. The system has been used for several algorithm development projects and has demonstrated its utility.

Algorithms↗
PubMed1992

A fast track to IAIMS: the Vanderbilt University strategy.

In July 1991, Vanderbilt University Medical Center (VUMC) initiated a fast track approach to the implementation of an Integrated Academic Information Management System (IAIMS). The fast track approach has four elements: 1) an integrated organizational structure combining various operational information management units and the academic informatics program into a single entity to enhance efficiency; 2) technology transfer and network access to remote resources in preference to de novo development; 3) parallel IAIMS planning and infrastructure construction; 4) restriction of the scope of the initial IAIMS to permit a manageable implementation project. The fast track approach is intended to provide a truly functional IAIMS within a time period (7 years) associated with other major construction projects such as the building of a replacement hospital.

Integrated Advanced Information Management Systems↗
PubMed1992

Work-sampling: a statistical approach to evaluation of the effect of computers on work patterns in the healthcare industry.

An increasing number of healthcare institutions are in the process of implementing clinical computing systems. The need for an accurate assessment of the clinical, administrative, social, and financial effects of such systems has been recognized. Techniques have been developed to evaluate these effects on the work patterns of healthcare workers including: 1) time-motion analysis, 2) subjective evaluations, 3) review of departmental statistics, 4) personal activity records, and 5) work-sampling. This manuscript reviews these techniques, discusses both positive and negative aspects, and presents a step-by-step description of work-sampling.

Evaluation Studies as Topic↗
PubMed1992

The multi-trellis software architecture and the intelligent cardiovascular monitor.

A real-time, intelligent cardiovascular monitor is complex. It must process multiple waveforms, recognize artifacts, extract pertinent parameters, recognize a patient's clinical state, analyze the problem and formulate a response. This paper presents the multi-trellis (a collection of process trellises), a software architecture for building such a monitor. A process trellis is a uniform hierarchical framework for heterogeneous program modules. The multi-trellis extension allows one to compile several process trellis programs with widely varying run-time requirements into a single executable program that it is efficient, predictable and usable. Our prototype consists of two process trellises. The lower trellis contains processes to analyze three different analog signals: the blood pressure from a non-invasive monitor and an arterial catheter, and the ECG. The upper trellis contains processes to help detect evolving hemodynamic trends, identify abnormalities, and present a succinct summary to the clinician. Our prototype shows that the multi-trellis is a demonstrably useful software architecture for building these real-time, intelligent monitors.

Artificial Intelligence↗
PubMed1991

A parallel computing approach to genetic sequence comparison: the master-worker paradigm with interworker communication.

We have implemented a parallel version of a dynamic programming biological sequence comparison algorithm to study the potential applicability of using parallel computers for genetic sequence comparisons. Our parallel program is built using C-Linda, a machine-independent parallel programming language, and was tested on both a 10 CPU Sequent Symmetry and a 64 CPU Intel Hypercube. C-Linda implements a shared associative memory model, "tuple space," through which multiple processes can communicate and coordinate control. In our master-worker (MW) parallel implementation, a master process creates several worker processes, extracts a test sequence and multiple library sequences from a database and stores them in tuple space. Each worker reads the test sequence and then repeatedly extracts library strings from tuple space, performs pairwise sequence comparison using a local comparison algorithm to generate a similarity score, and returns the similarity scores to tuple space. The master collects the scores from tuple space and identifies the best match over all library sequences. We also implemented a method of global interworker communication to reduce the total search time by stopping those string comparisons that had no chance of improving on the current best match. Comparisons of the total run time, speedup, and efficiency were made for parallel and sequential versions of a basic MW implementation as well as versions with the global abort threshold.

Algorithms↗
PubMed1991

Evaluation of a parallel implementation of the learning portion of the backward error propagation neural network: experiments in artifact identification.

Various methods have been proposed in an attempt to solve problems in artifact and/or alarm identification including expert systems, statistical signal processing techniques, and artificial neural networks (ANN). ANNs consist of a large number of simple processing units connected by weighted links. To develop truly robust ANNs, investigators are required to train their networks on huge training data sets, requiring enormous computing power. We implemented a parallel version of the backward error propagation neural network training algorithm in the widely portable parallel programming language C-Linda. A maximum speedup of 4.06 was obtained with six processors. This speedup represents a reduction in total run-time from approximately 6.4 hours to 1.5 hours. We conclude that use of the master-worker model of parallel computation is an excellent method for obtaining speedups in the backward error propagation neural network training algorithm.

Algorithms↗
PubMed1991

A parallel computing interface to facilitate development of physiologic signal processing algorithms.

We have created a system to aid in the development of algorithms related to the blood pressure waveform. The system performs all of the basic analyses of the waveform, and allows the user to utilize the results of those analyses for the algorithm under development. We have used a parallel programming architecture which keeps the different tasks, or processes, within the system independent of each other. Additionally, we have developed a graphics interface to enable the user to visualize the waveform, the system analysis, and the internal workings of the algorithm under development.

Algorithms↗
PubMed1990

A parallel software architecture for building intelligent medical monitors.

Intensive care units become more complicated each day as the number of devices developed to monitor various aspects of a patient's status continues to increase. Intelligent monitors attempt to reduce this complexity by interpreting the data and presenting a high level summary to the clinician. We propose an innovative parallel software architecture for constructing intelligent medical monitors: the process trellis. The process trellis is an explicitly parallel structure, and therefore can take advantage of the performance gains available from parallel computing hardware. It does not, however, presuppose any expertise in parallel programming on the part of the application programmer. A prototype cardiovascular monitor has been built using this parallel software architecture. Preliminary testing of the monitor has shown that real-time cardiovascular monitoring, including data calculations, symbolic classification, and interpretation can be accomplished in real-time.

Expert Systems↗
PubMed1990

Clinical evaluation of computer-based respiratory care algorithms.

A collection of computer-based respiratory care algorithms were implemented as a prototype computer-based patient advice system (COMPAS) within the existing HELP hospital information system. Detailed medical logic recommended ventilator adjustments for 5 different modes of ventilation: assist/control (A/C), intermittent mandatory ventilation (IMV), continuous positive airway pressure (CPAP), pressure controlled inverted ratio ventilation (PC-IRV), and extracorporeal carbon dioxide removal (ECCO2R). Suggestions for adjusting the mode of ventilation, fraction of inspired oxygen (FiO2), positive end-expiratory pressure (PEEP), peak inspiratory pressure, and several other therapeutic measures related to the treatment of severe arterial hypoxemia in adult respiratory distress syndrome (ARDS) patients were automatically presented to the clinical staff via bedside computer terminals. COMPAS was clinically evaluated for 624 hours of patient care on the first 5 ARDS patients in a randomized clinical trial. The clinical staff carried out 84% (320/379) of the computerized therapy suggestions. In response to a questionnaire distributed to clinical users of the system, 86% judged the system to be potentially valuable. Through implementation of COMPAS, a computer-based ventilatory therapy advice system, we have laid the groundwork for standardization of ventilator management of arterial hypoxemia in critically ill ARDS patients.

Algorithms↗
PubMed1990

Physiologic trend detection and artifact rejection: a parallel implementation of a multi-state Kalman filtering algorithm.

Using a parallel implementation of the multi-state Kalman filtering algorithm, we have developed an accurate method of reliably detecting and identifying trends, abrupt changes, and artifacts from multiple physiologic data streams in real-time. The Kalman filter algorithm was implemented within an innovative software architecture for parallel computation: a parallel process trellis. Examples, processed in real-time, of both simulated and actual data serve to illustrate the potential value of the Kalman filter as a tool in physiologic monitoring.

Algorithms↗
PubMed1989

Implementation of a computerized patient advice system using the HELP clinical information system.

A COMputerized Patient Advice System (COMPAS) was designed to test the feasibility of using the HELP clinical information system to direct the respiratory therapy of intensive care (ICU) patients acutely ill with adult respiratory distress syndrome. A modified black-board control architecture allowed the application of knowledge in either a forward or a backward chaining mode. Expert clinicians recommended decision logic and actions for five different modes of ventilatory support. The clinical staff used COMPAS to manage the ICU ventilatory support of five patients for a total of 624 hr. During that time there were 407 decision-making opportunities. COMPAS automatically generated therapy suggestions 379 (93.1%) times and the clinical staff accepted COMPAS's recommendation in 320 (84.4%) of these cases. These results suggest that the ventilatory support of severely ill ICU patients can be managed by a clinical information system using a blackboard control architecture.

Algorithms↗
PubMedDate not supplied

A strategy for development of computerized critical care decision support systems.

It is not enough to merely manage medical information. It is difficult to justify the cost of hospital information systems (HIS) or intensive care unit (ICU) patient data management systems (PDMS) on this basis alone. The real benefit of an integrated HIS or PDMS is in decision support. Although there are a variety of HIS and ICU PDMS systems available there are few that provide ICU decision support. The HELP system at the LDS Hospital is an example of a HIS which provides decision support on many different levels. In the ICU there are decision support tools for antibiotic therapy, nutritional management, and management of mechanical ventilation. Computer protocols for the management of mechanical ventilation (respiratory evaluation, ventilation, oxygenation, weaning and extubation) in patients with adult respiratory distress syndrome ((ARDS) have already been developed and clinically validated at the LDS Hospital. These protocols utilize the bedside intensive care unit (ICU) computer terminal to prompt the clinical care team with therapeutic and diagnostic suggestions. The protocols (in paper flow diagram and computerized form) have been used for over 40,000 hours in more than 125 adult respiratory distress syndrome (ARDS) patients. The protocols controlled care for 94% of the time. The remainder of the time patient care was not protocol controlled was a result of the patient being in states not covered by current protocol logic (e.g. hemodynamic instability, or transport for X-Ray studies). 52 of these ARDS patients met extra corporal membrane oxygenation (ECMO) criteria. The survival of the ECMO criteria ARDS patients was 41%, four times that expected (9%) from historical data (p less than 0.0002).(ABSTRACT TRUNCATED AT 250 WORDS)

Attitude of Health Personnel↗
PubMedDate not supplied

Computerized management of patient care in a complex, controlled clinical trial in the intensive care unit.

Acute respiratory distress syndrome (ARDS) is often not responsive to conventional supportive therapy and the mortality rate may exceed 90%. A new form of supportive care, extracorporeal carbon dioxide removal (ECCO2R), has shown a dramatic increase in survival (48%). A controlled clinical trial of the new ECCO2R therapy versus conventional continuous positive pressure ventilation (CPPV) is being initiated. Detailed care protocols have been developed by 'expert' critical care physicians for the management of patients. Using a blackboard control architecture, the protocols have been implemented on an existing hospital information system and will direct patient care and help manage the controlled clinical trial. Therapeutic instructions are automatically generated by the computer from data input by physicians, nurses, respiratory therapists, and the laboratory. Preliminary results show that the computerized protocol system can direct therapy for acutely ill patients.

Blood Gas Analysis↗
PubMedDate not supplied

The evaluation of clinical decision support systems: what is necessary versus what is interesting.

The evaluation of clinical decision support systems has long been recognized as an important part of the overall field of medical computing, as well as a complex and varied field in its own right. There are a number of purposes for which such an evaluation might be performed. This paper focuses on one of those purposes, the assurance of safety when preparing to release the system for general use. The assurance of safety involves two major stages: (1) a comprehensive validation of the system's knowledge and advice, and (2) a period of use which exposes the system's advice and interface to the clinical environment.

Decision Making, Computer-Assisted↗