Integrating data, models, and reasoning in critical care

Use of artificial intelligence in paediatric anaesthesia: a systematic review

Objectives

Although the development of artificial intelligence (AI) technologies in medicine has been significant, their application to paediatric anaesthesia is not well characterised. As the paediatric operating room is a data-rich environment that requires critical clinical decision-making, this systematic review aims to characterise the current use of AI in paediatric anaesthesia and to identify barriers to the successful integration of such technologies.

Methods

This review was registered with PROSPERO (CRD42022304610), the international registry for systematic reviews. The search strategy was prepared by a librarian and run in five electronic databases (Embase, Medline, Central, Scopus, and Web of Science). Collected articles were screened by two reviewers. Included studies described the use of AI for paediatric anaesthesia (<18 yr old) within the perioperative setting.

Results

From 3313 records identified in the initial search, 40 were included in this review. Identified applications of AI were described for patient risk factor prediction (24 studies; 60%), anaesthetic depth estimation (2; 5%), anaesthetic medication/technique decision guidance (2; 5%), intubation assistance (1; 2.5%), airway device selection (3; 7.5%), physiological variable monitoring (6; 15%), and operating room scheduling (2; 5%). Multiple domains of AI were discussed including machine learning, computer vision, fuzzy logic, and natural language processing.

Conclusion

There is an emerging literature regarding applications of AI for paediatric anaesthesia, and their clinical integration holds potential for ultimately improving patient outcomes. However, multiple barriers to their clinical integration remain including a lack of high-quality input data, lack of external validation/evaluation, and unclear generalisability to diverse settings.

Systematic review protocol

CRD42022304610 (PROSPERO).

Detecting Pathogen Exposure During the Non-symptomatic Incubation Period Using Physiological Data: Proof of Concept in Non-human Primates

Background and Objectives: Early warning of bacterial and viral infection, prior to the development of overt clinical symptoms, allows not only for improved patient care and outcomes but also enables faster implementation of public health measures (patient isolation and contact tracing). Our primary objectives in this effort are 3-fold. First, we seek to determine the upper limits of early warning detection through physiological measurements. Second, we investigate whether the detected physiological response is specific to the pathogen. Third, we explore the feasibility of extending early warning detection with wearable devices.

Research Methods: For the first objective, we developed a supervised random forest algorithm to detect pathogen exposure in the asymptomatic period prior to overt symptoms (fever). We used high-resolution physiological telemetry data (aortic blood pressure, intrathoracic pressure, electrocardiograms, and core temperature) from non-human primate animal models exposed to two viral pathogens: Ebola and Marburg (N = 20). Second, to determine reusability across different pathogens, we evaluated our algorithm against three independent physiological datasets from non-human primate models (N = 13) exposed to three different pathogens: Lassa and Nipah viruses and Y. pestis. For the third objective, we evaluated performance degradation when the algorithm was restricted to features derived from electrocardiogram (ECG) waveforms to emulate data from a non-invasive wearable device.

Results: First, our cross-validated random forest classifier provides a mean early warning of 51 ± 12 h, with an area under the receiver-operating characteristic curve (AUC) of 0.93 ± 0.01. Second, our algorithm achieved comparable performance when applied to datasets from different pathogen exposures – a mean early warning of 51 ± 14 h and AUC of 0.95 ± 0.01. Last, with a degraded feature set derived solely from ECG, we observed minimal degradation – a mean early warning of 46 ± 14 h and AUC of 0.91 ± 0.001.

Conclusion: Under controlled experimental conditions, physiological measurements can provide over 2 days of early warning with high AUC. Deviations in physiological signals following exposure to a pathogen are due to the underlying host’s immunological response and are not specific to the pathogen. Pre-symptomatic detection is strong even when features are limited to ECG-derivatives, suggesting that this approach may translate to non-invasive wearable devices.

Patient-adaptable intracranial pressure morphology analysis using a probabilistic model-based approach

OBJECTIVE

We present a framework for analyzing the morphology of intracranial pressure (ICP). The analysis of ICP signals is challenging due to the non-linear and non-Gaussian characteristics of the signal dynamics, inevitable corruption by noise and artifacts, and variations in ICP pulse morphology among individuals with different neurological conditions. Existing frameworks make unrealistic assumptions regarding ICP dynamics and are not tuned for individual patients.

APPROACH

We propose a dynamic Bayesian network for automated detection of three major ICP pulsatile components. The proposed model captures the non-linear and non-Gaussian dynamics of ICP morphology and further adapts to a patient as the individual's ICP measurements are received. To make the approach more robust, we leverage evidence reversal and present an inference algorithm to obtain the posterior distribution over the locations of pulsatile components.

MAIN RESULTS

We evaluate our approach on a dataset with over 700 h of recordings from 66 neurological patients, where the pulsatile components were annotated by prior studies. The algorithm obtains accuracies of 96.56%, 92.39%, and 94.04% for the detection of each pulsatile component in the test set, showing significant improvement over existing approaches.

SIGNIFICANCE

Continuous ICP monitoring is essential in guiding the treatment of neurological conditions such as traumatic brain injuries. An automated approach for ICP morphology analysis is a step towards enhancing patient care with minimal supervision. Compared to previous methods, our framework offers several advantages. It learns the parameters that model each patient's ICP in an unsupervised manner, resulting in an accurate morphology analysis. The Bayesian model-based framework provides uncertainty estimates and reveals interesting facts about the ICP dynamics. The framework can readily be applied to replace existing morphological analysis methods and support the use of ICP pulse morphological features to aid the monitoring of pathophysiological changes of relevance to the care of patients with acute brain injuries.

Augmented intelligence in pediatric anesthesia and pediatric critical care

PURPOSE OF REVIEW

Acute care technologies, including novel monitoring devices, big data, increased computing capabilities, machine-learning algorithms and automation, are converging. This enables the application of augmented intelligence for improved outcome predictions, clinical decision-making, and offers unprecedented opportunities to improve patient outcomes, reduce costs, and improve clinician workflow. This article briefly explores recent work in the areas of automation, artificial intelligence and outcome prediction models in pediatric anesthesia and pediatric critical care.

RECENT FINDINGS

Recent years have yielded little published research into pediatric physiological closed loop control (a type of automation) beyond studies focused on glycemic control for type 1 diabetes. However, there has been a greater range of research in augmented decision-making, leveraging artificial intelligence and machine-learning techniques, in particular, for pediatric ICU outcome prediction.

SUMMARY

Most studies focusing on artificial intelligence demonstrate good performance on prediction or classification, whether they use traditional statistical tools or novel machine-learning approaches. Yet the challenges of implementation, user acceptance, ethics and regulation cannot be underestimated. Areas in which there is easy access to routinely labeled data and robust outcomes, such as those collected through national networks and quality improvement programs, are likely to be at the forefront of the adoption of these advances.

Evaluation of standard versus nonstandard vital signs monitors in the prehospital and emergency departments: results and lessons learned from a trauma patient care protocol

BACKGROUND

This study aimed to determine the effectiveness of using a wireless, portable vital signs monitor (WVSM) for predicting the need for lifesaving interventions (LSIs) in the emergency department (ED) and use a multivariate logistic regression model to determine whether the WVSM was an improved predictor of LSIs in the ED over the standard of care monitor currently being used.

METHODS

This study analyzed 305 consecutive patients transported from the scene via helicopter to a Level I trauma center. For 104 patients in the study, a WVSM was also attached to the patient's arm and used to record and display prehospital and hospital physiologic data in real time on a handheld computer and in the trauma bay. Multivariate logistic regression analyses were performed for accuracy in predicting needs for LSIs in control and WVSM subjects. In addition, receiver operating characteristic curves were obtained to examine the discriminating power of the models for the outcome of one or more LSIs in the ED.

RESULTS

Of the 305 patients, 73 underwent 109 LSIs in the ED. Of these, 21 patients wore the WVSM during transport in addition to the standard monitor. Logistic regression analysis revealed that heart rate, respiratory rate, and systolic blood pressure were significantly associated with an increased risk for LSIs in the ED (p < 0.05). Receiver operating characteristic curve analysis also demonstrated better prediction for LSIs performed in the ED in WVSM subjects than in control subjects (area under the curve, 0.86 vs. 0.81, respectively).

CONCLUSION

The WVSM system leads to improved LSI accuracy in the ED. In addition, many important lessons have been learned in preparation for this study. Adoption of nonstandard vital signs monitors into critical care/trauma medicine may require a new paradigm of personnel education, training, and practice.

LEVEL OF EVIDENCE

Therapeutic/care management, level IV.

Integrating monitor alarms with laboratory test results to enhance patient deterioration prediction

Patient monitors in modern hospitals have become ubiquitous but they generate an excessive number of false alarms causing alarm fatigue. Our previous work showed that combinations of frequently co-occurring monitor alarms, called SuperAlarm patterns, were capable of predicting in-hospital code blue events at a lower alarm frequency. In the present study, we extend the conceptual domain of a SuperAlarm to incorporate laboratory test results along with monitor alarms so as to build an integrated data set to mine SuperAlarm patterns. We propose two approaches to integrate monitor alarms with laboratory test results and use a maximal frequent itemsets mining algorithm to find SuperAlarm patterns. Under an acceptable false positive rate FPRmax, optimal parameters including the minimum support threshold and the length of time window for the algorithm to find the combinations of monitor alarms and laboratory test results are determined based on a 10-fold cross-validation set. SuperAlarm candidates are generated under these optimal parameters. The final SuperAlarm patterns are obtained by further removing the candidates with false positive rate>FPRmax. The performance of SuperAlarm patterns are assessed using an independent test data set. First, we calculate the sensitivity with respect to prediction window and the sensitivity with respect to lead time. Second, we calculate the false SuperAlarm ratio (ratio of the hourly number of SuperAlarm triggers for control patients to that of the monitor alarms, or that of regular monitor alarms plus laboratory test results if the SuperAlarm patterns contain laboratory test results) and the work-up to detection ratio, WDR (ratio of the number of patients triggering any SuperAlarm patterns to that of code blue patients triggering any SuperAlarm patterns). The experiment results demonstrate that when varying FPRmax between 0.02 and 0.15, the SuperAlarm patterns composed of monitor alarms along with the last two laboratory test results are triggered at least once for [56.7-93.3%] of code blue patients within an 1-h prediction window before code blue events and for [43.3-90.0%] of code blue patients at least 1-h ahead of code blue events. However, the hourly number of these SuperAlarm patterns occurring in control patients is only [2.0-14.8%] of that of regular monitor alarms with WDR varying between 2.1 and 6.5 in a 12-h window. For a given FPRmax threshold, the SuperAlarm set generated from the integrated data set has higher sensitivity and lower WDR than the SuperAlarm set generated from the regular monitor alarm data set. In addition, the McNemar's test also shows that the performance of the SuperAlarm set from the integrated data set is significantly different from that of the SuperAlarm set from the regular monitor alarm data set. We therefore conclude that the SuperAlarm patterns generated from the integrated data set are better at predicting code blue events.

Space physiology IV: mathematical modeling of the cardiovascular system in space exploration

Mathematical modeling represents an important tool for analyzing cardiovascular function during spaceflight. This review describes how modeling of the cardiovascular system can contribute to space life science research and illustrates this process via modeling efforts to study postflight orthostatic intolerance (POI), a key issue for spaceflight. Examining this application also provides a context for considering broader applications of modeling techniques to the challenges of bioastronautics. POI, which affects a large fraction of astronauts in stand tests upon return to Earth, presents as dizziness, fainting and other symptoms, which can diminish crew performance and cause safety hazards. POI on the Moon or Mars could be more critical. In the field of bioastronautics, POI has been the dominant application of cardiovascular modeling for more than a decade, and a number of mechanisms for POI have been investigated. Modeling approaches include computational models with a range of incorporated factors and hemodynamic sophistication, and also physical models tested in parabolic and orbital flight. Mathematical methods such as parameter sensitivity analysis can help identify key system mechanisms. In the case of POI, this could lead to more effective countermeasures. Validation is a persistent issue in modeling efforts, and key considerations and needs for experimental data to synergistically improve understanding of cardiovascular responses are outlined. Future directions in cardiovascular modeling include subject-specific assessment of system status, as well as research on integrated physiological responses, leading, for instance, to assessment of subject-specific susceptibility to POI or effects of cardiovascular alterations on muscular, vision and cognitive function.

A qualitative study of expert and team cognition on complex patients in the pediatric intensive care unit

OBJECTIVES

To understand expert and team cognition of complex patients in the pediatric intensive care unit through the use of cognitive task analysis.

DESIGN

Qualitative study with semistructured interviews.

SETTING

Academic medical center pediatric intensive care unit.

PARTICIPANTS

Physicians, nurses, and nurse practitioners.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Semistructured interviews were conducted with members of the critical care team involved with the care of seven complex patients. Interviews were transcribed and themes were identified based on grounded theory and further divided into categories. A focus group of critical care team members further refined and validated the findings. From the interviews, 177 verbal fragments were sorted into 11 themes. Four broad thematic categories were identified and a cognitive framework for the care of complex patients was formulated. We found that at the center of this framework, critical care teams attempt to create and share mental models of their patients. These mental models serve as the framework for delivery of longitudinal care across handovers and shift changes. The analysis revealed that this process is limited by a number of factors such that team members utilize a variety of techniques to overcome these limitations and develop more complete and shared mental models.

CONCLUSIONS

An inadequately developed or inadequately shared mental model is a substantial cognitive limitation for expert and team cognition in the complex environment of the pediatric intensive care unit. Providers utilize techniques that may avoid or decrease the variable interpretations of patient condition that would otherwise impair mental model formation and sharing. Future studies should be designed to enhance mental model formation and communication in the pediatric intensive care unit and other environments that deal with complex patients.

Neonatal Informatics: Optimizing Clinical Data Entry and Display

Displaying the vast amount of clinical data that exist in electronic medical records without causing information overload or interfering with provider thought processes is a challenge. To support the transformation of data into information and knowledge, effective electronic displays must be flexible and guide physicians' thought processes. Applying research from cognitive science and human factors engineering offers promise in improving the electronic display of clinical information. OBJECTIVES: After completing this article, readers should be able to: Appreciate the importance of supporting provider thought processes during both data entry and data review.Recognize that information does not need to be displayed and reviewed in the same way the data are entered.

Model-based data integration in clinical environments

As a result of improved hospital information-technology infrastructure and declining costs of storage media, vast amounts of physiological waveform and trend data can now be continuously collected and archived from bedside monitors in operating rooms, intensive care units, or even regular hospital rooms. The real-time or off-line processing of such volumes of high-resolution data, in attempts to turn raw data into clinically actionable information, poses significant challenges. However, it also presents researchers - and eventually clinicians - with unprecedented opportunities to move beyond the traditional individual-channel analysis of waveform data, and towards an integrative patient-monitoring framework, with likely improvements in patient care and safety. We outline some of the challenges and opportunities, and propose strategies for model-based integration of physiological data to improve patient monitoring.

Reconstruction of ECG signals in presence of corruption

We present an approach to identifying and reconstructing corrupted regions in a multi-parameter physiological signal. The method, which uses information in correlated signals, is specifically designed to preserve clinically significant aspects of the signals. We use template matching to jointly segment the multi-parameter signal, morphological dissimilarity to estimate the quality of the signal segment, similarity search using features on a database of templates to find the closest match, and time-warping to reconstruct the corrupted segment with the matching template. In experiments carried out on the MIT-BIH Arrhythmia Database, a two-parameter database with many clinically significant arrhythmias, our method improved the classification accuracy of the beat type by more than 7 times on a signal corrupted with white Gaussian noise, and increased the similarity to the original signal, as measured by the normalized residual distance, by more than 2.5 times.