Real alerts and artifact classification in archived multi-signal vital sign monitoring data: implications for mining big data

A Comparison of Five Algorithmic Methods and Machine Learning Pattern Recognition for Artifact Detection in Electronic Records of Five Different Vital Signs: A Retrospective Analysis

BACKGROUND

Research on electronic health record physiologic data is common, invariably including artifacts. Traditionally, these artifacts have been handled using simple filter techniques. The authors hypothesized that different artifact detection algorithms, including machine learning, may be necessary to provide optimal performance for various vital signs and clinical contexts.

METHODS

In a retrospective single-center study, intraoperative operating room and intensive care unit (ICU) electronic health record datasets including heart rate, oxygen saturation, blood pressure, temperature, and capnometry were included. All records were screened for artifacts by at least two human experts. Classical artifact detection methods (cutoff, multiples of SD [z-value], interquartile range, and local outlier factor) and a supervised learning model implementing long short-term memory neural networks were tested for each vital sign against the human expert reference dataset. For each artifact detection algorithm, sensitivity and specificity were calculated.

RESULTS

A total of 106 (53 operating room and 53 ICU) patients were randomly selected, resulting in 392,808 data points. Human experts annotated 5,167 (1.3%) data points as artifacts. The artifact detection algorithms demonstrated large variations in performance. The specificity was above 90% for all detection methods and all vital signs. The neural network showed significantly higher sensitivities than the classic methods for heart rate (ICU, 33.6%; 95% CI, 33.1 to 44.6), systolic invasive blood pressure (in both the operating room [62.2%; 95% CI, 57.5 to 71.9] and the ICU [60.7%; 95% CI, 57.3 to 71.8]), and temperature in the operating room (76.1%; 95% CI, 63.6 to 89.7). The CI for specificity overlapped for all methods. Generally, sensitivity was low, with only the z-value for oxygen saturation in the operating room reaching 88.9%. All other sensitivities were less than 80%.

CONCLUSIONS

No single artifact detection method consistently performed well across different vital signs and clinical settings. Neural networks may be a promising artifact detection method for specific vital signs.

EDITOR’S PERSPECTIVE

PRECISION-TBI: a study protocol for a vanguard prospective cohort study to enhance understanding and management of moderate to severe traumatic brain injury in Australia

INTRODUCTION

Traumatic brain injury (TBI) is a heterogeneous condition in terms of pathophysiology and clinical course. Outcomes from moderate to severe TBI (msTBI) remain poor despite concerted research efforts. The heterogeneity of clinical management represents a barrier to progress in this area. PRECISION-TBI is a prospective, observational, cohort study that will establish a clinical research network across major neurotrauma centres in Australia. This network will enable the ongoing collection of injury and clinical management data from patients with msTBI, to quantify variations in processes of care between sites. It will also pilot high-frequency data collection and analysis techniques, novel clinical interventions, and comparative effectiveness methodology.

METHODS AND ANALYSIS

PRECISION-TBI will initially enrol 300 patients with msTBI with Glasgow Coma Scale (GCS) <13 requiring intensive care unit (ICU) admission for invasive neuromonitoring from 10 Australian neurotrauma centres. Demographic data and process of care data (eg, prehospital, emergency and surgical intervention variables) will be collected. Clinical data will include prehospital and emergency department vital signs, and ICU physiological variables in the form of high frequency neuromonitoring data. ICU treatment data will also be collected for specific aspects of msTBI care. Six-month extended Glasgow Outcome Scores (GOSE) will be collected as the key outcome. Statistical analysis will focus on measures of between and within-site variation. Reports documenting performance on selected key quality indicators will be provided to participating sites.

ETHICS AND DISSEMINATION

Ethics approval has been obtained from The Alfred Human Research Ethics Committee (Alfred Health, Melbourne, Australia). All eligible participants will be included in the study under a waiver of consent (hospital data collection) and opt-out (6 months follow-up). Brochures explaining the rationale of the study will be provided to all participants and/or an appropriate medical treatment decision-maker, who can act on the patient's behalf if they lack capacity. Study findings will be disseminated by peer-review publications.

TRIAL REGISTRATION NUMBER

NCT05855252.

Weakly Supervised Classification of Vital Sign Alerts as Real or Artifact

A significant proportion of clinical physiologic monitoring alarms are false. This often leads to alarm fatigue in clinical personnel, inevitably compromising patient safety. To combat this issue, researchers have attempted to build Machine Learning (ML) models capable of accurately adjudicating Vital Sign (VS) alerts raised at the bedside of hemodynamically monitored patients as real or artifact. Previous studies have utilized supervised ML techniques that require substantial amounts of hand-labeled data. However, manually harvesting such data can be costly, time-consuming, and mundane, and is a key factor limiting the widespread adoption of ML in healthcare (HC). Instead, we explore the use of multiple, individually imperfect heuristics to automatically assign probabilistic labels to unlabeled training data using weak supervision. Our weakly supervised models perform competitively with traditional supervised techniques and require less involvement from domain experts, demonstrating their use as efficient and practical alternatives to supervised learning in HC applications of ML.

Patient Deterioration in the Adult Progressive Care Unit: A Scoping Review

BACKGROUND

The American Heart Association set a goal in 2010 to double the percentage of hospitalized adult patients who survive cardiac arrest by 2020. Because of acuity and interventions, progressive care patients are a population of interest to address this goal. The state of the literature involving patient deterioration, which can lead to cardiac arrest, in the progressive care setting has yet to be explored.

OBJECTIVE

A scoping review was done to investigate the literature involving patient deterioration in adult progressive care units in order to map knowledge, identify themes, and discover areas for research potential.

METHODS

The scoping review began with an extensive literature search and a multistep review. The characteristics of the final group of studies were charted and grouped according to common themes.

RESULTS

There were 13 studies in the final group. All studies were conducted in the United States and most by interprofessional teams. Three themes were evident in the review, training methods, surveillance, and monitoring systems.

DISCUSSION

Patient deterioration in the progressive care unit may benefit from team-based training methods involving checklists or protocols. Nurses can use surveillance, including physical assessment and technology, to recognize early warning signs. Lastly, the use of systems that identify patterns in vital signs can be useful to reduce patient harm. Further research in this area care is warranted and could potentially improve patient outcomes and nursing practice.

Factor analysis for the clustering of cardiometabolic risk factors and sedentary behavior, a cross-sectional study

Background

Few studies have reported on the clustering pattern of CVD risk factors, including sedentary behavior, systemic inflammation, and cadiometabolic components in the general population.

Objective

We aimed to explore the clustering pattern of CVD risk factors using exploratory factor analysis to investigate the underlying relationships between various CVD risk factors.

Methods

A total of 5606 subjects (3157 male, 51.5±11.7 y/o) were enrolled, and 14 cardiovascular risk factors were analyzed in an exploratory group (n = 3926) and a validation group (n = 1676), including sedentary behaviors.

Results

Five factor clusters were identified to explain 69.4% of the total variance, including adiposity (BMI, TG, HDL, UA, and HsCRP; 21.3%), lipids (total cholesterol and LDL-cholesterol; 14.0%), blood pressure (SBP and DBP; 13.3%), glucose (HbA1C, fasting glucose; 12.9%), and sedentary behavior (MET and sitting time; 8.0%). The inflammation biomarker HsCRP was clustered with only adiposity factors and not with other cardiometabolic risk factors, and the clustering pattern was verified in the validation group.

Conclusion

This study confirmed the clustering structure of cardiometabolic risk factors in the general population, including sedentary behavior. HsCRP was clustered with adiposity factors, while physical inactivity and sedentary behavior were clustered with each other.

Automatic detection of artifacts in EEG by combining deep learning and histogram contour processing

This paper introduces a simple approach combining deep learning and histogram contour processing for automatic detection of various types of artifact contaminating the raw electroencephalogram (EEG). The proposed method considers both spatial and temporal information of raw EEG, without additional need for reference signals like ECG or EOG. The proposed method was evaluated with data including 785 EEG sequences contaminated by artifacts and 785 artifact-free EEG sequences collected from 15 intensive care patients. The obtained results showed an overall accuracy of 0.98, representing high reliability of proposed technique in detecting different types of artifacts and being comparable or outperforming the approaches proposed earlier in the literature.

The Byzantine Role of Big Data Application in Nursing Science: A Systematic Review

Big data have the potential to determine enhanced decision-making process and to personalize the approach of delivering care when applied in nursing science. So far, the literature on this topic is still not synthesized for the period between 2014 and 2018. Thus, this systematic review aimed to identify and synthesize the most recent evidence on big data application in nursing research. The systematic search was undertaken for the evidence published from January 2014 to May 2018, and the outputs were formatted using the PRISMA Flow Diagram, whereas the quality appraisal was addressed by recommendations consistent with the Critical Appraisal Skills Program. Twelve studies on big data in nursing were included and divided into two themes: the majority of the studies aimed to determine prediction assessment, while only four studies were related to the impact of big data applications to support clinical practice. This review tracks the recent state of knowledge on big data applications in nursing science, revealing the potential for nursing engagement in big data science, even if currently limited to some fields. Big data applications in nursing might have a tremendous potential impact, but are currently underused in research and clinical practice.

Artifact Processing Methods Influence on Intraoperative Hypotension Quantification and Outcome Effect Estimates

BACKGROUND

Physiologic data that is automatically collected during anesthesia is widely used for medical record keeping and clinical research. These data contain artifacts, which are not relevant in clinical care, but may influence research results. The aim of this study was to explore the effect of different methods of filtering and processing artifacts in anesthesiology data on study findings in order to demonstrate the importance of proper artifact filtering.

METHODS

The authors performed a systematic literature search to identify artifact filtering methods. Subsequently, these methods were applied to the data of anesthesia procedures with invasive blood pressure monitoring. Different hypotension measures were calculated (i.e., presence, duration, maximum deviation below threshold, and area under threshold) across different definitions (i.e., thresholds for mean arterial pressure of 50, 60, 65, 70 mmHg). These were then used to estimate the association with postoperative myocardial injury.

RESULTS

After screening 3,585 papers, the authors included 38 papers that reported artifact filtering methods. The authors applied eight of these methods to the data of 2,988 anesthesia procedures. The occurrence of hypotension (defined with a threshold of 50 mmHg) varied from 24% with a median filter of seven measurements to 55% without an artifact filtering method, and between 76 and 90% with a threshold of 65 mmHg. Standardized odds ratios for presence of hypotension ranged from 1.16 (95% CI, 1.07 to 1.26) to 1.24 (1.14 to 1.34) when hypotension was defined with a threshold of 50 mmHg. Similar variations in standardized odds ratios were found when applying methods to other hypotension measures and definitions.

CONCLUSIONS

The method of artifact filtering can have substantial effects on estimates of hypotension prevalence. The effect on the association between intraoperative hypotension and postoperative myocardial injury was relatively small. Nevertheless, the authors recommend that researchers carefully consider artifacts handling and report the methodology used.

Increasing Cardiovascular Data Sampling Frequency and Referencing It to Baseline Improve Hemorrhage Detection

Supplemental Digital Content is available in the text.

We hypothesize that knowledge of a stable personalized baseline state and increased data sampling frequency would markedly improve the ability to detect progressive hypovolemia during hemorrhage earlier and with a lower false positive rate than when using less granular data.

Unexpected sawtooth artifact in beat-to-beat pulse transit time measured from patient monitor data

Object

It is increasingly popular to collect as much data as possible in the hospital setting from clinical monitors for research purposes. However, in this setup the data calibration issue is often not discussed and, rather, implicitly assumed, while the clinical monitors might not be designed for the data analysis purpose. We hypothesize that this calibration issue for a secondary analysis may become an important source of artifacts in patient monitor data. We test an off-the-shelf integrated photoplethysmography (PPG) and electrocardiogram (ECG) monitoring device for its ability to yield a reliable pulse transit time (PTT) signal.

Approach

This is a retrospective clinical study using two databases: one containing 35 subjects who underwent laparoscopic cholecystectomy, another containing 22 subjects who underwent spontaneous breathing test in the intensive care unit. All data sets include recordings of PPG and ECG using a commonly deployed patient monitor. We calculated the PTT signal offline.

Main results

We report a novel constant oscillatory pattern in the PTT signal and identify this pattern as a sawtooth artifact. We apply an approach based on the de-shape method to visualize, quantify and validate this sawtooth artifact.

Significance

The PPG and ECG signals not designed for the PTT evaluation may contain unwanted artifacts. The PTT signal should be calibrated before analysis to avoid erroneous interpretation of its physiological meaning.

Applying machine learning to continuously monitored physiological data

The use of machine learning (ML) in healthcare has enormous potential for improving disease detection, clinical decision support, and workflow efficiencies. In this commentary, we review published and potential applications for the use of ML for monitoring within the hospital environment. We present use cases as well as several questions regarding the application of ML to the analysis of the vast amount of complex data that clinicians must interpret in the realm of continuous physiological monitoring. ML, especially employed in bidirectional conjunction with electronic health record data, has the potential to extract much more useful information out of this currently under-analyzed data source from a population level. As a data driven entity, ML is dependent on copious, high quality input data so that error can be introduced by low quality data sources. At present, while ML is being studied in hybrid formulations along with static expert systems for monitoring applications, it is not yet actively incorporated in the formal artificial learning sense of an algorithm constantly learning and updating its rules without external intervention. Finally, innovations in monitoring, including those supported by ML, will pose regulatory and medico-legal challenges, as well as questions regarding precisely how to incorporate these features into clinical care and medical education. Rigorous evaluation of ML techniques compared to traditional methods or other AI methods will be required to validate the algorithms developed with consideration of database limitations and potential learning errors. Demonstration of value on processes and outcomes will be necessary to support the use of ML as a feature in monitoring system development: Future research is needed to evaluate all AI based programs before clinical implementation in non-research settings.

A call to alarms: Current state and future directions in the battle against alarm fatigue

Research demonstrates that the majority of alarms derived from continuous bedside monitoring devices are non-actionable. This avalanche of unreliable alerts causes clinicians to experience sensory overload when attempting to sort real from false alarms, causing desensitization and alarm fatigue, which in turn leads to adverse events when true instability is neither recognized nor attended to despite the alarm. The scope of the problem of alarm fatigue is broad, and its contributing mechanisms are numerous. Current and future approaches to defining and reacting to actionable and non-actionable alarms are being developed and investigated, but challenges in impacting alarm modalities, sensitivity and specificity, and clinical activity in order to reduce alarm fatigue and adverse events remain. A multi-faceted approach involving clinicians, computer scientists, industry, and regulatory agencies is needed to battle alarm fatigue.

Dynamic and Personalized Risk Forecast in Step-Down Units. Implications for Monitoring Paradigms

RATIONALE

Cardiorespiratory insufficiency (CRI) is a term applied to the manifestations of loss of normal cardiorespiratory reserve and portends a bad outcome. CRI occurs commonly in hospitalized patients, but its risk escalation patterns are unexplored.

OBJECTIVES

To describe the dynamic and personal character of CRI risk evolution observed through continuous vital sign monitoring of individual step-down unit patients.

METHODS

Using a machine learning model, we estimated risk trends for CRI (defined as exceedance of vital sign stability thresholds) for each of 1,971 admissions (1,880 unique patients) to a 24-bed adult surgical trauma step-down unit at an urban teaching hospital in Pittsburgh, Pennsylvania using continuously recorded vital signs from standard bedside monitors. We compared and contrasted risk trends during initial 4-hour periods after step-down unit admission, and again during the 4 hours immediately before the CRI event, between cases (ever had a CRI) and control subjects (never had a CRI). We further explored heterogeneity of risk escalation patterns during the 4 hours before CRI among cases, comparing personalized to nonpersonalized risk.

MEASUREMENTS AND MAIN RESULTS

Estimated risk was significantly higher for cases (918) than control subjects (1,053; P ≤ 0.001) during the initial 4-hour stable periods. Among cases, the aggregated nonpersonalized risk trend increased 2 hours before the CRI, whereas the personalized risk trend became significantly different from control subjects 90 minutes ahead. We further discovered several unique phenotypes of risk escalation patterns among cases for nonpersonalized (14.6% persistently high risk, 18.6% early onset, 66.8% late onset) and personalized risk (7.7% persistently high risk, 8.9% early onset, 83.4% late onset).

CONCLUSIONS

Insights from this proof-of-concept analysis may guide design of dynamic and personalized monitoring systems that predict CRI, taking into account the triage and real-time monitoring utility of vital signs. These monitoring systems may prove useful in the dynamic allocation of technological and clinical personnel resources in acute care hospitals.

Clinical Information Systems as the Backbone of a Complex Information Logistics Process: Findings from the Clinical Information Systems Perspective for 2016

Objective: To summarize recent research and to propose a selection of best papers published in 2016 in the field of Clinical Information Systems (CIS). Method: The query used to retrieve the articles for the CIS section of the 2016 edition of the IMIA Yearbook of Medical Informatics was reused. It again aimed at identifying relevant publications in the field of CIS from PubMed and Web of Science and comprised search terms from the Medical Subject Headings (MeSH) catalog as well as additional free text search terms. The retrieved articles were categorized in a multi-pass review carried out by the two section editors. The final selection of candidate papers was then peer-reviewed by Yearbook editors and external reviewers. Based on the review results, the best papers were then chosen at the selection meeting with the IMIA Yearbook editorial board. Text mining, term co-occurrence mapping, and topic modelling techniques were used to get an overview on the content of the retrieved articles. Results: The query was carried out in mid-January 2017, yielding a consolidated result set of 2,190 articles published in 921 different journals. Out of them, 14 papers were nominated as candidate best papers and three of them were finally selected as the best papers of the CIS field. The content analysis of the articles revealed the broad spectrum of topics covered by CIS research. Conclusions: The CIS field is multi-dimensional and complex. It is hard to draw a well-defined outline between CIS and other domains or other sections of the IMIA Yearbook. The trends observed in the previous years are progressing. Clinical information systems are more than just sociotechnical systems for data collection, processing, exchange, presentation, and archiving. They are the backbone of a complex, trans-institutional information logistics process.

Sensor, Signal, and Imaging Informatics

Objective: To summarize significant contributions to sensor, signal, and imaging informatics published in 2016. Methods: We conducted an extensive search using PubMed® and Web of Science® to identify the scientific contributions published in 2016 that addressed sensors, signals, and imaging in medical informatics. The three section editors selected 15 candidate best papers by consensus. Each candidate article was reviewed by the section editors and at least two other external reviewers. The final selection of the six best papers was conducted by the editorial board of the Yearbook. Results: The selected papers of 2016 demonstrate the important scientific advances in management and analysis of sensor, signal, and imaging information. Conclusion: The growing volume of signal and imaging data provides exciting new challenges and opportunities for research in medical informatics. Evolving technologies provide faster and more effective approaches for pattern recognition and diagnostic evaluation. The papers selected here offer a small glimpse of the high-quality scientific work published in 2016 in the domain of sensor, signal, and imaging informatics.

Risk for Cardiorespiratory Instability Following Transfer to a Monitored Step-Down Unit

BACKGROUND

Hospitalized patients who develop at least one instance of cardiorespiratory instability (CRI) have poorer outcomes. We sought to describe the admission characteristics, drivers, and time to onset of initial CRI events in monitored step-down unit (SDU) patients.

METHODS

Admission characteristics and continuous monitoring data (frequency 1/20 Hz) were recorded in 307 subjects. Vital sign deviations beyond local instability trigger threshold criteria, with a tolerance of 40 s and cumulative duration of 4 of 5 min, were classified as CRI events. The CRI driver was defined as the first vital sign to cross a threshold and meet persistence criteria. Time to onset of initial CRI was the number of days from SDU admission to initial CRI, and duration was length of the initial CRI epoch.

RESULTS

Subjects transferred to the SDU from units with higher monitoring capability were more likely to develop CRI (CRI n = 133 [44%] vs no CRI n = 174 [31%] P = .042). Time to onset varied according to the CRI driver. Subjects with at least one CRI event had a longer hospital stay (CRI 11.3 ± 10.2 d vs no CRI 7.8 ± 9.2 d, P < .001) and SDU stay (CRI 6.1 ± 4.9 d vs no CRI 3.5 ± 2.9 d, P < .001). First events were more often due to S_pO2 , whereas breathing frequency was the most common driver of all CRI.

CONCLUSIONS

Initial CRI most commonly occurred due to S_pO2 and was associated with prolonged SDU and hospital stay. Findings suggest the need for clinicians to more closely monitor SDU patients transferred from an ICU and parameters (S_pO2 , breathing frequency) that more commonly precede CRI events.