Implementation of artifact detection in critical care: a methodological review

Visual saliency detection approach for long-term ECG analysis

BACKGROUND AND OBJECTIVE

Detection and analysis of QRS-complex as well as the processing of electrocardiogram (ECG) signal using computers are being practiced for over the last fifty-eight years, approximately, and yet the thirst of designing superior ECG processing and recognition algorithms still captures researchers' attention around the globe. A saliency detection-based technique for the processing of one-dimensional biomedical signals such as ECG is proposed here for the first time, to the best or our knowledge.

METHODS AND RESULTS

In this proposed research work, first, a trigonometric threshold-based technique is used to identify the QRS-complexes from the ECG signal. Motion-artifact (MA) and sudden-change-in-baseline (SCB) types of noises are considered to be the toughest among others to filter out from the ECG signals as the bandwidths of these two types of noises overlap with that of the ECG. Only one feature is extracted from each of the QRS-complex-intervals, and the normalised values of this feature are arranged in the form of a gray-scale image. Then, a saliency detection-based technique is applied iteratively on the gray-scale image to detect those regions of the ECG signals, which are highly corrupted with MA and (or) SCB noises. Next, three unique geometric-features are extracted from the rest of the QRS-complexes, which are not corrupted with MA or SCB noises, and the normalised values of these three features are arranged in the form of an Red-Green-Blue (RGB) image. Again, the saliency detection-based technique is applied to identify the abnormal QRS-complexes from the RGB image.

CONCLUSIONS

The technique is tested on long-term ECG signals; totaling a duration of 17.54 days, and its performance is evaluated through both quantitative and qualitative measures. The applicability, scope of implement in real-time scenarios, advantage of the proposed technique over the existing ones are discussed with a group of clinicians and cardiologists, and very affirmative and encouraging responses are received from them.

A review on utilizing machine learning technology in the fields of electronic emergency triage and patient priority systems in telemedicine: Coherent taxonomy, motivations, open research challenges and recommendations for intelligent future work

BACKGROUND

With the remarkable increasing in the numbers of patients, the triaging and prioritizing patients into multi-emergency level is required to accommodate all the patients, save more lives, and manage the medical resources effectively. Triaging and prioritizing patients becomes particularly challenging especially for the patients who are far from hospital and use telemedicine system. To this end, the researchers exploiting the useful tool of machine learning to address this challenge. Hence, carrying out an intensive investigation and in-depth study in the field of using machine learning in E-triage and patient priority are essential and required.

OBJECTIVES

This research aims to (1) provide a literature review and an in-depth study on the roles of machine learning in the fields of electronic emergency triage (E-triage) and prioritize patients for fast healthcare services in telemedicine applications. (2) highlight the effectiveness of machine learning methods in terms of algorithms, medical input data, output results, and machine learning goals in remote healthcare telemedicine systems. (3) present the relationship between machine learning goals and the electronic triage processes specifically on the: triage levels, medical features for input, outcome results as outputs, and the relevant diseases. (4), the outcomes of our analyses are subjected to organize and propose a cross-over taxonomy between machine learning algorithms and telemedicine structure. (5) present lists of motivations, open research challenges and recommendations for future intelligent work for both academic and industrial sectors in telemedicine and remote healthcare applications.

METHODS

An intensive research is carried out by reviewing all articles related to the field of E-triage and remote priority systems that utilise machine learning algorithms and sensors. We have searched all related keywords to investigate the databases of Science Direct, IEEE Xplore, Web of Science, PubMed, and Medline for the articles, which have been published from January 2012 up to date.

RESULTS

A new crossover matching between machine learning methods and telemedicine taxonomy is proposed. The crossover-taxonomy is developed in this study to identify the relationship between machine learning algorithm and the equivalent telemedicine categories whereas the machine learning algorithm has been utilized. The impact of utilizing machine learning is composed in proposing the telemedicine architecture based on synchronous (real-time/ online) and asynchronous (store-and-forward / offline) structure. In addition to that, list of machine learning algorithms, list of the performance metrics, list of inputs data and outputs results are presented. Moreover, open research challenges, the benefits of utilizing machine learning and the recommendations for new research opportunities that need to be addressed for the synergistic integration of multidisciplinary works are organized and presented accordingly.

DISCUSSION

The state-of-the-art studies on the E-triage and priority systems that utilise machine learning algorithms in telemedicine architecture are discussed. This approach allows the researchers to understand the modernisation of healthcare systems and the efficient use of artificial intelligence and machine learning. In particular, the growing worldwide population and various chronic diseases such as heart chronic diseases, blood pressure and diabetes, require smart health monitoring systems in E-triage and priority systems, in which machine learning algorithms could be greatly beneficial.

CONCLUSIONS

Although research directions on E-triage and priority systems that use machine learning algorithms in telemedicine vary, they are equally essential and should be considered. Hence, we provide a comprehensive review to emphasise the advantages of the existing research in multidisciplinary works of artificial intelligence, machine learning and healthcare services.

Electrocardiogram Quality Assessment using Unsupervised Deep Learning

OBJECTIVE

Noise and disturbances hinder effective interpretation of recorded ECG. To identify the clean parts of a recording, free from such disturbances, various quality indicators have been developed. Previous instances of these indicators focus on human-defined desirable properties of a clean signal. The reliance on human-specified properties places an inherent limitation on the potential power of signal quality indicators. To move away from this limitation, we propose a data-driven quality indicator.

METHODS

We use an unsupervised deep learning model, the auto-encoder, to derive the quality indicator. For different quality assessment settings we compare the performance of our quality indicator with traditional indicators.

RESULTS

The data-driven method performs consistently strong across tasks while performance of the traditional indicators varies strongly from task to task.

CONCLUSION

This strong performance indicates the potential of data-driven quality indicators for use in ECG processing, removing the reliance on expert-specified desirable properties.

SIGNIFICANCE

The proposed methodology can easily be extended towards learning quality indicators in other data modalities.

Integrating Physiological Data Artifacts Detection With Clinical Decision Support Systems: Observational Study

BACKGROUND

Clinical decision support systems (CDSS) have the potential to lower the patient mortality and morbidity rates. However, signal artifacts present in physiological data affect the reliability and accuracy of the CDSS. Moreover, patient monitors and other medical devices generate false alarms while processing physiological data, further leading to alarm fatigue because of increased noise levels, staff disruption, and staff desensitization in busy critical care environments. This adversely affects the quality of care at the patient bedside. Hence, artifact detection (AD) algorithms play a crucial role in assessing the quality of physiological data and mitigating the impact of these artifacts.

OBJECTIVE

The aim of this study is to evaluate a novel AD framework for integrating AD algorithms with CDSS. We designed the framework with features that support real-time implementation within critical care. In this study, we evaluated the framework and its features in a false alarm reduction study. We developed static framework component models, followed by dynamic framework compositions to formulate four CDSS. We evaluated these formulations using neonatal patient data and validated the six framework features: flexibility, reusability, signal quality indicator standardization, scalability, customizability, and real-time implementation support.

METHODS

We developed four exemplar static AD components with standardized requirements and provisions interfaces that facilitate the interoperability of framework components. These AD components were mixed and matched into four different AD compositions to mitigate the artifacts' effects. We developed a novel static clinical event detection component that is integrated with each AD composition to formulate and evaluate a dynamic CDSS for peripheral oxygen saturation (SpO2) alarm generation. This study collected data from 11 patients with diverse pathologies in the neonatal intensive care unit. Collected data streams and corresponding alarms include pulse rate and SpO2 measured from a pulse oximeter (Masimo SET SmartPod) integrated with an Infinity Delta monitor and the heart rate derived from electrocardiography leads attached to a second Infinity Delta monitor.

RESULTS

A total of 119 SpO2 alarms were evaluated. The lowest achievable SpO2 false alarm rate was 39%, with a sensitivity of 80%. This demonstrates the framework's utility in identifying the best possible dynamic composition to serve the clinical need for false SpO2 alarm reduction and subsequent alarm fatigue, given the limitations of a small sample size.

CONCLUSIONS

The framework features, including reusability, signal quality indicator standardization, scalability, and customizability, allow the evaluation and comparison of novel CDSS formulations. The optimal solution for a CDSS can then be hard-coded and integrated within clinical workflows for real-time implementation. The flexibility to serve different clinical needs and standardized component interoperability of the framework supports the potential for a real-time clinical implementation of AD.

The Importance of Respiratory Rate Monitoring: From Healthcare to Sport and Exercise

Respiratory rate is a fundamental vital sign that is sensitive to different pathological conditions (e.g., adverse cardiac events, pneumonia, and clinical deterioration) and stressors, including emotional stress, cognitive load, heat, cold, physical effort, and exercise-induced fatigue. The sensitivity of respiratory rate to these conditions is superior compared to that of most of the other vital signs, and the abundance of suitable technological solutions measuring respiratory rate has important implications for healthcare, occupational settings, and sport. However, respiratory rate is still too often not routinely monitored in these fields of use. This review presents a multidisciplinary approach to respiratory monitoring, with the aim to improve the development and efficacy of respiratory monitoring services. We have identified thirteen monitoring goals where the use of the respiratory rate is invaluable, and for each of them we have described suitable sensors and techniques to monitor respiratory rate in specific measurement scenarios. We have also provided a physiological rationale corroborating the importance of respiratory rate monitoring and an original multidisciplinary framework for the development of respiratory monitoring services. This review is expected to advance the field of respiratory monitoring and favor synergies between different disciplines to accomplish this goal.

Generalizable deep temporal models for predicting episodes of sudden hypotension in critically ill patients: a personalized approach

The vast quantities of data generated and collected in the Intensive Care Unit (ICU) have given rise to large retrospective datasets that are frequently used for observational studies. The temporal nature and fine granularity of much of the data collected in the ICU enable the pursuit of predictive modeling. In particular, forecasting acute hypotensive episodes (AHE) in intensive care patients has been of interest to researchers in critical care medicine. Given an advance warning of an AHE, care providers may be prompted to search for evolving disease processes and help mitigate negative clinical outcomes. However, the conventionally adopted definition of an AHE does not account for inter-patient variability and is restrictive. To reflect the wider trend of global clinical and research efforts in precision medicine, we introduce a patient-specific definition of AHE in this study and propose deep learning based models to predict this novel definition of AHE in data from multiple independent institutions. We provide extensive evaluation of the models by studying their accuracies in detecting patient-specific AHEs with lead-times ranging from 10 min to 1 hour before the onset of the event. The resulting models achieve AUROC values ranging from 0.57–0.87 depending on the lead time of the prediction. We demonstrate the generalizability and robustness of our approach through the use of independent multi-institutional data.

Estimating Reliability of Signal Quality of Physiological Data from Data Statistics Itself for Real-time Wearables

Artificial intelligence (AI) algorithms including machine and deep learning relies on proper data for classification and subsequent action. However, real-time unsupervised streaming data might not be reliable, which can lead to reduced accuracy or high error rates. Estimating reliability of signals, such as from wearable sensors for disease monitoring, is thus important but challenging since signals can be noisy and vulnerable to artifacts. In this paper, we propose a novel "Data Reliability Metric (DReM)" and demonstrate the proof-of-concept with two bio signals: electrocardiogram (ECG) and photoplethysmogram (PPG). We explored various statistical features and developed Artificial Neural Network (ANN), Random Forest (RF) and Support Vector Machine (SVM) models to autonomously classify good quality signals from the bad quality signals. Our results demonstrate the performance of the classification with a cross-validation accuracy of 99.7%, sensitivity of 100%, precision of 97% and F-score of 96%. This work demonstrates the potential of DReM to objectively and automatically estimate signal quality in unsupervised real-time settings with low computational requirement suitable for low-power digital signal processing techniques on wearables.

Artifact removal from neurophysiological signals: impact on intracranial and arterial pressure monitoring in traumatic brain injury

OBJECTIVE

Monitoring intracranial and arterial blood pressure (ICP and ABP, respectively) provides crucial information regarding the neurological status of patients with traumatic brain injury (TBI). However, these signals are often heavily affected by artifacts, which may significantly reduce the reliability of the clinical determinations derived from the signals. The goal of this work was to eliminate signal artifacts from continuous ICP and ABP monitoring via deep learning techniques and to assess the changes in the prognostic capacities of clinical parameters after artifact elimination.

METHODS

The first 24 hours of monitoring ICP and ABP in a total of 309 patients with TBI was retrospectively analyzed. An artifact elimination model for ICP and ABP was constructed via a stacked convolutional autoencoder (SCAE) and convolutional neural network (CNN) with 10-fold cross-validation tests. The prevalence and prognostic capacity of ICP- and ABP-related clinical events were compared before and after artifact elimination.

RESULTS

The proposed SCAE-CNN model exhibited reliable accuracy in eliminating ABP and ICP artifacts (net prediction rates of 97% and 94%, respectively). The prevalence of ICP- and ABP-related clinical events (i.e., systemic hypotension, intracranial hypertension, cerebral hypoperfusion, and poor cerebrovascular reactivity) all decreased significantly after artifact removal.

CONCLUSIONS

The SCAE-CNN model can be reliably used to eliminate artifacts, which significantly improves the reliability and efficacy of ICP- and ABP-derived clinical parameters for prognostic determinations after TBI.

A practical approach to storage and retrieval of high-frequency physiological signals

OBJECTIVE

Storage of physiological waveform data for retrospective analysis presents significant challenges. Resultant data can be very large, and therefore becomes expensive to store and complicated to manage. Traditional database approaches are not appropriate for large scale storage of physiological waveforms. Our goal was to apply modern time series compression and indexing techniques to the problem of physiological waveform storage and retrieval.

APPROACH

We deployed a vendor-agnostic data collection system and developed domain-specific compression approaches that allowed long term storage of physiological waveform data and other associated clinical and medical device data. The database (called AtriumDB) also facilitates rapid retrieval of retrospective data for high-performance computing and machine learning applications.

MAIN RESULTS

A prototype system has been recording data in a 42-bed pediatric critical care unit at The Hospital for Sick Children in Toronto, Ontario since February 2016. As of December 2019, the database contains over 720,000 patient-hours of data collected from over 5300 patients, all with complete waveform capture. One year of full resolution physiological waveform storage from this 42-bed unit can be losslessly compressed and stored in less than 300 GB of disk space. Retrospective data can be delivered to analytical applications at a rate of up to 50 million time-value pairs per second.

SIGNIFICANCE

Stored data are not pre-processed or filtered. Having access to a large retrospective dataset with realistic artefacts lends itself to the process of anomaly discovery and understanding. Retrospective data can be replayed to simulate a realistic streaming data environment where analytical tools can be rapidly tested at scale.

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.

Groundtruth: A Matlab GUI for Artifact and Feature Identification in Physiological Signals

Groundtruth is a Matlab Graphical User Interface (GUI) developed for the identification of key features and artifacts within physiological signals. The ultimate aim of this GUI is to provide a simple means of assessing the performance of new sensors. Secondary, to this is providing a means of providing marked data, enabling assessment of automated artifact rejection and feature identification algorithms. With the emergence of new wearable sensor technologies, there is an unmet need for convenient assessment of device performance, and a faster means of assessing new algorithms. The proposed GUI allows interactive marking of artifact regions as well as simultaneous interactive identification of key features, e.g., respiration peaks in respiration signals, R-peaks in Electrocardiography signals, etc. In this paper, we present the base structure of the system, together with an example of its use for two simultaneously worn respiration sensors. The respiration rates are computed for both original as well as artifact removed data and validated using Bland–Altman plots. The respiration rates computed based on the proposed GUI (after artifact removal process) demonstrated consistent results for two respiration sensors after artifact removal process. Groundtruth is customizable, and alternative processing modules are easy to add/remove. Groundtruth is intended for open-source use.

Hemodynamic Instability and Cardiovascular Events After Traumatic Brain Injury Predict Outcome After Artifact Removal With Deep Belief Network Analysis

BACKGROUND

Hemodynamic instability and cardiovascular events heavily affect the prognosis of traumatic brain injury. Physiological signals are monitored to detect these events. However, the signals are often riddled with faulty readings, which jeopardize the reliability of the clinical parameters obtained from the signals. A machine-learning model for the elimination of artifactual events shows promising results for improving signal quality. However, the actual impact of the improvements on the performance of the clinical parameters after the elimination of the artifacts is not well studied.

MATERIALS AND METHODS

The arterial blood pressure of 99 subjects with traumatic brain injury was continuously measured for 5 consecutive days, beginning on the day of admission. The machine-learning deep belief network was constructed to automatically identify and remove false incidences of hypotension, hypertension, bradycardia, tachycardia, and alterations in cerebral perfusion pressure (CPP).

RESULTS

The prevalences of hypotension and tachycardia were significantly reduced by 47.5% and 13.1%, respectively, after suppressing false incidents (P=0.01). Hypotension was particularly effective at predicting outcome favorability and mortality after artifact elimination (P=0.015 and 0.027, respectively). In addition, increased CPP was also statistically significant in predicting outcomes (P=0.02).

CONCLUSIONS

The prevalence of false incidents due to signal artifacts can be significantly reduced using machine-learning. Some clinical events, such as hypotension and alterations in CPP, gain particularly high predictive capacity for patient outcomes after artifacts are eliminated from physiological signals.

Electrocardiogram Signal Quality Assessment Based on Structural Image Similarity Metric

OBJECTIVE

We developed an image-based electrocardiographic (ECG) quality assessment technique that mimics how clinicians annotate ECG signal quality.

METHODS

We adopted the structural similarity measure (SSIM) to compare images of two ECG records that are obtained from displaying ECGs in a standard scale. Then, a subset of representative ECG images from the training set was selected as templates through a clustering method. SSIM between each image and all the templates were used as the feature vector for the linear discriminant analysis classifier. We also employed three commonly used ECG signal quality index (SQI) measures: baseSQI, kSQI, and sSQI to compare with the proposed image quality index (IQI) approach. We used 1926 annotated ECGs, recorded from patient monitors, and associated with six different ECG arrhythmia alarm types which were obtained previously from an ECG alarm study at the University of California, San Francisco (UCSF). In addition, we applied the templates from the UCSF database to test the SSIM approach on the publicly available PhysioNet Challenge 2011 data.

RESULTS

For the UCSF database, the proposed IQI algorithm achieved an accuracy of 93.1% and outperformed all the SQI metrics, baseSQI, kSQI, and sSQI, with accuracies of 85.7%, 63.7%, and 73.8% respectively. Moreover, evaluation of our algorithm on the PhysioNet data showed an accuracy of 82.5%.

CONCLUSION

The proposed algorithm showed better performance for assessing ECG signal quality than traditional signal processing methods.

SIGNIFICANCE

A more accurate assessment of ECG signal quality can lead to a more robust ECG-based diagnosis of cardiovascular conditions.

Robust estimation of respiratory rate via ECG- and PPG-derived respiratory quality indices

Respiratory rate (RR) is one of the most informative indicators of a patient's health status. However, automated, non-invasive measurements of RR are insufficiently robust for use in clinical practice. A number of methods have been described in the literature to estimate RR from both photo-plethysmography (PPG) and electrocardiography (ECG) based on three physiological modulations of respiration: amplitude modulation (AM), frequency modulation (FM), and baseline wander (BW). However, the quality of the respiratory information acquired is highly patient-dependent and often too noisy to be used. We address this by proposing respiratory quality indices (RQIs) that quantify the quality of the respiratory signal that can be extracted from each modulation from both PPG and ECG waveforms. Signal quality indices (SQIs) detect artefact in the ECG and PPG, which is relatively straight-forward. RQIs have a different role: they quantify if an individual patient's physiology is modulating the sensor waveforms. We have designed four RQIs based on Fourier transform (RQIFFT), autocorrelation (RQIAC), autoregression (RQIAR), and Hjorth complexity (RQIHC). We validated the approach using PPG and ECG data in the CapnoBase and MIMIC II datasets. We conclude that the novel implementation of an RQI-based preprocessing step has the potential to improve substantially the performance of RR estimation algorithms.

Hypotension Risk Prediction via Sequential Contrast Patterns of ICU Blood Pressure

Acute hypotension is a significant risk factor for in-hospital mortality at intensive care units. Prolonged hypotension can cause tissue hypoperfusion, leading to cellular dysfunction and severe injuries to multiple organs. Prompt medical interventions are thus extremely important for dealing with acute hypotensive episodes (AHE). Population level prognostic scoring systems for risk stratification of patients are suboptimal in such scenarios. However, the design of an efficient risk prediction system can significantly help in the identification of critical care patients, who are at risk of developing an AHE within a future time span. Toward this objective, a pattern mining algorithm is employed to extract informative sequential contrast patterns from hemodynamic data, for the prediction of hypotensive episodes. The hypotensive and normotensive patient groups are extracted from the MIMIC-II critical care research database, following an appropriate clinical inclusion criteria. The proposed method consists of a data preprocessing step to convert the blood pressure time series into symbolic sequences, using a symbolic aggregate approximation algorithm. Then, distinguishing subsequences are identified using the sequential contrast mining algorithm. These subsequences are used to predict the occurrence of an AHE in a future time window separated by a user-defined gap interval. Results indicate that the method performs well in terms of the prediction performance as well as in the generation of sequential patterns of clinical significance. Hence, the novelty of sequential patterns is in their usefulness as potential physiological biomarkers for building optimal patient risk stratification systems and for further clinical investigation of interesting patterns in critical care patients.

Real-time arrhythmia detection with supplementary ECG quality and pulse wave monitoring for the reduction of false alarms in ICUs

False intensive care unit (ICU) alarms induce stress in both patients and clinical staff and decrease the quality of care, thus significantly increasing both the hospital recovery time and rehospitalization rates. In the PhysioNet/CinC Challenge 2015 for reducing false arrhythmia alarms in ICU bedside monitor data, this paper validates the application of a real-time arrhythmia detection library (ADLib, Schiller AG) for the robust detection of five types of life-threatening arrhythmia alarms. The strength of the application is to give immediate feedback on the arrhythmia event within a scan interval of 3 s-7.5 s, and to increase the noise immunity of electrocardiogram (ECG) arrhythmia analysis by fusing its decision with supplementary ECG quality interpretation and real-time pulse wave monitoring (quality and hemodynamics) using arterial blood pressure or photoplethysmographic signals. We achieved the third-ranked real-time score (79.41) in the challenge (Event 1), however, the rank was not officially recognized due to the 'closed-source' entry. This study shows the optimization of the alarm decision module, using tunable parameters such as the scan interval, lead quality threshold, and pulse wave features, with a follow-up improvement of the real-time score (80.07). The performance (true positive rate, true negative rate) is reported in the blinded challenge test set for different arrhythmias: asystole (83%, 96%), extreme bradycardia (100%, 90%), extreme tachycardia (98%, 80%), ventricular tachycardia (84%, 82%), and ventricular fibrillation (78%, 84%). Another part of this study considers the validation of ADLib with four reference ECG databases (AHA, EDB, SVDB, MIT-BIH) according to the international recommendations for performance reports in ECG monitors (ANSI/AAMI EC57). The sensitivity (Se) and positive predictivity (+P) are: QRS detector QRS (Se, +P)  >  99.7%, ventricular ectopic beat (VEB) classifier VEB (Se, +P)  =  95%, and ventricular fibrillation detector VFIB (P  +  = 94.8%)  >  VFIB (Se  =  86.4%), adjusted to the clinical setting requirements, giving preference to low false positive alarms.

Machine Learning and Decision Support in Critical Care

Clinical data management systems typically provide caregiver teams with useful information, derived from large, sometimes highly heterogeneous, data sources that are often changing dynamically. Over the last decade there has been a significant surge in interest in using these data sources, from simply re-using the standard clinical databases for event prediction or decision support, to including dynamic and patient-specific information into clinical monitoring and prediction problems. However, in most cases, commercial clinical databases have been designed to document clinical activity for reporting, liability and billing reasons, rather than for developing new algorithms. With increasing excitement surrounding "secondary use of medical records" and "Big Data" analytics, it is important to understand the limitations of current databases and what needs to change in order to enter an era of "precision medicine." This review article covers many of the issues involved in the collection and preprocessing of critical care data. The three challenges in critical care are considered: compartmentalization, corruption, and complexity. A range of applications addressing these issues are covered, including the modernization of static acuity scoring; on-line patient tracking; personalized prediction and risk assessment; artifact detection; state estimation; and incorporation of multimodal data sources such as genomic and free text data.

Real-Time and Retrospective Health-Analytics-as-a-Service: A Novel Framework

BACKGROUND

Analytics-as-a-service (AaaS) is one of the latest provisions emerging from the cloud services family. Utilizing this paradigm of computing in health informatics will benefit patients, care providers, and governments significantly. This work is a novel approach to realize health analytics as services in critical care units in particular.

OBJECTIVE

To design, implement, evaluate, and deploy an extendable big-data compatible framework for health-analytics-as-a-service that offers both real-time and retrospective analysis.

METHODS

We present a novel framework that can realize health data analytics-as-a-service. The framework is flexible and configurable for different scenarios by utilizing the latest technologies and best practices for data acquisition, transformation, storage, analytics, knowledge extraction, and visualization. We have instantiated the proposed method, through the Artemis project, that is, a customization of the framework for live monitoring and retrospective research on premature babies and ill term infants in neonatal intensive care units (NICUs).

RESULTS

We demonstrated the proposed framework in this paper for monitoring NICUs and refer to it as the Artemis-In-Cloud (Artemis-IC) project. A pilot of Artemis has been deployed in the SickKids hospital NICU. By infusing the output of this pilot set up to an analytical model, we predict important performance measures for the final deployment of Artemis-IC. This process can be carried out for other hospitals following the same steps with minimal effort. SickKids' NICU has 36 beds and can classify the patients generally into 5 different types including surgical and premature babies. The arrival rate is estimated as 4.5 patients per day, and the average length of stay was calculated as 16 days. Mean number of medical monitoring algorithms per patient is 9, which renders 311 live algorithms for the whole NICU running on the framework. The memory and computation power required for Artemis-IC to handle the SickKids NICU will be 32 GB and 16 CPU cores, respectively. The required amount of storage was estimated as 8.6 TB per year. There will always be 34.9 patients in SickKids NICU on average. Currently, 46% of patients cannot get admitted to SickKids NICU due to lack of resources. By increasing the capacity to 90 beds, all patients can be accommodated. For such a provisioning, Artemis-IC will need 16 TB of storage per year, 55 GB of memory, and 28 CPU cores.

CONCLUSIONS

Our contributions in this work relate to a cloud architecture for the analysis of physiological data for clinical decisions support for tertiary care use. We demonstrate how to size the equipment needed in the cloud for that architecture based on a very realistic assessment of the patient characteristics and the associated clinical decision support algorithms that would be required to run for those patients. We show the principle of how this could be performed and furthermore that it can be replicated for any critical care setting within a tertiary institution.

Performance Evaluation of New-Generation Pulse Oximeters in the NICU: Observational Study

This crossover observational study compares the data characteristics and performance of new-generation Nellcor OXIMAX and Masimo SET SmartPod pulse oximeter technologies. The study was conducted independent of either original equipment manufacturer (OEM) across eleven preterm infants in a Neonatal Intensive Care Unit (NICU). The SmartPods were integrated with Dräger Infinity Delta monitors. The Delta monitor measured the heart rate (HR) using an independent electrocardiogram sensor, and the two SmartPods collected arterial oxygen saturation (SpO2) and pulse rate (PR). All patient data were non-Gaussian. Nellcor PR showed a higher correlation with the HR as compared to Masimo PR. The statistically significant difference found in their median values (1% for SpO2, 1 bpm for PR) was deemed clinically insignificant. SpO2 alarms generated by both SmartPods were observed and categorized for performance evaluation. Results for sensitivity, positive predictive value, accuracy and false alarm rates were Nellcor (80.3, 50, 44.5, 50%) and Masimo (72.2, 48.2, 40.6, 51.8%) respectively. These metrics were not statistically significantly different between the two pulse oximeters. Despite claims by OEMs, both pulse oximeters exhibited high false alarm rates, with no statistically or clinically significant difference in performance. These findings have a direct impact on alarm fatigue in the NICU. Performance evaluation studies can also impact medical device purchase decisions made by hospital administrators.

MS-QI: A Modulation Spectrum-Based ECG Quality Index for Telehealth Applications

As telehealth applications emerge, the need for accurate and reliable biosignal quality indices has increased. One typical modality used in remote patient monitoring is the electrocardiogram (ECG), which is inherently susceptible to several different noise sources, including environmental (e.g., powerline interference), experimental (e.g., movement artifacts), and physiological (e.g., muscle and breathing artifacts). Accurate measurement of ECG quality can allow for automated decision support systems to make intelligent decisions about patient conditions. This is particularly true for in-home monitoring applications, where the patient is mobile and the ECG signal can be severely corrupted by movement artifacts. In this paper, we propose an innovative ECG quality index based on the so-called modulation spectral signal representation. The representation quantifies the rate of change of ECG spectral components, which are shown to be different from the rate of change of typical ECG noise sources. The proposed modulation spectral-based quality index, MS-QI, was tested on 1) synthetic ECG signals corrupted by varying levels of noise, 2) single-lead recorded data using the Hexoskin garment during three activity levels (sitting, walking, running), 3) 12-lead recorded data using conventional ECG machines (Computing in Cardiology 2011 dataset), and 4) two-lead ambulatory ECG recorded from arrhythmia patients (MIT-BIH Arrhythmia Database). Experimental results showed the proposed index outperforming two conventional benchmark quality measures, particularly in the scenarios involving recorded data in real-world environments.

Signal-quality indices for the electrocardiogram and photoplethysmogram: derivation and applications to wireless monitoring

The identification of invalid data in recordings obtained using wearable sensors is of particular importance since data obtained from mobile patients is, in general, noisier than data obtained from nonmobile patients. In this paper, we present a signal quality index (SQI), which is intended to assess whether reliable heart rates (HRs) can be obtained from electrocardiogram (ECG) and photoplethysmogram (PPG) signals collected using wearable sensors. The algorithms were validated on manually labeled data. Sensitivities and specificities of 94% and 97% were achieved for the ECG and 91% and 95% for the PPG. Additionally, we propose two applications of the SQI. First, we demonstrate that, by using the SQI as a trigger for a power-saving strategy, it is possible to reduce the recording time by up to 94% for the ECG and 93% for the PPG with only minimal loss of valid vital-sign data. Second, we demonstrate how an SQI can be used to reduce the error in the estimation of respiratory rate (RR) from the PPG. The performance of the two applications was assessed on data collected from a clinical study on hospital patients who were able to walk unassisted.