MIMIC-IV, a freely accessible electronic health record dataset.

Digital data collection during routine clinical practice is now ubiquitous within hospitals. The data contains valuable information on the care of patients and their response to treatments, offering exciting opportunities for research. Typically, data are stored within archival systems that are not intended to support research. These systems are often inaccessible to researchers and structured for optimal storage, rather than interpretability and analysis. Here we present MIMIC-IV, a publicly available database sourced from the electronic health record of the Beth Israel Deaconess Medical Center. Information available includes patient measurements, orders, diagnoses, procedures, treatments, and deidentified free-text clinical notes. MIMIC-IV is intended to support a wide array of research studies and educational material, helping to reduce barriers to conducting clinical research.

A contrastive learning approach for ICU false arrhythmia alarm reduction.

The high rate of false arrhythmia alarms in Intensive Care Units (ICUs) can lead to disruption of care, negatively impacting patients' health through noise disturbances, and slow staff response time due to alarm fatigue. Prior false-alarm reduction approaches are often rule-based and require hand-crafted features from physiological waveforms as inputs to machine learning classifiers. Despite considerable prior efforts to address the problem, false alarms are a continuing problem in the ICUs. In this work, we present a deep learning framework to automatically learn feature representations of physiological waveforms using convolutional neural networks (CNNs) to discriminate between true vs. false arrhythmia alarms. We use Contrastive Learning to simultaneously minimize a binary cross entropy classification loss and a proposed similarity loss from pair-wise comparisons of waveform segments over time as a discriminative constraint. Furthermore, we augment our deep models with learned embeddings from a rule-based method to leverage prior domain knowledge for each alarm type. We evaluate our method using the dataset from the 2015 PhysioNet Computing in Cardiology Challenge. Ablation analysis demonstrates that Contrastive Learning significantly improves the performance of a combined deep learning and rule-based-embedding approach. Our results indicate that the final proposed deep learning framework achieves superior performance in comparison to the winning entries of the Challenge.

Post-extrasystolic characteristics in the arterial blood pressure waveform are associated with right ventricular dysfunction in intensive care patients.

Right ventricular dysfunction (RVD) is associated with end-organ dysfunction and mortality, but has been an overlooked condition in the ICU. We hypothesized that analysis of the arterial waveform in the presence of ventricular extrasystoles could differentiate patients with RVD from patients with a normally functioning right ventricle, because the 2nd and 3rd post-ectopic beat could reflect right ventricular state (pulmonary transit time) during the preceding ectopy. We retrospectively identified patients with echocardiographic evidence of moderate-to-severe RVD and patients with a normal functioning right ventricle (control) from the MIMIC database. We identified waveform records where ECG and arterial pressure were available in combination, simultaneously with echocardiographic evaluation. Ventricular extrasystoles were visually confirmed and the median systolic blood pressure (SBP) of the 2nd and 3rd post-ectopic beats compared with the median SBP of the ten sinus beats preceding the extrasystole. We identified 34 patients in the control group and 24 patients in the RVD group with ventricular extrasystoles. The mean SBP reduction at the 2nd and 3rd beat was lower in the RVD group compared with the control group [- 1.7 (SD: 1.9) % vs. - 3.6 (SD: 1.9) %, p < 0.001], and this characteristic differentiated RVD subjects from control subjects with an AUC of 0.76 (CI [0.64; 0.89]), with a specificity of 91% and sensitivity of 50%. In this proof-of-concept study, we found that post-extrasystolic ABP characteristics were associated with RVD.

You Snooze, You Win: the PhysioNet/Computing in Cardiology Challenge 2018.

The PhysioNet/Computing in Cardiology Challenge 2018 focused on the use of various physiological signals (EEG, EOG, EMG, ECG, SaO2) collected during polysomnographic sleep studies to detect sources of arousal (non-apnea) during sleep. A total of 1,983 polysomnographic recordings were made available to the entrants. The arousal labels for 994 of the recordings were made available in a public training set while 989 labels were retained in a hidden test set. Challengers were asked to develop an algorithm that could label the presence of arousals within the hidden test set. The performance metric used to assess entrants was the area under the precision-recall curve. A total of twenty-two independent teams entered the Challenge, deploying a variety of methods from generalized linear models to deep neural networks.

AF Classification from a Short Single Lead ECG Recording: the PhysioNet/Computing in Cardiology Challenge 2017.

The PhysioNet/Computing in Cardiology (CinC) Challenge 2017 focused on differentiating AF from noise, normal or other rhythms in short term (from 9-61 s) ECG recordings performed by patients. A total of 12,186 ECGs were used: 8,528 in the public training set and 3,658 in the private hidden test set. Due to the high degree of inter-expert disagreement between a significant fraction of the expert labels we implemented a mid-competition bootstrap approach to expert relabeling of the data, levering the best performing Challenge entrants' algorithms to identify contentious labels. A total of 75 independent teams entered the Challenge using a variety of traditional and novel methods, ranging from random forests to a deep learning approach applied to the raw data in the spectral domain. Four teams won the Challenge with an equal high F1 score (averaged across all classes) of 0.83, although the top 11 algorithms scored within 2% of this. A combination of 45 algorithms identified using LASSO achieved an F1 of 0.87, indicating that a voting approach can boost performance.

An open access database for the evaluation of heart sound algorithms.

In the past few decades, analysis of heart sound signals (i.e. the phonocardiogram or PCG), especially for automated heart sound segmentation and classification, has been widely studied and has been reported to have the potential value to detect pathology accurately in clinical applications. However, comparative analyses of algorithms in the literature have been hindered by the lack of high-quality, rigorously validated, and standardized open databases of heart sound recordings. This paper describes a public heart sound database, assembled for an international competition, the PhysioNet/Computing in Cardiology (CinC) Challenge 2016. The archive comprises nine different heart sound databases sourced from multiple research groups around the world. It includes 2435 heart sound recordings in total collected from 1297 healthy subjects and patients with a variety of conditions, including heart valve disease and coronary artery disease. The recordings were collected from a variety of clinical or nonclinical (such as in-home visits) environments and equipment. The length of recording varied from several seconds to several minutes. This article reports detailed information about the subjects/patients including demographics (number, age, gender), recordings (number, location, state and time length), associated synchronously recorded signals, sampling frequency and sensor type used. We also provide a brief summary of the commonly used heart sound segmentation and classification methods, including open source code provided concurrently for the Challenge. A description of the PhysioNet/CinC Challenge 2016, including the main aims, the training and test sets, the hand corrected annotations for different heart sound states, the scoring mechanism, and associated open source code are provided. In addition, several potential benefits from the public heart sound database are discussed.

False alarm reduction in critical care.

High false alarm rates in the ICU decrease quality of care by slowing staff response times while increasing patient delirium through noise pollution. The 2015 PhysioNet/Computing in Cardiology Challenge provides a set of 1250 multi-parameter ICU data segments associated with critical arrhythmia alarms, and challenges the general research community to address the issue of false alarm suppression using all available signals. Each data segment was 5 minutes long (for real time analysis), ending at the time of the alarm. For retrospective analysis, we provided a further 30 seconds of data after the alarm was triggered. A total of 750 data segments were made available for training and 500 were held back for testing. Each alarm was reviewed by expert annotators, at least two of whom agreed that the alarm was either true or false. Challenge participants were invited to submit a complete, working algorithm to distinguish true from false alarms, and received a score based on their program's performance on the hidden test set. This score was based on the percentage of alarms correct, but with a penalty that weights the suppression of true alarms five times more heavily than acceptance of false alarms. We provided three example entries based on well-known, open source signal processing algorithms, to serve as a basis for comparison and as a starting point for participants to develop their own code. A total of 38 teams submitted a total of 215 entries in this year's Challenge. This editorial reviews the background issues for this challenge, the design of the challenge itself, the key achievements, and the follow-up research generated as a result of the Challenge, published in the concurrent special issue of Physiological Measurement. Additionally we make some recommendations for future changes in the field of patient monitoring as a result of the Challenge.

MIMIC-III, a freely accessible critical care database.

MIMIC-III ('Medical Information Mart for Intensive Care') is a large, single-center database comprising information relating to patients admitted to critical care units at a large tertiary care hospital. Data includes vital signs, medications, laboratory measurements, observations and notes charted by care providers, fluid balance, procedure codes, diagnostic codes, imaging reports, hospital length of stay, survival data, and more. The database supports applications including academic and industrial research, quality improvement initiatives, and higher education coursework.

Robust detection of heart beats in multimodal data.

This editorial reviews the background issues, the design, the key achievements, and the follow-up research generated as a result of the PhysioNet/Computing in Cardiology (CinC) Challenge 2014, published in the concurrent focus issue of Physiological Measurement. Our major focus was to accelerate the development and facilitate the comparison of robust methods for locating heart beats in long-term multi-channel recordings. A public (training) database consisting of 151 032 annotated beats was compiled from records that contained ECGs as well as pulsatile signals that directly reflect cardiac activity, and other signals that may have few or no observable markers of heart beats. A separate hidden test data set (consisting of 152 478 beats) is permanently stored at PhysioNet, and a public framework has been developed to provide researchers with the ability to continue to automatically score and compare the performance of their algorithms. A scoring criteria based on the averaging of gross sensitivity, gross positive predictivity, average sensitivity, and average positive predictivity is proposed. The top three scores (as of March 2015) on the hidden test data set were 93.64%, 91.50%, and 90.70%.

The PhysioNet/Computing in Cardiology Challenge 2015: Reducing False Arrhythmia Alarms in the ICU.

High false alarm rates in the ICU decrease quality of care by slowing staff response times while increasing patient delirium through noise pollution. The 2015 Physio-Net/Computing in Cardiology Challenge provides a set of 1,250 multi-parameter ICU data segments associated with critical arrhythmia alarms, and challenges the general research community to address the issue of false alarm suppression using all available signals. Each data segment was 5 minutes long (for real time analysis), ending at the time of the alarm. For retrospective analysis, we provided a further 30 seconds of data after the alarm was triggered. A collection of 750 data segments was made available for training and a set of 500 was held back for testing. Each alarm was reviewed by expert annotators, at least two of whom agreed that the alarm was either true or false. Challenge participants were invited to submit a complete, working algorithm to distinguish true from false alarms, and received a score based on their program's performance on the hidden test set. This score was based on the percentage of alarms correct, but with a penalty that weights the suppression of true alarms five times more heavily than acceptance of false alarms. We provided three example entries based on well-known, open source signal processing algorithms, to serve as a basis for comparison and as a starting point for participants to develop their own code. A total of 38 teams submitted a total of 215 entries in this year's Challenge.

Acute lower urinary tract dysfunction (LUTD) following traumatic brain injury (TBI) in rats.

AIMS: The aim of this study was to assess experimental traumatic brain injury (TBI)-induced lower urinary tract dysfunction (LUTD) by monitoring systemic and urodynamic parameters using an implantable telemetry system. METHODS: A single lateral fluid percussion TBI (FP-TBI; 3.4 atm) was administered to 10 female rats. Pressure micro-catheters were implanted in the abdominal aorta and bladder dome for simultaneous data recording. Hemodynamic and urodynamic variables recorded 24 hr before and 24 hr after injury were analyzed and compared. RESULTS: TBI in the acute phase resulted in LUTD affecting bladder emptying, characterized by failure of voiding reflex, high capacity bladder, increased voided volume, prolonged intermicturition intervals, and loss of compliance. The dominant symptom was urinary retention (100%) and incontinence (60%). The effects followed a pattern of initial loss of bladder function followed by either altered recovery of reflex micturition or a period of incontinence. With a moderate injury symptoms were temporary in 90% of animals and permanent in 10% of animals. Injury produced only transient hypertension (≤1 hr) with a maximum systolic pressure of 172.64 ± 14.53 mmHg (70% of animals). CONCLUSIONS: The results demonstrate that experimental FP-TBI causes temporary bladder dysfunction that in more severe cases becomes permanent. Telemetry recordings revealed a sequence of events following injury that establishes moderate TBI as a risk factor for neurogenic bladder disorder. Results also suggest a correlation between lateral FP-TBI and incontinence.

Multiparameter Intelligent Monitoring in Intensive Care II: a public-access intensive care unit database.

OBJECTIVE: We sought to develop an intensive care unit research database applying automated techniques to aggregate high-resolution diagnostic and therapeutic data from a large, diverse population of adult intensive care unit patients. This freely available database is intended to support epidemiologic research in critical care medicine and serve as a resource to evaluate new clinical decision support and monitoring algorithms. DESIGN: Data collection and retrospective analysis. SETTING: All adult intensive care units (medical intensive care unit, surgical intensive care unit, cardiac care unit, cardiac surgery recovery unit) at a tertiary care hospital. PATIENTS: Adult patients admitted to intensive care units between 2001 and 2007. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC-II) database consists of 25,328 intensive care unit stays. The investigators collected detailed information about intensive care unit patient stays, including laboratory data, therapeutic intervention profiles such as vasoactive medication drip rates and ventilator settings, nursing progress notes, discharge summaries, radiology reports, provider order entry data, International Classification of Diseases, 9th Revision codes, and, for a subset of patients, high-resolution vital sign trends and waveforms. Data were automatically deidentified to comply with Health Insurance Portability and Accountability Act standards and integrated with relational database software to create electronic intensive care unit records for each patient stay. The data were made freely available in February 2010 through the Internet along with a detailed user's guide and an assortment of data processing tools. The overall hospital mortality rate was 11.7%, which varied by critical care unit. The median intensive care unit length of stay was 2.2 days (interquartile range, 1.1-4.4 days). According to the primary International Classification of Diseases, 9th Revision codes, the following disease categories each comprised at least 5% of the case records: diseases of the circulatory system (39.1%); trauma (10.2%); diseases of the digestive system (9.7%); pulmonary diseases (9.0%); infectious diseases (7.0%); and neoplasms (6.8%). CONCLUSIONS: MIMIC-II documents a diverse and very large population of intensive care unit patient stays and contains comprehensive and detailed clinical data, including physiological waveforms and minute-by-minute trends for a subset of records. It establishes a new public-access resource for critical care research, supporting a diverse range of analytic studies spanning epidemiology, clinical decision-rule development, and electronic tool development.

In situ monitoring of adipogenesis with human-adipose-derived stem cells using surface-enhanced Raman spectroscopy.

Methods capable of nondestructively collecting high-quality, real-time chemical information from living human stem cells are of increasing importance given the escalating relevance of stem cells in therapeutic and regenerative medicines. Raman spectroscopy is one such technique that can nondestructively collect real-time chemical information. Living cells uptake gold nanoparticles and transport these particles through an endosomal pathway. Once inside the endosome, nanoparticles aggregate into clusters that give rise to large spectroscopic enhancements that can be used to elucidate local chemical environments through the use of surface-enhanced Raman spectroscopy. This report uses 40-nm colloidal gold nanoparticles to create volumes of surface-enhanced Raman scattering (SERS) within living human-adipose-derived adult stem cells enabling molecular information to be monitored. We exploit this method to spectroscopically observe chemical changes that occur during the adipogenic differentiation of human-adipose-derived stem cells over a period of 22 days. It is shown that intracellular SERS is able to detect the production of lipids as little as one day after the onset of adipogenesis and that a complex interplay between lipids, proteins, and chemical messengers can be observed shortly thereafter. After 22 days of differentiation, the cells show visible and spectroscopic indications of completed adipogenesis yet still share spectral features common to the progenitor stem cells.

Using surface-enhanced Raman spectroscopy to probe for genetic markers on single-stranded DNA.

Methods capable of quickly and inexpensively collecting genetic information are of increasing importance. We report a method of using surface-enhanced Raman spectroscopy to probe single-stranded DNA for genetic markers. This unique approach is used to analyze unmodified genes of moderate length for genetic markers by hybridizing native test oligonucleotides into a surface-enhanced Raman complex, vastly increasing detection sensitivity as compared to traditional Raman spectroscopy. The Raman complex is formed by sandwiching the test DNA between 40-nm gold nanoparticles and a photolithographically defined gold surface. With this design, we are able to collect characteristic Raman spectra about the test DNA and to detect genetic markers such as single-nucleotide polymorphisms (SNPs) and polymorphic regions. Results show that strands containing one of three different types of polymorphism can be differentiated using statistically significant trends regarding Raman intensity.

Enhancing Electrochemical Detection by Scaling Solid State Nanogaps.

The ability to quickly and inexpensively fabricate planar solid state nanogaps has enabled research to be effectively performed on devices down to just a few nanometers. Here, nanofabricated electrode pairs with electrode-to-electrode spacings of <4, 6 and 20 nm are utilized for monitoring an electroactive molecules, dopamine, in ionic solution. The results show a several order of magnitude enhancement of the electrochemical signal, collected current, for the solid state nanogaps with 6 nm electrode-electrode spacings as compared to traditional microelectrodes. The data from the <4 nm and 20 nm solid state nanogaps verify that this enhancement is due to cycling of the redox molecules in the confined geometry of the nanogap. In addition the data collected for the <4 nm nanogap emphasizes and reinforces that scaling does have limits and that as device sizes move to the few nanometer scale, the influence of a molecule's size and other physical properties becomes increasingly important and can eventually dominate the generated signals.

Specific oxygen-containing functional groups on the carbon surface underlie an enhanced sensitivity to dopamine at electrochemically pretreated carbon fiber microelectrodes.

The in vivo use of carbon-fiber microelectrodes for neurochemical investigation has proven to be selective and sensitive when coupled with background-subtracted fast-scan cyclic voltammetry (FSCV). Various electrochemical pretreatments have been established to enhance the sensitivity of these sensors; however, the fundamental chemical mechanisms underlying these enhancement strategies remain poorly understood. We have investigated an electrochemical pretreatment in which an extended triangular waveform from -0.5 to 1.8 V is applied to the electrode prior to the voltammetric detection of dopamine using a more standard waveform ranging from -0.4 to 1.3 V. This pretreatment enhances the electron-transfer kinetics and significantly improves sensitivity. To gain insight into the chemical mechanism, the electrodes were studied using common analytical techniques. Contact atomic force microscopy (AFM) was used to demonstrate that the surface roughness was not altered on the nanoscale by electrochemical pretreatment. Raman spectroscopy was utilized to investigate oxide functionalities on the carbon surface and confirmed that carbonyl and hydroxyl functional groups were increased by electrochemical conditioning. Spectra collected after the selective chemical modification of these groups implicate the hydroxyl functionality, rather than the carbonyl, as the major contributor to the enhanced electrochemical signal. Finally, we have demonstrated that this electrochemical pretreatment can be used to create carbon microdisc electrodes with sensitivities comparable to those associated with larger, conventionally treated cylindrical carbon fiber microelectrodes.

Simultaneous decoupled detection of dopamine and oxygen using pyrolyzed carbon microarrays and fast-scan cyclic voltammetry.

Microfabricated structures utilizing pyrolyzed photoresist have been shown to be useful for monitoring electrochemical processes. These previous studies, however, were limited to constant-potential measurements and slow-scan voltammetry. The work described in this paper utilizes microfabrication processes to produce devices that enable multiple fast-scan cyclic voltammetry (FSCV) waveforms to be applied to different electrodes on a single substrate. This enabled the simultaneous, decoupled detection of dopamine and oxygen. In this paper we describe the fabrication process of these arrays and show that pyrolyzed photoresist electrodes possess surface chemistry and electrochemical properties comparable to PAN-type, T-650, carbon fiber microelectrodes using background-subtracted FSCV. The functionality of the array is discussed in terms of the degree of cross talk in response to flow injections of physiologically relevant concentrations of dopamine and oxygen. Finally, other applications of pyrolyzed photoresist microelectrode arrays are shown, including spatially resolved detection of analytes and combining FSCV with amperometry for the detection of dopamine.