Dataset Information

SWIFT: A Deep Learning Approach to Prediction of Hypoxemic Events in Critically-Ill Patients Using SpO 2 Waveform Prediction.

ABSTRACT: Hypoxemia is a significant driver of mortality and poor clinical outcomes in conditions such as brain injury and cardiac arrest in critically ill patients, including COVID-19 patients. Given the host of negative clinical outcomes attributed to hypoxemia, identifying patients likely to experience hypoxemia would offer valuable opportunities for early and thus more effective intervention. We present SWIFT (SpO 2 W aveform I CU F orecasting T echnique), a deep learning model that predicts blood oxygen saturation (SpO 2 ) waveforms 5 and 30 minutes in the future using only prior SpO 2 values as inputs. When tested on novel data, SWIFT predicts more than 80% and 60% of hypoxemic events in critically ill and COVID-19 patients, respectively. SWIFT also predicts SpO 2 waveforms with average MSE below .0007. SWIFT provides information on both occurrence and magnitude of potential hypoxemic events 30 minutes in advance, allowing it to be used to inform clinical interventions, patient triaging, and optimal resource allocation. SWIFT may be used in clinical decision support systems to inform the management of critically ill patients during the COVID-19 pandemic and beyond.

SUBMITTER: Annapragada AV

PROVIDER: S-EPMC7941630 | biostudies-literature | 2021 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

SWIFT: A Deep Learning Approach to Prediction of Hypoxemic Events in Critically-Ill Patients Using SpO <sub>2</sub> Waveform Prediction.

Annapragada Akshaya V AV Greenstein Joseph L JL Bose Sanjukta N SN Winters Bradford D BD Sarma Sridevi V SV Winslow Raimond L RL

medRxiv : the preprint server for health sciences 20210305

Hypoxemia is a significant driver of mortality and poor clinical outcomes in conditions such as brain injury and cardiac arrest in critically ill patients, including COVID-19 patients. Given the host of negative clinical outcomes attributed to hypoxemia, identifying patients likely to experience hypoxemia would offer valuable opportunities for early and thus more effective intervention. We present SWIFT (SpO <sub>2</sub> W aveform I CU F orecasting T echnique), a deep learning model that predict ...[more]

PMID: 33688661

Similar Datasets

Project description:Rationale & objectiveRisk prediction tools for assisting acute kidney injury (AKI) management have focused on AKI onset but have infrequently addressed kidney recovery. We developed clinical models for risk stratification of mortality and major adverse kidney events (MAKE) in critically ill patients with incident AKI.Study designMulticenter cohort study.Setting & participants9,587 adult patients admitted to heterogeneous intensive care units (ICUs; March 2009 to February 2017) who experienced AKI within the first 3 days of their ICU stays.PredictorsMultimodal clinical data consisting of 71 features collected in the first 3 days of ICU stay.Outcomes(1) Hospital mortality and (2) MAKE, defined as the composite of death during hospitalization or within 120 days of discharge, receipt of kidney replacement therapy in the last 48 hours of hospital stay, initiation of maintenance kidney replacement therapy within 120 days, or a ≥50% decrease in estimated glomerular filtration rate from baseline to 120 days from hospital discharge.Analytical approachFour machine-learning algorithms (logistic regression, random forest, support vector machine, and extreme gradient boosting) and the SHAP (Shapley Additive Explanations) framework were used for feature selection and interpretation. Model performance was evaluated by 10-fold cross-validation and external validation.ResultsOne developed model including 15 features outperformed the SOFA (Sequential Organ Failure Assessment) score for the prediction of hospital mortality, with areas under the curve of 0.79 (95% CI, 0.79-0.80) and 0.71 (95% CI, 0.71-0.71) in the development cohort and 0.74 (95% CI, 0.73-0.74) and 0.71 (95% CI, 0.71-0.71) in the validation cohort (P < 0.001 for both). A second developed model including 14 features outperformed KDIGO (Kidney Disease: Improving Global Outcomes) AKI severity staging for the prediction of MAKE: 0.78 (95% CI, 0.78-0.78) versus 0.66 (95% CI, 0.66-0.66) in the development cohort and 0.73 (95% CI, 0.72-0.74) versus 0.67 (95% CI, 0.67-0.67) in the validation cohort (P < 0.001 for both).LimitationsThe models are applicable only to critically ill adult patients with incident AKI within the first 3 days of an ICU stay.ConclusionsThe reported clinical models exhibited better performance for mortality and kidney recovery prediction than standard scoring tools commonly used in critically ill patients with AKI in the ICU. Additional validation is needed to support the utility and implementation of these models.Plain-language summaryAcute kidney injury (AKI) occurs commonly in critically ill patients admitted to the intensive care unit (ICU) and is associated with high morbidity and mortality rates. Prediction of mortality and recovery after an episode of AKI may assist bedside decision making. In this report, we describe the development and validation of a clinical model using data from the first 3 days of an ICU stay to predict hospital mortality and major adverse kidney events occurring as long as 120 days after hospital discharge among critically ill adult patients who experienced AKI within the first 3 days of an ICU stay. The proposed clinical models exhibited good performance for outcome prediction and, if further validated, could enable risk stratification for timely interventions that promote kidney recovery.

Project description:Background Waveform parameters provide approximate data about aortic wave reflection. However, their association with cardiovascular events remains controversial and their role in cardiovascular prediction is unknown. Methods and Results We analyzed participants aged between 40 and 69 from the population-based CARTaGENE cohort. Baseline pulse wave analysis (central pulse pressure, augmentation index) and wave separation analysis (forward pressure, backward pressure, reflection magnitude) parameters were derived from radial artery tonometry. Associations between each parameter and major adverse atherosclerotic events (MACE; cardiovascular death, stroke, myocardial infarction) were obtained using adjusted Cox models. The incremental predictive value of each parameter compared with the 10-year atherosclerotic cardiovascular disease score alone was assessed using hazard ratios, c-index differences, continuous net reclassification indexes, and integrated discrimination indexes. From 17 561 eligible patients, 2315 patients had a MACE during a median follow-up of 10.1 years. Central pulse pressure, forward pressure, and backward pressure, but not augmentation index and reflection magnitude, were significantly associated with MACE after full adjustment. All parameters except forward pressure statistically improved MACE prediction compared with the atherosclerotic cardiovascular disease score alone. The greatest prediction improvement was seen with augmentation index and reflection magnitude but remained small in magnitude. These 2 parameters enhanced predictive performance more strongly in patients with low baseline atherosclerotic cardiovascular disease scores. Up to 5.7% of individuals were reclassified into a different risk stratum by adding waveform parameters to atherosclerotic cardiovascular disease scores. Conclusions Some waveform parameters are independently associated with MACEs in a population-based cohort. Augmentation index and reflection magnitude slightly improve risk prediction, especially in patients at low cardiovascular risk.

Project description:Sepsis is defined by life-threatening organ dysfunction during infection and is the leading cause of death in hospitals. During sepsis, there is a high risk that new onset of atrial fibrillation (AF) can occur, which is associated with significant morbidity and mortality. Consequently, early prediction of AF during sepsis would allow testing of interventions in the intensive care unit (ICU) to prevent AF and its severe complications. In this paper, we present a novel automated AF prediction algorithm for critically ill sepsis patients using electrocardiogram (ECG) signals. From the heart rate signal collected from 5-min ECG, feature extraction is performed using the traditional time, frequency, and nonlinear domain methods. Moreover, variable frequency complex demodulation and tunable Q-factor wavelet-transform-based time-frequency methods are applied to extract novel features from the heart rate signal. Using a selected feature subset, several machine learning classifiers, including support vector machine (SVM) and random forest (RF), were trained using only the 2001 Computers in Cardiology data set. For testing the proposed method, 50 critically ill ICU subjects from the Medical Information Mart for Intensive Care (MIMIC) III database were used in this study. Using distinct and independent testing data from MIMIC III, the SVM achieved 80% sensitivity, 100% specificity, 90% accuracy, 100% positive predictive value, and 83.33% negative predictive value for predicting AF immediately prior to the onset of AF, while the RF achieved 88% AF prediction accuracy. When we analyzed how much in advance we can predict AF events in critically ill sepsis patients, the algorithm achieved 80% accuracy for predicting AF events 10 min early. Our algorithm outperformed a state-of-the-art method for predicting AF in ICU patients, further demonstrating the efficacy of our proposed method. The annotations of patients' AF transition information will be made publicly available for other investigators. Our algorithm to predict AF onset is applicable for any ECG modality including patch electrodes and wearables, including Holter, loop recorder, and implantable devices.

Project description:ObjectiveGuidelines recommend that encephalopathic critically ill children undergo continuous electroencephalographic (CEEG) monitoring for electrographic seizure (ES) identification and management. However, limited data exist on antiseizure medication (ASM) safety for ES treatment in critically ill children.MethodsWe performed a single-center prospective observational study of encephalopathic critically ill children undergoing CEEG. Clinical and EEG features and ASM utilization patterns were evaluated. We determined the incidence, types, and risk factors for adverse events associated with ASM administration.ResultsA total of 472 consecutive critically ill children undergoing CEEG were enrolled. ES occurred in 131 children (28%). Clinicians administered ASM to 108 children with ES (82%). ES terminated after the initial ASM in 38% of patients who received one ASM, after the second ASM in 35% of patients who received two ASMs, after the third ASM in 50% of patients who received three ASMs, and after the fourth ASM in 53% of patients who received four ASMs. Thirty patients (28%) received anesthetic infusions for ES management. Adverse events occurred in 18 patients (17%). Adverse effects were expected and resolved in all patients, and they were generally serious (in 15 patients) and definitely related (in 12 patients). Adverse events were rare in patients with acute symptomatic seizures requiring only one to two ASMs for treatment, but were more common in children with epilepsy, ictal-interictal continuum EEG patterns, or patients requiring more extensive ASM management.SignificanceES ceased after one ASM in only 38% of critically ill children but ceased after two ASMs in 73% of critically ill children. Thus, ES management was often accomplished with readily available medications, but optimization of multistep ES management strategies might be beneficial. Adverse events were rare and manageable in children with acute symptomatic seizures requiring only one to two ASMs for treatment. Future studies are needed to determine whether management of acute symptomatic ES improves neurobehavioral outcomes.

Dataset Information

SWIFT: A Deep Learning Approach to Prediction of Hypoxemic Events in Critically-Ill Patients Using SpO 2 Waveform Prediction.

Publications

SWIFT: A Deep Learning Approach to Prediction of Hypoxemic Events in Critically-Ill Patients Using SpO <sub>2</sub> Waveform Prediction.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets