Data-driven discovery of a novel sepsis pre-shock state predicts impending septic shock in the ICU

Rethinking Human-AI Collaboration in Complex Medical Decision Making: A Case Study in Sepsis Diagnosis

Today's AI systems for medical decision support often succeed on benchmark datasets in research papers but fail in real-world deployment. This work focuses on the decision making of sepsis, an acute life-threatening systematic infection that requires an early diagnosis with high uncertainty from the clinician. Our aim is to explore the design requirements for AI systems that can support clinical experts in making better decisions for the early diagnosis of sepsis. The study begins with a formative study investigating why clinical experts abandon an existing AI-powered Sepsis predictive module in their electrical health record (EHR) system. We argue that a human-centered AI system needs to support human experts in the intermediate stages of a medical decision-making process (e.g., generating hypotheses or gathering data), instead of focusing only on the final decision. Therefore, we build SepsisLab based on a state-of-the-art AI algorithm and extend it to predict the future projection of sepsis development, visualize the prediction uncertainty, and propose actionable suggestions (i.e., which additional laboratory tests can be collected) to reduce such uncertainty. Through heuristic evaluation with six clinicians using our prototype system, we demonstrate that SepsisLab enables a promising human-AI collaboration paradigm for the future of AI-assisted sepsis diagnosis and other high-stakes medical decision making.

Use of Artificial Intelligence in Improving Outcomes in Heart Disease: A Scientific Statement From the American Heart Association

A major focus of academia, industry, and global governmental agencies is to develop and apply artificial intelligence and other advanced analytical tools to transform health care delivery. The American Heart Association supports the creation of tools and services that would further the science and practice of precision medicine by enabling more precise approaches to cardiovascular and stroke research, prevention, and care of individuals and populations. Nevertheless, several challenges exist, and few artificial intelligence tools have been shown to improve cardiovascular and stroke care sufficiently to be widely adopted. This scientific statement outlines the current state of the art on the use of artificial intelligence algorithms and data science in the diagnosis, classification, and treatment of cardiovascular disease. It also sets out to advance this mission, focusing on how digital tools and, in particular, artificial intelligence may provide clinical and mechanistic insights, address bias in clinical studies, and facilitate education and implementation science to improve cardiovascular and stroke outcomes. Last, a key objective of this scientific statement is to further the field by identifying best practices, gaps, and challenges for interested stakeholders.

Predicting septic shock in patients with sepsis at emergency department triage using systolic and diastolic shock index

INTRODUCTION

Identifying patients with at a high risk of progressing to septic shock is essential. Due to systemic vasodilation in the pathophysiology of septic shock, the use of diastolic blood pressure (DBP) has emerged. We hypothesized that the initial shock index (SI) and diastolic SI (DSI) at the emergency department (ED) triage can predict septic shock.

METHOD

This observational study used the prospectively collected sepsis registry. The primary outcome was progression to septic shock. Secondary outcomes were the time to vasopressor requirement, vasopressor dose, and severity according to SI and DSI. Patients were classified by tertiles according to the first principal component of shock index and diastolic shock index.

RESULTS

A total of 1267 patients were included in the analysis. The area under the receiver operating characteristic curve (AUC) for predicting progression to septic shock for DSI was 0.717, while that for SI was 0.707. The AUC for predicting progression to septic shock for DSI and SI were significantly higher than those for conventional early warning scores. Middle tertile showed adjusted Odd ratio (aOR) of 1.448 (95% CI 1.074-1.953), and that of upper tertile showed 3.704 (95% CI 2.299-4.111).

CONCLUSION

The SI and DSI were significant predictors of progression to septic shock. Our findings suggest an association between DSI and vasopressor requirement. We propose stratifying lower tertile as being at low risk, middle tertile as being at intermediate risk, and upper tertile as being at high risk of progression to septic shock. This system can be applied simply at the ED triage.

Attention-based multimodal fusion with contrast for robust clinical prediction in the face of missing modalities

OBJECTIVE

With the increasing amount and growing variety of healthcare data, multimodal machine learning supporting integrated modeling of structured and unstructured data is an increasingly important tool for clinical machine learning tasks. However, it is non-trivial to manage the differences in dimensionality, volume, and temporal characteristics of data modalities in the context of a shared target task. Furthermore, patients can have substantial variations in the availability of data, while existing multimodal modeling methods typically assume data completeness and lack a mechanism to handle missing modalities.

METHODS

We propose a Transformer-based fusion model with modality-specific tokens that summarize the corresponding modalities to achieve effective cross-modal interaction accommodating missing modalities in the clinical context. The model is further refined by inter-modal, inter-sample contrastive learning to improve the representations for better predictive performance. We denote the model as Attention-based cRoss-MOdal fUsion with contRast (ARMOUR). We evaluate ARMOUR using two input modalities (structured measurements and unstructured text), six clinical prediction tasks, and two evaluation regimes, either including or excluding samples with missing modalities.

RESULTS

Our model shows improved performances over unimodal or multimodal baselines in both evaluation regimes, including or excluding patients with missing modalities in the input. The contrastive learning improves the representation power and is shown to be essential for better results. The simple setup of modality-specific tokens enables ARMOUR to handle patients with missing modalities and allows comparison with existing unimodal benchmark results.

CONCLUSION

We propose a multimodal model for robust clinical prediction to achieve improved performance while accommodating patients with missing modalities. This work could inspire future research to study the effective incorporation of multiple, more complex modalities of clinical data into a single model.

Updating mortality risk estimation in intensive care units from high-dimensional electronic health records with incomplete data

BACKGROUND

The risk of mortality in intensive care units (ICUs) is currently addressed by the implementation of scores using admission data. Their performances are satisfactory when complications occur early after admission; however, they may become irrelevant in the case of long hospital stays. In this study, we developed predictive models of short-term mortality in the ICU from longitudinal data.

METHODS

Using data collected throughout patients' stays of at least 48 h from the MIMIC-III database, several statistical learning approaches were compared, including deep neural networks and penalized regression. Missing data were handled using complete-case analysis or multiple imputation.

RESULTS

Complete-case analyses from 19 predictors showed good discrimination (AUC > 0.77 for several approaches) to predict death between 12 and 24 h onward, yet excluded 75% of patients from the initial target cohort, as data was missing for some of the predictors. Multiple imputation allowed us to include 70 predictors and keep 95% of patients, with similar performances.

CONCLUSION

This proof-of-concept study supports that automated analysis of electronic health records can be of great interest throughout patients' stays as a surveillance tool. Although this framework relies on a large set of predictors, it is robust to data imputation and may be effective early after admission, when data are still scarce.

Clinical Sepsis Phenotypes in Critically Ill Patients

Sepsis, defined as the life-threatening dysregulated host response to an infection leading to organ dysfunction, is considered as one of the leading causes of mortality worldwide, especially in intensive care units (ICU). Moreover, sepsis remains an enigmatic clinical syndrome, with complex pathophysiology incompletely understood and a great heterogeneity both in terms of clinical expression, patient response to currently available therapeutic interventions and outcomes. This heterogeneity proves to be a major obstacle in our quest to deliver improved treatment in septic critical care patients; thus, identification of clinical phenotypes is absolutely necessary. Although this might be seen as an extremely difficult task, nowadays, artificial intelligence and machine learning techniques can be recruited to quantify similarities between individuals within sepsis population and differentiate them into distinct phenotypes regarding not only temperature, hemodynamics or type of organ dysfunction, but also fluid status/responsiveness, trajectories in ICU and outcome. Hopefully, we will eventually manage to determine both the subgroup of septic patients that will benefit from a therapeutic intervention and the correct timing of applying the intervention during the disease process.

lab: an R package for generating analysis-ready data from laboratory records

Background

Electronic health records (EHRs) play a crucial role in healthcare decision-making by giving physicians insights into disease progression and suitable treatment options. Within EHRs, laboratory test results are frequently utilized for predicting disease progression. However, processing laboratory test results often poses challenges due to variations in units and formats. In addition, leveraging the temporal information in EHRs can improve outcomes, prognoses, and diagnosis predication. Nevertheless, the irregular frequency of the data in these records necessitates data preprocessing, which can add complexity to time-series analyses.

Methods

To address these challenges, we developed an open-source R package that facilitates the extraction of temporal information from laboratory records. The proposed lab package generates analysis-ready time series data by segmenting the data into time-series windows and imputing missing values. Moreover, users can map local laboratory codes to the Logical Observation Identifier Names and Codes (LOINC), an international standard. This mapping allows users to incorporate additional information, such as reference ranges and related diseases. Moreover, the reference ranges provided by LOINC enable us to categorize results into normal or abnormal. Finally, the analysis-ready time series data can be further summarized using descriptive statistics and utilized to develop models using machine learning technologies.

Results

Using the lab package, we analyzed data from MIMIC-III, focusing on newborns with patent ductus arteriosus (PDA). We extracted time-series laboratory records and compared the differences in test results between patients with and without 30-day in-hospital mortality. We then identified significant variations in several laboratory test results 7 days after PDA diagnosis. Leveraging the time series-analysis-ready data, we trained a prediction model with the long short-term memory algorithm, achieving an area under the receiver operating characteristic curve of 0.83 for predicting 30-day in-hospital mortality in model training. These findings demonstrate the lab package's effectiveness in analyzing disease progression.

Conclusions

The proposed lab package simplifies and expedites the workflow involved in laboratory records extraction. This tool is particularly valuable in assisting clinical data analysts in overcoming the obstacles associated with heterogeneous and sparse laboratory records.

The NUTRIC Score as a Tool to Predict Mortality and Increased Resource Utilization in Intensive Care Patients with Sepsis

The Nutrition Risk in Critically Ill score (NUTRIC) is an important nutritional risk assessment instrument for patients in the intensive care unit (ICU). The purpose of this study was to evaluate the power of the score to predict mortality in patients treated for sepsis and to forecast increased resource utilization and nursing workload in the ICU. The NUTRIC score predicted mortality (AUC 0.833, p < 0.001) with the optimal cut-off value of 6 points. Among patients with a score ≥ 6 on ICU admission, the 28-day mortality was 61%, and 10% with a score < 6 (p < 0.001). In addition, a NUTRIC score of ≥6 was associated with a more intense use of ICU resources, as evidenced by a higher proportion of patients requiring vasopressor infusion (98 vs. 82%), mechanical ventilation (99 vs. 87%), renal replacement therapy (54 vs. 26%), steroids (68 vs. 31%), and blood products (60 vs. 43%); the nursing workload was also significantly higher in this group. In conclusion, the NUTRIC score obtained at admission to the ICU provided a good discriminative value for mortality and makes it possible to identify patients who will ultimately require intense use of ICU resources and an associated increase in the nursing workload during treatment.

Septic Cardiomyopathy: From Pathophysiology to the Clinical Setting

The onset of cardiomyopathy is a common feature in sepsis, with relevant effects on its pathophysiology and clinical care. Septic cardiomyopathy is characterized by reduced left ventricular (LV) contractility eventually associated with LV dilatation with or without right ventricle failure. Unfortunately, such a wide range of ultrasonographic findings does not reflect a deep comprehension of sepsis-induced cardiomyopathy, but rather a lack of consensus about its definition. Several echocardiographic parameters intrinsically depend on loading conditions (both preload and afterload) so that it may be challenging to discriminate which is primitive and which is induced by hemodynamic perturbances. Here, we explore the state of the art in sepsis-related cardiomyopathy. We focus on the shortcomings in its definition and point out how cardiac performance dynamically changes in response to different hemodynamic clusters. A special attention is also given to update the knowledge about molecular mechanisms leading to myocardial dysfunction and that recall those of myocardial hibernation. Ultimately, the aim of this review is to highlight the unsolved issue in the field of sepsis-induced cardiomyopathy as their implementation would lead to improve risk stratification and clinical care.

Risk Management In Intensive Care Units With Artificial Intelligence Technologies: Systematic Review of Prediction Models Using Electronic Health Records

Background and aim: Clinical risk assessments should be made to protect patients from negative outcomes, and the definition, frequency and severity of the risk should be determined. The information contained in the electronic health records (EHRs) can use in different areas such as risk prediction, estimation of treatment effect ect. Many prediction models using artificial intelligence (AI) technologies that can be used in risk assessment have been developed. The aim of this study is to bring together the researches on prediction models developed with AI technologies using the EHRs of patients hospitalized in the intensive care unit (ICU) and to evaluate them in terms of risk management in healthcare. Methods: The study restricted the search to the Web of Science, Pubmed, Science Direct, and Medline databases to retrieve research articles published in English in 2010 and after. Studies with a prediction model using data obtained from EHRs in the ICU are included. The study focused solely on research conducted in ICU to predict a health condition that poses a significant risk to patient safety using artificial intellegence (AI) technologies. Results: Recognized prediction subcategories were mortality (n=6), sepsis (n=4), pressure ulcer (n=4), acute kidney injury (n=3), and other areas (n=10). It has been found that EHR-based prediction models are good risk management and decision support tools and adoption of such models in ICUs may reduce the prevalence of adverse conditions. Conclusions: The article results remarks that developed models was found to have higher performance and better selectivity than previously developed risk models, so they are better at predicting risks and serious adverse events in ICU. It is recommended to use AI based prediction models developed using EHRs in risk management studies. Future work is still needed to researches to predict different health conditions risks.

Enhancing sepsis management through machine learning techniques: A review

Sepsis is a major public health problem and a leading cause of death in the world, where delay in the beginning of treatment, along with clinical guidelines non-adherence have been proved to be associated with higher mortality. Machine Learning is increasingly being adopted in developing innovative Clinical Decision Support Systems in many areas of medicine, showing a great potential for automatic prediction of diverse patient conditions, as well as assistance in clinical decision making. In this context, this work conducts a narrative review to provide an overview of how specific Machine Learning techniques can be used to improve sepsis management, discussing the main tasks addressed, the most popular methods and techniques, as well as the obtained results, in terms of both intelligent system accuracy and clinical outcomes improvement.

Sepsis prediction, early detection, and identification using clinical text for machine learning: a systematic review

OBJECTIVE

To determine the effects of using unstructured clinical text in machine learning (ML) for prediction, early detection, and identification of sepsis.

MATERIALS AND METHODS

PubMed, Scopus, ACM DL, dblp, and IEEE Xplore databases were searched. Articles utilizing clinical text for ML or natural language processing (NLP) to detect, identify, recognize, diagnose, or predict the onset, development, progress, or prognosis of systemic inflammatory response syndrome, sepsis, severe sepsis, or septic shock were included. Sepsis definition, dataset, types of data, ML models, NLP techniques, and evaluation metrics were extracted.

RESULTS

The clinical text used in models include narrative notes written by nurses, physicians, and specialists in varying situations. This is often combined with common structured data such as demographics, vital signs, laboratory data, and medications. Area under the receiver operating characteristic curve (AUC) comparison of ML methods showed that utilizing both text and structured data predicts sepsis earlier and more accurately than structured data alone. No meta-analysis was performed because of incomparable measurements among the 9 included studies.

DISCUSSION

Studies focused on sepsis identification or early detection before onset; no studies used patient histories beyond the current episode of care to predict sepsis. Sepsis definition affects reporting methods, outcomes, and results. Many methods rely on continuous vital sign measurements in intensive care, making them not easily transferable to general ward units.

CONCLUSIONS

Approaches were heterogeneous, but studies showed that utilizing both unstructured text and structured data in ML can improve identification and early detection of sepsis.

Ground truth labels challenge the validity of sepsis consensus definitions in critical illness

BACKGROUND

Sepsis is the leading cause of death in the intensive care unit (ICU). Expediting its diagnosis, largely determined by clinical assessment, improves survival. Predictive and explanatory modelling of sepsis in the critically ill commonly bases both outcome definition and predictions on clinical criteria for consensus definitions of sepsis, leading to circularity. As a remedy, we collected ground truth labels for sepsis.

METHODS

In the Ground Truth for Sepsis Questionnaire (GTSQ), senior attending physicians in the ICU documented daily their opinion on each patient's condition regarding sepsis as a five-category working diagnosis and nine related items. Working diagnosis groups were described and compared and their SOFA-scores analyzed with a generalized linear mixed model. Agreement and discriminatory performance measures for clinical criteria of sepsis and GTSQ labels as reference class were derived.

RESULTS

We analyzed 7291 questionnaires and 761 complete encounters from the first survey year. Editing rates for all items were > 90%, and responses were consistent with current understanding of critical illness pathophysiology, including sepsis pathogenesis. Interrater agreement for presence and absence of sepsis was almost perfect but only slight for suspected infection. ICU mortality was 19.5% in encounters with SIRS as the "worst" working diagnosis compared to 5.9% with sepsis and 5.9% with severe sepsis without differences in admission and maximum SOFA. Compared to sepsis, proportions of GTSQs with SIRS plus acute organ dysfunction were equal and macrocirculatory abnormalities higher (p < 0.0001). SIRS proportionally ranked above sepsis in daily assessment of illness severity (p < 0.0001). Separate analyses of neurosurgical referrals revealed similar differences. Discriminatory performance of Sepsis-1/2 and Sepsis-3 compared to GTSQ labels was similar with sensitivities around 70% and specificities 92%. Essentially no difference between the prevalence of SIRS and SOFA ≥ 2 yielded sensitivities and specificities for detecting sepsis onset close to 55% and 83%, respectively.

CONCLUSIONS

GTSQ labels are a valid measure of sepsis in the ICU. They reveal suspicion of infection as an unclear clinical concept and refute an illness severity hierarchy in the SIRS-sepsis-severe sepsis spectrum. Ground truth challenges the accuracy of Sepsis-1/2 and Sepsis-3 in detecting sepsis onset. It is an indispensable intermediate step towards advancing diagnosis and therapy in the ICU and, potentially, other health care settings.

Early Identification of Resuscitated Patients with a Significant Coronary Disease in Out-of-Hospital Cardiac Arrest Survivors without ST-Segment Elevation

This study aimed to develop a machine learning (ML)-based model for identifying patients who had a significant coronary artery disease among out-of-hospital cardiac arrest (OHCA) survivors without ST-segment elevation (STE). This multicenter observational study used data from the Korean Hypothermia Network prospective registry (KORHN-PRO) gathered between October 2015 and December 2018. We used information available before targeted temperature management (TTM) as predictor variables, and the primary outcome was a significant coronary artery lesion in coronary angiography (CAG). Among 1373 OHCA patients treated with TTM, 331 patients without STE who underwent CAG were enrolled. Among them, 127 patients (38.4%) had a significant coronary artery lesion. Four ML algorithms, namely regularized logistic regression (RLR), random forest classifier (RF), CatBoost classifier (CBC), and voting classifier (VC), were used with data collected before CAG. The VC model showed the highest accuracy for predicting significant lesions (area under the curve of 0.751). Eight variables (older age, male, initial shockable rhythm, shorter total collapse duration, higher glucose and creatinine, and lower pH and lactate) were significant to ML models. These results showed that ML models may be useful in developing early predictive tools for identifying high-risk patients with a significant stenosis in CAG.

The impact of recency and adequacy of historical information on sepsis predictions using machine learning

Sepsis is a major public and global health concern. Every hour of delay in detecting sepsis significantly increases the risk of death, highlighting the importance of accurately predicting sepsis in a timely manner. A growing body of literature has examined developing new or improving the existing machine learning (ML) approaches for timely and accurate predictions of sepsis. This study contributes to this literature by providing clear insights regarding the role of the recency and adequacy of historical information in predicting sepsis using ML. To this end, we implemented a deep learning model using a bidirectional long short-term memory (BiLSTM) algorithm and compared it with six other ML algorithms based on numerous combinations of the prediction horizons (to capture information recency) and observation windows (to capture information adequacy) using different measures of predictive performance. Our results indicated that the BiLSTM algorithm outperforms all other ML algorithms and provides a great separability of the predicted risk of sepsis among septic versus non-septic patients. Moreover, decreasing the prediction horizon (in favor of information recency) always boosts the predictive performance; however, the impact of expanding the observation window (in favor of information adequacy) depends on the prediction horizon and the purpose of prediction. More specifically, when the prediction is responsive to the positive label (i.e., Sepsis), increasing historical data improves the predictive performance when the prediction horizon is short-moderate.

Machine learning identification of specific changes in myeloid cell phenotype during bloodstream infections

The early identification of bacteremia is critical for ensuring appropriate treatment of nosocomial infections in intensive care unit (ICU) patients. The aim of this study was to use flow cytometric data of myeloid cells as a biomarker of bloodstream infection (BSI). An eight-color antibody panel was used to identify seven monocyte and two dendritic cell subsets. In the learning cohort, immunophenotyping was applied to (1) control subjects, (2) postoperative heart surgery patients, as a model of noninfectious inflammatory responses, and (3) blood culture-positive patients. Of the complex changes in the myeloid cell phenotype, a decrease in myeloid and plasmacytoid dendritic cell numbers, increase in CD14⁺CD16⁺ inflammatory monocyte numbers, and upregulation of neutrophils CD64 and CD123 expression were prominent in BSI patients. An extreme gradient boosting (XGBoost) algorithm called the “infection detection and ranging score” (iDAR), ranging from 0 to 100, was developed to identify infection-specific changes in 101 phenotypic variables related to neutrophils, monocytes and dendritic cells. The tenfold cross-validation achieved an area under the receiver operating characteristic (AUROC) of 0.988 (95% CI 0.985–1) for the detection of bacteremic patients. In an out-of-sample, in-house validation, iDAR achieved an AUROC of 0.85 (95% CI 0.71–0.98) in differentiating localized from bloodstream infection and 0.95 (95% CI 0.89–1) in discriminating infected from noninfected ICU patients. In conclusion, a machine learning approach was used to translate the changes in myeloid cell phenotype in response to infection into a score that could identify bacteremia with high specificity in ICU patients.

A Machine Learning Sepsis Prediction Algorithm for Intended Intensive Care Unit Use (NAVOY Sepsis): Proof-of-Concept Study

Background

Despite decades of research, sepsis remains a leading cause of mortality and morbidity in intensive care units worldwide. The key to effective management and patient outcome is early detection, for which no prospectively validated machine learning prediction algorithm is currently available for clinical use in Europe.

Objective

We aimed to develop a high-performance machine learning sepsis prediction algorithm based on routinely collected intensive care unit data, designed to be implemented in European intensive care units.

Methods

The machine learning algorithm was developed using convolutional neural networks, based on Massachusetts Institute of Technology Lab for Computational Physiology MIMIC-III clinical data from intensive care unit patients aged 18 years or older. The model uses 20 variables to produce hourly predictions of onset of sepsis, defined by international Sepsis-3 criteria. Predictive performance was externally validated using hold-out test data.

Results

The algorithm—NAVOY Sepsis—uses 4 hours of input and can identify patients with high risk of developing sepsis, with high performance (area under the receiver operating characteristics curve 0.90; area under the precision-recall curve 0.62) for predictions up to 3 hours before sepsis onset.

Conclusions

The prediction performance of NAVOY Sepsis was superior to that of existing sepsis early warning scoring systems and comparable with those of other prediction algorithms designed to predict sepsis onset. The algorithm has excellent predictive properties and uses variables that are routinely collected in intensive care units.

Early Prediction of Multiple Organ Dysfunction in the Pediatric Intensive Care Unit

Objective: The objective of the study is to build models for early prediction of risk for developing multiple organ dysfunction (MOD) in pediatric intensive care unit (PICU) patients. Design: The design of the study is a retrospective observational cohort study. Setting: The setting of the study is at a single academic PICU at the Johns Hopkins Hospital, Baltimore, MD. Patients: The patients included in the study were <18 years of age admitted to the PICU between July 2014 and October 2015. Measurements and main results: Organ dysfunction labels were generated every minute from preceding 24-h time windows using the International Pediatric Sepsis Consensus Conference (IPSCC) and Proulx et al. MOD criteria. Early MOD prediction models were built using four machine learning methods: random forest, XGBoost, GLMBoost, and Lasso-GLM. An optimal threshold learned from training data was used to detect high-risk alert events (HRAs). The early prediction models from all methods achieved an area under the receiver operating characteristics curve ≥0.91 for both IPSCC and Proulx criteria. The best performance in terms of maximum F1-score was achieved with random forest (sensitivity: 0.72, positive predictive value: 0.70, F1-score: 0.71) and XGBoost (sensitivity: 0.8, positive predictive value: 0.81, F1-score: 0.81) for IPSCC and Proulx criteria, respectively. The median early warning time was 22.7 h for random forest and 37 h for XGBoost models for IPSCC and Proulx criteria, respectively. Applying spectral clustering on risk-score trajectories over 24 h following early warning provided a high-risk group with ≥0.93 positive predictive value. Conclusions: Early predictions from risk-based patient monitoring could provide more than 22 h of lead time for MOD onset, with ≥0.93 positive predictive value for a high-risk group identified pre-MOD.

Multitask prediction of organ dysfunction in the intensive care unit using sequential subnetwork routing

Objective

Multitask learning (MTL) using electronic health records allows concurrent prediction of multiple endpoints. MTL has shown promise in improving model performance and training efficiency; however, it often suffers from negative transfer – impaired learning if tasks are not appropriately selected. We introduce a sequential subnetwork routing (SeqSNR) architecture that uses soft parameter sharing to find related tasks and encourage cross-learning between them.

Materials and Methods

Using the MIMIC-III (Medical Information Mart for Intensive Care-III) dataset, we train deep neural network models to predict the onset of 6 endpoints including specific organ dysfunctions and general clinical outcomes: acute kidney injury, continuous renal replacement therapy, mechanical ventilation, vasoactive medications, mortality, and length of stay. We compare single-task (ST) models with naive multitask and SeqSNR in terms of discriminative performance and label efficiency.

Results

SeqSNR showed a modest yet statistically significant performance boost across 4 of 6 tasks compared with ST and naive multitasking. When the size of the training dataset was reduced for a given task (label efficiency), SeqSNR outperformed ST for all cases showing an average area under the precision-recall curve boost of 2.1%, 2.9%, and 2.1% for tasks using 1%, 5%, and 10% of labels, respectively.

Conclusions

The SeqSNR architecture shows superior label efficiency compared with ST and naive multitasking, suggesting utility in scenarios in which endpoint labels are difficult to ascertain.

Prediction of Impending Septic Shock in Children With Sepsis

Objectives

Sepsis and septic shock are leading causes of in-hospital mortality. Timely treatment is crucial in improving patient outcome, yet treatment delays remain common. Early prediction of those patients with sepsis who will progress to its most severe form, septic shock, can increase the actionable window for interventions. We aim to extend a time-evolving risk score, previously developed in adult patients, to predict pediatric sepsis patients who are likely to develop septic shock before its onset, and to determine whether or not these risk scores stratify into groups with distinct temporal evolution once this prediction is made.

Design

Retrospective cohort study.

Setting

Academic medical center from July 1, 2016, to December 11, 2020.

Patients

Six-thousand one-hundred sixty-one patients under 18 admitted to the Johns Hopkins Hospital PICU.

Interventions

None.

Measurements and Main Results

We trained risk models to predict impending transition into septic shock and compute time-evolving risk scores representative of a patient's probability of developing septic shock. We obtain early prediction performance of 0.90 area under the receiver operating curve, 43% overall positive predictive value, patient-specific positive predictive value as high as 62%, and an 8.9-hour median early warning time using Sepsis-3 labels based on age-adjusted Sequential Organ Failure Assessment score. Using spectral clustering, we stratified pediatric sepsis patients into two clusters differing in septic shock prevalence, mortality, and proportion of patients adequately fluid resuscitated.

CONCLUSIONS

We demonstrate the applicability of our methodology for early prediction and stratification for risk of septic shock in pediatric sepsis patients. Through analyses of risk score evolution over time, we corroborate our past finding of an abrupt transition preceding onset of septic shock in children and are able to stratify pediatric sepsis patients using their risk score trajectories into low and high-risk categories.

Temporal patterns of organ dysfunction after severe trauma

Background

Understanding temporal patterns of organ dysfunction (OD) may aid early recognition of complications after trauma and assist timing and modality of treatment strategies. Our aim was to analyse and characterise temporal patterns of OD in intensive care unit-admitted trauma patients.

Methods

We used group-based trajectory modelling to identify temporal trajectories of OD after trauma. Modelling was based on the joint development of all six subdomains comprising the sequential organ failure assessment score measured daily during the first two weeks post trauma. Further, the time for trajectories to stabilise and transition to final group assignments were evaluated.

Results

Six-hundred and sixty patients were included in the final model. Median age was 40 years, and median ISS was 26 (IQR 17–38). We identified five distinct trajectories of OD. Group 1, mild OD (n = 300), median ISS of 20 (IQR 14–27), had an early resolution of OD and a low mortality. Group 2, moderate OD (n = 135), and group 3, severe OD (n = 87), were fairly similar in admission characteristics and initial OD but differed in subsequent OD trajectories, the latter experiencing an extended course and higher mortality. In group 3, 56% of the patients developed sepsis as compared with 19% in group 2. Group 4, extreme OD (n = 40), received most blood transfusions, had the highest proportion of shock at admission and a median ISS of 41 (IQR 29–50). They experienced significant and sustained OD affecting all organ systems and a 28-day mortality of 30%. Group 5, traumatic brain injury with OD (n = 98), had the highest mortality of 35% and the shortest time to death for non-survivors, median 3.5 (IQR 2.4–4.8) days. Groups 1 and 5 reached their final group assignment early, > 80% of the patients within 48 h. In contrast, groups 2 and 3 had a prolonged time to final group assignment.

Conclusions

We identified five distinct trajectories of OD after severe trauma during the first two weeks post-trauma. Our findings underline the heterogeneous course after trauma and describe some potentially important clinical insights that are suggested by the groupings and temporal trajectories.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13054-021-03586-6.

Critical carE Database for Advanced Research (CEDAR): An automated method to support intensive care units with electronic health record data

Patients treated in an intensive care unit (ICU) are critically ill and require life-sustaining organ failure support. Existing critical care data resources are limited to a select number of institutions, contain only ICU data, and do not enable the study of local changes in care patterns. To address these limitations, we developed the Critical carE Database for Advanced Research (CEDAR), a method for automating extraction and transformation of data from an electronic health record (EHR) system. Compared to an existing gold standard of manually collected data at our institution, CEDAR was statistically similar in most measures, including patient demographics and sepsis-related organ failure assessment (SOFA) scores. Additionally, CEDAR automated data extraction obviated the need for manual collection of 550 variables. Critically, during the spring 2020 COVID-19 surge in New York City, a modified version of CEDAR supported pandemic response efforts, including clinical operations and research. Other academic medical centers may find value in using the CEDAR method to automate data extraction from EHR systems to support ICU activities.

Enhancement in Performance of Septic Shock Prediction Using National Early Warning Score, Initial Triage Information, and Machine Learning Analysis

BACKGROUND

Several studies reported that the National Early Warning Score (NEWS) has shown superiority over other screening tools in discriminating emergency department (ED) patients who are likely to progress to septic shock.

OBJECTIVES

To improve the performance of the NEWS for septic shock prediction by adding variables collected during ED triage, and to implement a machine-learning algorithm.

METHODS

The study population comprised adult ED patients with suspected infection. To detect septic shock within 24 h after ED arrival, the Sepsis-3 clinical criteria and nine variables were used: NEWS, age, gender, systolic blood pressure, diastolic blood pressure, pulse rate, respiratory rate, body temperature, and oxygen saturation. The model was developed using logistic regression (LR), extreme gradient boosting (XGB), and artificial neural network (ANN) algorithms. The evaluations were performed using an area under the receiver operating characteristic curve (AUROC), Hosmer-Lemeshow test, and net reclassification index (NRI).

RESULTS

Overall, 41,687 patients were enrolled. The AUROC of the model with NEWS, age, gender, and the six vital signs (0.835-0.845) was better than that of the baseline model (0.804). The XGB model (AUROC 0.845) was the most accurate, compared with LR (0.844) and ANN (0.835). The LR and XGB models were well calibrated; however, the ANN showed poor calibration power. The LR and XGB models showed better reclassification than the baseline model with positive NRI.

CONCLUSION

The discrimination power of the model for screening septic shock using NEWS, age, gender, and the six vital signs collected at ED triage outperformed the baseline NEWS model.

Prediction of Neurologically Intact Survival in Cardiac Arrest Patients without Pre-Hospital Return of Spontaneous Circulation: Machine Learning Approach

Current multimodal approaches for the prognostication of out-of-hospital cardiac arrest (OHCA) are based mainly on the prediction of poor neurological outcomes; however, it is challenging to identify patients expected to have a favorable outcome, especially before the return of spontaneous circulation (ROSC). We developed and validated a machine learning-based system to predict good outcome in OHCA patients before ROSC. This prospective, multicenter, registry-based study analyzed non-traumatic OHCA data collected between October 2015 and June 2017. We used information available before ROSC as predictor variables, and the primary outcome was neurologically intact survival at discharge, defined as cerebral performance category 1 or 2. The developed models’ robustness were evaluated and compared with various score metrics to confirm their performance. The model using a voting classifier had the best performance in predicting good neurological outcome (area under the curve = 0.926). We confirmed that the six top-weighted variables predicting neurological outcomes, such as several duration variables after the instant of OHCA and several electrocardiogram variables in the voting classifier model, showed significant differences between the two neurological outcome groups. These findings demonstrate the potential utility of a machine learning model to predict good neurological outcome of OHCA patients before ROSC.

Application of Machine Learning in Intensive Care Unit (ICU) Settings Using MIMIC Dataset: Systematic Review

Modern Intensive Care Units (ICUs) provide continuous monitoring of critically ill patients susceptible to many complications affecting morbidity and mortality. ICU settings require a high staff-to-patient ratio and generates a sheer volume of data. For clinicians, the real-time interpretation of data and decision-making is a challenging task. Machine Learning (ML) techniques in ICUs are making headway in the early detection of high-risk events due to increased processing power and freely available datasets such as the Medical Information Mart for Intensive Care (MIMIC). We conducted a systematic literature review to evaluate the effectiveness of applying ML in the ICU settings using the MIMIC dataset. A total of 322 articles were reviewed and a quantitative descriptive analysis was performed on 61 qualified articles that applied ML techniques in ICU settings using MIMIC data. We assembled the qualified articles to provide insights into the areas of application, clinical variables used, and treatment outcomes that can pave the way for further adoption of this promising technology and possible use in routine clinical decision-making. The lessons learned from our review can provide guidance to researchers on application of ML techniques to increase their rate of adoption in healthcare.

Assessment of the timeliness and robustness for predicting adult sepsis

Sepsis is a leading cause of death among inpatients at hospitals. However, with early detection, death rate can drop substantially. In this study, we present the top-performing algorithm for Sepsis II prediction in the DII National Data Science Challenge using the Cerner Health Facts data involving more than 100,000 adult patients. This large sample size allowed us to dissect the predictability by age-groups, race, genders, and care settings and up to 192 hr of sepsis onset. This large data collection also allowed us to conclude that the last six biometric records on average are informative to the prediction of sepsis. We identified biomarkers that are common across the treatment time and novel biomarkers that are uniquely presented for early prediction. The algorithms showed meaningful signals days ahead of sepsis onset, supporting the potential of reducing death rate by focusing on high-risk populations identified from heterogeneous data integration.

Systemic Inflammatory Response Syndrome After Surgery: Mechanisms and Protection

The immune system is an evolutionary hallmark of higher organisms that defends the host against invading pathogens and exogenous infections. This defense includes the recruitment of immune cells to the site of infection and the initiation of an inflammatory response to contain and eliminate pathogens. However, an inflammatory response may also be triggered by noninfectious stimuli such as major surgery, and, in case of an overshooting, still not comprehensively understood reaction, lead to tissue destruction and organ dysfunction. Unfortunately, in some cases, the immune system may not effectively distinguish between stimuli elicited by major surgery, which ideally should only require a modest inflammatory response, and those elicited by trauma or pathogenic infection. Surgical procedures thus represent a potential trigger for systemic inflammation that causes the secretion of proinflammatory cytokines, endothelial dysfunction, glycocalyx damage, activation of neutrophils, and ultimately tissue and multisystem organ destruction. In this review, we discuss and summarize currently available mechanistic knowledge on surgery-associated systemic inflammation, demarcation toward other inflammatory complications, and possible therapeutic options. These options depend on uncovering the underlying mechanisms and could include pharmacologic agents, remote ischemic preconditioning protocols, cytokine blockade or clearance, and optimization of surgical procedures, anesthetic regimens, and perioperative inflammatory diagnostic assessment. Currently, a large gap between basic science and clinically confirmed data exists due to a limited evidence base of translational studies. We thus summarize important steps toward the understanding of the precise time- and space-regulated processes in systemic perioperative inflammation.

Predicting 30-days mortality for MIMIC-III patients with sepsis-3: a machine learning approach using XGboost

Background

Sepsis is a significant cause of mortality in-hospital, especially in ICU patients. Early prediction of sepsis is essential, as prompt and appropriate treatment can improve survival outcomes. Machine learning methods are flexible prediction algorithms with potential advantages over conventional regression and scoring system. The aims of this study were to develop a machine learning approach using XGboost to predict the 30-days mortality for MIMIC-III Patients with sepsis-3 and to determine whether such model performs better than traditional prediction models.

Methods

Using the MIMIC-III v1.4, we identified patients with sepsis-3. The data was split into two groups based on death or survival within 30 days and variables, selected based on clinical significance and availability by stepwise analysis, were displayed and compared between groups. Three predictive models including conventional logistic regression model, SAPS-II score prediction model and XGBoost algorithm model were constructed by R software. Then, the performances of the three models were tested and compared by AUCs of the receiver operating characteristic curves and decision curve analysis. At last, nomogram and clinical impact curve were used to validate the model.

Results

A total of 4559 sepsis-3 patients are included in the study, in which, 889 patients were death and 3670 survival within 30 days, respectively. According to the results of AUCs (0.819 [95% CI 0.800–0.838], 0.797 [95% CI 0.781–0.813] and 0.857 [95% CI 0.839–0.876]) and decision curve analysis for the three models, the XGboost model performs best. The risk nomogram and clinical impact curve verify that the XGboost model possesses significant predictive value.

Conclusions

Using machine learning technique by XGboost, more significant prediction model can be built. This XGboost model may prove clinically useful and assist clinicians in tailoring precise management and therapy for the patients with sepsis-3.

Survival prediction of patients with sepsis from age, sex, and septic episode number alone

Sepsis is a life-threatening condition caused by an exaggerated reaction of the body to an infection, that leads to organ failure or even death. Since sepsis can kill a patient even in just one hour, survival prediction is an urgent priority among the medical community: even if laboratory tests and hospital analyses can provide insightful information about the patient, in fact, they might not come in time to allow medical doctors to recognize an immediate death risk and treat it properly. In this context, machine learning can be useful to predict survival of patients within minutes, especially when applied to few medical features easily retrievable. In this study, we show that it is possible to achieve this goal by applying computational intelligence algorithms to three features of patients with sepsis, recorded at hospital admission: sex, age, and septic episode number. We applied several data mining methods to a cohort of 110,204 admissions of patients, and obtained high prediction scores both on this complete dataset (top precision-recall area under the curve PR AUC = 0.966) and on its subset related to the recent Sepsis-3 definition (top PR AUC = 0.860). Additionally, we tested our models on an external validation cohort of 137 patients, and achieved good results in this case too (top PR AUC = 0.863), confirming the generalizability of our approach. Our results can have a huge impact on clinical settings, allowing physicians to forecast the survival of patients by sex, age, and septic episode number alone.

Supervised Machine Learning Applied to Automate Flash and Prolonged Capillary Refill Detection by Pulse Oximetry

Objective

Develop an automated approach to detect flash (<1.0 s) or prolonged (>2.0 s) capillary refill time (CRT) that correlates with clinician judgment by applying several supervised machine learning (ML) techniques to pulse oximeter plethysmography data.

Materials and Methods

Data was collected in the Pediatric Intensive Care Unit (ICU), Cardiac ICU, Progressive Care Unit, and Operating Suites in a large academic children’s hospital. Ninety-nine children and 30 adults were enrolled in testing and validation cohorts, respectively. Patients had 5 paired CRT measurements by a modified pulse oximeter device and a clinician, generating 485 waveform pairs for model training. Supervised ML models using gradient boosting (XGBoost), logistic regression (LR), and support vector machines (SVMs) were developed to detect flash (<1 s) or prolonged CRT (≥2 s) using clinician CRT assessment as the reference standard. Models were compared using Area Under the Receiver Operating Curve (AUC) and precision-recall curve (positive predictive value vs. sensitivity) analysis. The best performing model was externally validated with 90 measurement pairs from adult patients. Feature importance analysis was performed to identify key waveform characteristics.

Results

For flash CRT, XGBoost had a greater mean AUC (0.79, 95% CI 0.75–0.83) than logistic regression (0.77, 0.71–0.82) and SVM (0.72, 0.67–0.76) models. For prolonged CRT, XGBoost had a greater mean AUC (0.77, 0.72–0.82) than logistic regression (0.73, 0.68–0.78) and SVM (0.75, 0.70–0.79) models. Pairwise testing showed statistically significant improved performance comparing XGBoost and SVM; all other pairwise model comparisons did not reach statistical significance. XGBoost showed good external validation with AUC of 0.88. Feature importance analysis of XGBoost identified distinct key waveform characteristics for flash and prolonged CRT, respectively.

Conclusion

Novel application of supervised ML to pulse oximeter waveforms yielded multiple effective models to identify flash and prolonged CRT, using clinician judgment as the reference standard.

Tweet

Supervised machine learning applied to pulse oximeter waveform features predicts flash or prolonged capillary refill.

Spectral clustering of risk score trajectories stratifies sepsis patients by clinical outcome and interventions received

Sepsis is not a monolithic disease, but a loose collection of symptoms with diverse outcomes. Thus, stratification and subtyping of sepsis patients is of great importance. We examine the temporal evolution of patient state using our previously-published method for computing risk of transition from sepsis into septic shock. Risk trajectories diverge into four clusters following early prediction of septic shock, stratifying by outcome: the highest-risk and lowest-risk groups have a 76.5% and 10.4% prevalence of septic shock, and 43% and 18% mortality, respectively. These clusters differ also in treatments received and median time to shock onset. Analyses reveal the existence of a rapid (30–60 min) transition in risk at the time of threshold crossing. We hypothesize that this transition occurs as a result of the failure of compensatory biological systems to cope with infection, resulting in a bifurcation of low to high risk. Such a collapse, we believe, represents the true onset of septic shock. Thus, this rapid elevation in risk represents a potential new data-driven definition of septic shock.

Detection of Bacteremia in Surgical In-Patients Using Recurrent Neural Network Based on Time Series Records: Development and Validation Study

Background

Detecting bacteremia among surgical in-patients is more obscure than other patients due to the inflammatory condition caused by the surgery. The previous criteria such as systemic inflammatory response syndrome or Sepsis-3 are not available for use in general wards, and thus, many clinicians usually rely on practical senses to diagnose postoperative infection.

Objective

This study aims to evaluate the performance of continuous monitoring with a deep learning model for early detection of bacteremia for surgical in-patients in the general ward and the intensive care unit (ICU).

Methods

In this retrospective cohort study, we included 36,023 consecutive patients who underwent general surgery between October and December 2017 at a tertiary referral hospital in South Korea. The primary outcome was the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC) for detecting bacteremia by the deep learning model, and the secondary outcome was the feature explainability of the model by occlusion analysis.

Results

Out of the 36,023 patients in the data set, 720 cases of bacteremia were included. Our deep learning–based model showed an AUROC of 0.97 (95% CI 0.974-0.981) and an AUPRC of 0.17 (95% CI 0.147-0.203) for detecting bacteremia in surgical in-patients. For predicting bacteremia within the previous 24-hour period, the AUROC and AUPRC values were 0.93 and 0.15, respectively. Occlusion analysis showed that vital signs and laboratory measurements (eg, kidney function test and white blood cell group) were the most important variables for detecting bacteremia.

Conclusions

A deep learning model based on time series electronic health records data had a high detective ability for bacteremia for surgical in-patients in the general ward and the ICU. The model may be able to assist clinicians in evaluating infection among in-patients, ordering blood cultures, and prescribing antibiotics with real-time monitoring.

Paediatric sepsis

PURPOSE OF REVIEW

Sepsis remains among the leading causes of childhood mortality worldwide. This review serves to highlight key areas of knowledge gain and ongoing controversies pertinent to sepsis in children.

RECENT FINDINGS

Several recent publications describe the epidemiology of paediatric sepsis, demonstrating the impact on child health in terms of mortality and morbidity, and the shortcomings of current paediatric sepsis definitions. Although emerging data support the importance of organ dysfunction as a hallmark of paediatric sepsis, the understanding of host susceptibility to sepsis and to sepsis severity remains very limited. Next-generation sequencing and host transcriptomics have the potential to provide new insights into the pathogenesis of sepsis and may enable personalized medicine approaches. Despite good observational data indicating benefit of sepsis recognition and treatment bundles, the evidence for the individual bundle components remains scarce, implying an urgent need for large trials.

SUMMARY

Recent studies have demonstrated distinct epidemiological patterns pertinent to age groups, healthcare settings, and comorbidities in the era post meningococcal epidemics. Although sepsis quality improvement initiatives have led to substantial outcome improvements, there is urgency for innovative trials to reduce uncertainty around the optimal approach for the recognition and treatment of sepsis in children.

How good is our diagnostic intuition? Clinician prediction of bacteremia in critically ill children

Background

Clinical intuition and nonanalytic reasoning play a major role in clinical hypothesis generation; however, clinicians’ intuition about whether a critically ill child is bacteremic has not been explored. We endeavored to assess pediatric critical care clinicians’ ability to predict bacteremia and to evaluate what affected the accuracy of those predictions.

Methods

We conducted a retrospective review of clinicians’ responses to a sepsis screening tool (“Early Sepsis Detection Tool” or “ESDT”) over 6 months. The ESDT was completed during the initial evaluation of a possible sepsis episode. If a culture was ordered, they were asked to predict if the culture would be positive or negative. Culture results were compared to predictions for each episode as well as vital signs and laboratory data from the preceding 24 h.

Results

From January to July 2017, 266 ESDTs were completed. Of the 135 blood culture episodes, 15% of cultures were positive. Clinicians correctly predicted patients with bacteremia in 82% of cases, but the positive predictive value was just 28% as there was a tendency to overestimate the presence of bacteremia. The negative predictive value was 96%. The presence of bandemia, thrombocytopenia, and abnormal CRP were associated with increased likelihood of correct positive prediction.

Conclusions

Clinicians are accurate in predicting critically ill children whose blood cultures, obtained for symptoms of sepsis, will be negative. Clinicians frequently overestimate the presence of bacteremia. The combination of evidence-based practice guidelines and bedside judgment should be leveraged to optimize diagnosis of bacteremia.

Natural Language Processing of Clinical Notes for Improved Early Prediction of Septic Shock in the ICU

Sepsis and septic shock are major concerns in public health as the leading contributors to hospital mortality and cost of treatment in the United States. Early treatment is instrumental for improving patient outcome; to this end, algorithmic methods for early prediction of septic shock have been developed using electronic health record data, with the goal of decreasing treatment delay. We extend a previously-developed method, using a gradient boosting algorithm (XG-Boost) to compute a time-evolving risk of impending transition into septic shock, by combining physiological data from the electronic health record with features obtained from natural language processing of clinical note data. We compare two different methods for generating natural language processing features, with the best method obtaining improved performance of 0.92 AUC, 84% sensitivity, 82% specificity, 49% positive predictive value, and a median early warning time of 7.0 hours. This degree of early warning is sufficient to enable intervention many hours in advance of septic shock onset, with the improved prediction performance of this method resulting in fewer false alarms and thus more actionable predictions.

Enhancing sepsis management through machine learning techniques: A review

Sepsis is a major public health problem and a leading cause of death in the world, where delay in the beginning of treatment, along with clinical guidelines non-adherence have been proved to be associated with higher mortality. Machine Learning is increasingly being adopted in developing innovative Clinical Decision Support Systems in many areas of medicine, showing a great potential for automatic prediction of diverse patient conditions, as well as assistance in clinical decision making. In this context, this work conducts a narrative review to provide an overview of how specific Machine Learning techniques can be used to improve sepsis management, discussing the main tasks addressed, the most popular methods and techniques, as well as the obtained results, in terms of both intelligent system accuracy and clinical outcomes improvement.

Collaborative Antimicrobial Stewardship: Working with Information Technology

Information technology (IT) is vitally important to making antimicrobial stewardship a scalable endeavor in modern health care systems. Without IT, many antimicrobial interventions in patient care would be missed. Clinical decision support systems and smartphone apps, either stand-alone or integrated into electronic health records, can all be effective tools to help augment the work of antimicrobial stewardship programs and support the management of infectious diseases in any health care setting.

Ethical considerations about artificial intelligence for prognostication in intensive care

Background

Prognosticating the course of diseases to inform decision-making is a key component of intensive care medicine. For several applications in medicine, new methods from the field of artificial intelligence (AI) and machine learning have already outperformed conventional prediction models. Due to their technical characteristics, these methods will present new ethical challenges to the intensivist.

Results

In addition to the standards of data stewardship in medicine, the selection of datasets and algorithms to create AI prognostication models must involve extensive scrutiny to avoid biases and, consequently, injustice against individuals or groups of patients. Assessment of these models for compliance with the ethical principles of beneficence and non-maleficence should also include quantification of predictive uncertainty. Respect for patients’ autonomy during decision-making requires transparency of the data processing by AI models to explain the predictions derived from these models. Moreover, a system of continuous oversight can help to maintain public trust in this technology. Based on these considerations as well as recent guidelines, we propose a pathway to an ethical implementation of AI-based prognostication. It includes a checklist for new AI models that deals with medical and technical topics as well as patient- and system-centered issues.

Conclusion

AI models for prognostication will become valuable tools in intensive care. However, they require technical refinement and a careful implementation according to the standards of medical ethics.

A Data-Driven Approach to Predicting Septic Shock in the Intensive Care Unit

Early diagnosis of sepsis and septic shock has been unambiguously linked to lower mortality and better patient outcomes. Despite this, there is a strong unmet need for a reliable clinical tool that can be used for large-scale automated screening to identify high-risk patients. We addressed the following questions: Can a novel algorithm to identify patients at high risk of septic shock 24 hours before diagnosis be discovered using available clinical data? What are performance characteristics of this predictive algorithm? Can current metrics for evaluation of sepsis be improved using novel algorithm? Publicly available data from the intensive care unit setting was used to build septic shock and control patient cohorts. Using Bayesian networks, causal relationships between diagnosis groups, procedure groups, laboratory results, and demographic data were inferred. Predictive model for septic shock 24 hours prior to digital diagnosis was built based on inferred causal networks. Sepsis risk scores were augmented by de novo inferred model and performance was evaluated. A novel predictive model to identify high-risk patients 24 hours ahead of time, with area under curve of 0.81, negative predictive value of 0.87, and a positive predictive value as high as 0.65 was built. The specificity of quick sequential organ failure assessment, systemic inflammatory response syndrome, and modified early warning score was improved when augmented with the novel model, whereas no improvements were made to the sequential organ failure assessment score. We used a data-driven, expert knowledge agnostic method to build a screening algorithm for early detection of septic shock. The model demonstrates strong performance in the data set used and provides a basis for expanding this work toward building an algorithm that is used to screen patients based on electronic medical record data in real time.

LiSep LSTM: A Machine Learning Algorithm for Early Detection of Septic Shock

Sepsis is a major health concern with global estimates of 31.5 million cases per year. Case fatality rates are still unacceptably high, and early detection and treatment is vital since it significantly reduces mortality rates for this condition. Appropriately designed automated detection tools have the potential to reduce the morbidity and mortality of sepsis by providing early and accurate identification of patients who are at risk of developing sepsis. In this paper, we present “LiSep LSTM”; a Long Short-Term Memory neural network designed for early identification of septic shock. LSTM networks are typically well-suited for detecting long-term dependencies in time series data. LiSep LSTM was developed using the machine learning framework Keras with a Google TensorFlow back end. The model was trained with data from the Medical Information Mart for Intensive Care database which contains vital signs, laboratory data, and journal entries from approximately 59,000 ICU patients. We show that LiSep LSTM can outperform a less complex model, using the same features and targets, with an AUROC 0.8306 (95% confidence interval: 0.8236, 0.8376) and median offsets between prediction and septic shock onset up to 40 hours (interquartile range, 20 to 135 hours). Moreover, we discuss how our classifier performs at specific offsets before septic shock onset, and compare it with five state-of-the-art machine learning algorithms for early detection of sepsis.