Pediatric Severe Sepsis Prediction Using Machine Learning

The application of metagenomics, radiomics and machine learning for diagnosis of sepsis

Introduction

Sepsis poses a serious threat to individual life and health. Early and accessible diagnosis and targeted treatment are crucial. This study aims to explore the relationship between microbes, metabolic pathways, and blood test indicators in sepsis patients and develop a machine learning model for clinical diagnosis.

Methods

Blood samples from sepsis patients were sequenced. α-diversity and β-diversity analyses were performed to compare the microbial diversity between the sepsis group and the normal group. Correlation analysis was conducted on microbes, metabolic pathways, and blood test indicators. In addition, a model was developed based on medical records and radiomic features using machine learning algorithms.

Results

The results of α-diversity and β-diversity analyses showed that the microbial diversity of sepsis group was significantly higher than that of normal group (p < 0.05). The top 10 microbial abundances in the sepsis and normal groups were Vitis vinifera, Mycobacterium canettii, Solanum pennellii, Ralstonia insidiosa, Ananas comosus, Moraxella osloensis, Escherichia coli, Staphylococcus hominis, Camelina sativa, and Cutibacterium acnes. The enriched metabolic pathways mainly included Protein families: genetic information processing, Translation, Protein families: signaling and cellular processes, and Unclassified: genetic information processing. The correlation analysis revealed a significant positive correlation (p < 0.05) between IL-6 and Membrane transport. Metabolism of other amino acids showed a significant positive correlation (p < 0.05) with Cutibacterium acnes, Ralstonia insidiosa, Moraxella osloensis, and Staphylococcus hominis. Ananas comosus showed a significant positive correlation (p < 0.05) with Poorly characterized and Unclassified: metabolism. Blood test-related indicators showed a significant negative correlation (p < 0.05) with microorganisms. Logistic regression (LR) was used as the optimal model in six machine learning models based on medical records and radiomic features. The nomogram, calibration curves, and AUC values demonstrated that LR performed best for prediction.

Discussion

This study provides insights into the relationship between microbes, metabolic pathways, and blood test indicators in sepsis. The developed machine learning model shows potential for aiding in clinical diagnosis. However, further research is needed to validate and improve the model.

An early sepsis prediction model utilizing machine learning and unbalanced data processing in a clinical context

Background

Early and accurate diagnoses of sepsis patients are essential to reduce the mortality. However, the sepsis is still diagnosed in a traditional way in China despite the increasing number of related studies, which may to some extent lead to delays in the treatment.

Methods

The study included 2,385 patients, including 364 with sepsis, collected from the First Affiliated Hospital of Anhui Medical University and partner hospitals from April to July 2022. External validation was conducted using the MIMIC-III database (over 60,000 patients from 2001 to 2012) and the eICU Collaborative Research Database (139,000 patients from 2014 to 2015). Multiple algorithm models, along with the SHapley Additive exPlanations (SHAP) analysis, are applied to explore the main risk factors for the accurate prediction of the sepsis. Multiple Imputations for filling missing data and the Synthetic Minority Oversampling (SMOTE) balancing method for balancing data are used for the data processing.

Result

Eighteen diagnostic features are used in the predictive model for early sepsis. The Random Forest model has the best performance among all the models, with an Area Under the Curve (AUC) of 87% and an F1-score (F1) of 77%. Moreover, the interpretation from the SHAP analysis is generally consistent with the current clinical situation.

Conclusion

The study revealed the relationship between these 18 clinical features and diagnostic outcomes. The results indicate that patients with laboratory values of Systolic Blood Pressure, Albumin, and Heart Rate exceeding certain thresholds are at a high likelihood of developing sepsis.

Pediatric septic shock estimation using deep learning and electronic medical records

BACKGROUND

Diagnosing pediatric septic shock is difficult due to the complex and often impractical traditional criteria, such as systemic inflammatory response syndrome (SIRS), which result in delays and higher risks. This study aims to develop a deep learning-based model using SIRS data for early diagnosis in pediatric septic shock cases.

METHODS

The study analyzed data from pediatric patients (<18 years old) admitted to a tertiary hospital from January 2010 to July 2023. Vital signs, lab tests, and clinical information were collected. Septic shock cases were identified using SIRS criteria and inotrope use. A deep learning model was trained and evaluated using the area under the receiver operating characteristics curve (AUROC) and area under the precision-recall curve (AUPRC). Variable contributions were analyzed using the Shapley additive explanation value.

RESULTS

The analysis, involving 9,616,115 measurements, identified 34,696 septic shock cases (0.4%). Oxygen supply was crucial for 41.5% of the control group and 20.8% of the septic shock group. The final model showed strong performance, with an AUROC of 0.927 and AUPRC of 0.879. Key influencers were age, oxygen supply, sex, and partial pressure of carbon dioxide, while body temperature had minimal impact on estimation.

CONCLUSIONS

The proposed deep learning model simplifies early septic shock diagnosis in pediatric patients, reducing the diagnostic workload. Its high accuracy allows timely treatment, but external validation through prospective studies is needed.

Artificial Intelligence in Pediatric Emergency Medicine: Applications, Challenges, and Future Perspectives

The dawn of Artificial intelligence (AI) in healthcare stands as a milestone in medical innovation. Different medical fields are heavily involved, and pediatric emergency medicine is no exception. We conducted a narrative review structured in two parts. The first part explores the theoretical principles of AI, providing all the necessary background to feel confident with these new state-of-the-art tools. The second part presents an informative analysis of AI models in pediatric emergencies. We examined PubMed and Cochrane Library from inception up to April 2024. Key applications include triage optimization, predictive models for traumatic brain injury assessment, and computerized sepsis prediction systems. In each of these domains, AI models outperformed standard methods. The main barriers to a widespread adoption include technological challenges, but also ethical issues, age-related differences in data interpretation, and the paucity of comprehensive datasets in the pediatric context. Future feasible research directions should address the validation of models through prospective datasets with more numerous sample sizes of patients. Furthermore, our analysis shows that it is essential to tailor AI algorithms to specific medical needs. This requires a close partnership between clinicians and developers. Building a shared knowledge platform is therefore a key step.

Deciphering the molecular classification of pediatric sepsis: integrating WGCNA and machine learning-based classification with immune signatures for the development of an advanced diagnostic model

Introduction: Pediatric sepsis (PS) is a life-threatening infection associated with high mortality rates, necessitating a deeper understanding of its underlying pathological mechanisms. Recently discovered programmed cell death induced by copper has been implicated in various medical conditions, but its potential involvement in PS remains largely unexplored. Methods: We first analyzed the expression patterns of cuproptosis-related genes (CRGs) and assessed the immune landscape of PS using the GSE66099 dataset. Subsequently, PS samples were isolated from the same dataset, and consensus clustering was performed based on differentially expressed CRGs. We applied weighted gene co-expression network analysis to identify hub genes associated with PS and cuproptosis. Results: We observed aberrant expression of 27 CRGs and a specific immune landscape in PS samples. Our findings revealed that patients in the GSE66099 dataset could be categorized into two cuproptosis clusters, each characterized by unique immune landscapes and varying functional classifications or enriched pathways. Among the machine learning approaches, Extreme Gradient Boosting demonstrated optimal performance as a diagnostic model for PS. Discussion: Our study provides valuable insights into the molecular mechanisms underlying PS, highlighting the involvement of cuproptosis-related genes and immune cell infiltration.

Automated machine learning for early prediction of acute kidney injury in acute pancreatitis

BACKGROUND

Acute kidney injury (AKI) represents a frequent and grave complication associated with acute pancreatitis (AP), substantially elevating both mortality rates and the financial burden of hospitalization. The aim of our study is to construct a predictive model utilizing automated machine learning (AutoML) algorithms for the early prediction of AKI in patients with AP.

METHODS

We retrospectively analyzed patients who were diagnosed with AP in our hospital from January 2017 to December 2021. These patients were randomly allocated into a training set and a validation set at a ratio of 7:3. To develop predictive models for each set, we employed the least absolute shrinkage and selection operator (LASSO) algorithm along with AutoML. A nomogram was developed based on multivariate logistic regression analysis outcomes. The model's efficacy was assessed using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Additionally, the performance of the model constructed via AutoML was evaluated using decision curve analysis (DCA), feature importance, SHapley Additive exPlanations (SHAP) plots, and locally interpretable model-agnostic explanations (LIME).

RESULTS

This study incorporated a total of 437 patients who met the inclusion criteria. Out of these, 313 were assigned to the training cohort and 124 to the validation cohort. In the training and validation cohorts, AKI occurred in 68 (21.7%) and 29(23.4%) patients, respectively. Comparative analysis revealed that the AutoML models exhibited enhanced performance over traditional logistic regression (LR). Furthermore, the deep learning (DL) model demonstrated superior predictive accuracy, evidenced by an area under the ROC curve of 0.963 in the training set and 0.830 in the validation set, surpassing other comparative models. The key variables identified as significant in the DL model within the training dataset included creatinine (Cr), urea (Urea), international normalized ratio (INR), etiology, smoking, alanine aminotransferase (ALT), hypertension, prothrombin time (PT), lactate dehydrogenase (LDH), and diabetes.

CONCLUSION

The AutoML model, utilizing DL algorithm, offers considerable clinical significance in the early detection of AKI among patients with AP.

Self-supervised machine learning using adult inpatient data produces effective models for pediatric clinical prediction tasks

OBJECTIVE

Development of electronic health records (EHR)-based machine learning models for pediatric inpatients is challenged by limited training data. Self-supervised learning using adult data may be a promising approach to creating robust pediatric prediction models. The primary objective was to determine whether a self-supervised model trained in adult inpatients was noninferior to logistic regression models trained in pediatric inpatients, for pediatric inpatient clinical prediction tasks.

MATERIALS AND METHODS

This retrospective cohort study used EHR data and included patients with at least one admission to an inpatient unit. One admission per patient was randomly selected. Adult inpatients were 18 years or older while pediatric inpatients were more than 28 days and less than 18 years. Admissions were temporally split into training (January 1, 2008 to December 31, 2019), validation (January 1, 2020 to December 31, 2020), and test (January 1, 2021 to August 1, 2022) sets. Primary comparison was a self-supervised model trained in adult inpatients versus count-based logistic regression models trained in pediatric inpatients. Primary outcome was mean area-under-the-receiver-operating-characteristic-curve (AUROC) for 11 distinct clinical outcomes. Models were evaluated in pediatric inpatients.

RESULTS

When evaluated in pediatric inpatients, mean AUROC of self-supervised model trained in adult inpatients (0.902) was noninferior to count-based logistic regression models trained in pediatric inpatients (0.868) (mean difference = 0.034, 95% CI=0.014-0.057; P < .001 for noninferiority and P = .006 for superiority).

CONCLUSIONS

Self-supervised learning in adult inpatients was noninferior to logistic regression models trained in pediatric inpatients. This finding suggests transferability of self-supervised models trained in adult patients to pediatric patients, without requiring costly model retraining.

Prediction of the development of contrast‑induced nephropathy following percutaneous coronary artery intervention by machine learning

Contrast-induced nephropathy (CIN) is associated with increased mortality and morbidity in patients with coronary artery disease undergoing elective percutaneous coronary intervention(PCI). We developed a machine learning-based risk stratification model to predict contrast-induced nephropathy after PCI. A study retrospectively enrolling 240 patients eligible for PCI from December 2017 to May 2020 was performed. CIN was defined as a rise in serum creatinine levels ≥0.5 mg/dL or ≥25% from baseline within 72 h after surgery. Eight machine learning methods were performed based on clinical variables. Shapley Additive exPlanation values were also used to interpret the best-performing prediction models. Development of CIN was found in 37 patients(16.5%) after PCI. There were 11 significant predictors of CIN, including uric acid, peripheral vascular disease, cystatin C, creatine kinase-MB, haemoglobin, N-terminal pro-brain natriuretic peptide, age, diabetes, systemic immune-inflammatory index, total protein, and low-density lipoprotein. Regarding the efficacy of the machine learning model that accurately predicted CIN, SVM exhibited the most outstanding AUC value of 0.784. The SHAP and radar plots were used to illustrate the positive and negative effects of the 11 features attributed to the SVM. Machine learning models have the potential to identify the risk of CIN for elective PCI patients.

Elaborating the potential of Artificial Intelligence in automated CAR-T cell manufacturing

This paper discusses the challenges of producing CAR-T cells for cancer treatment and the potential for Artificial Intelligence (AI) for its improvement. CAR-T cell therapy was approved in 2018 as the first Advanced Therapy Medicinal Product (ATMP) for treating acute leukemia and lymphoma. ATMPs are cell- and gene-based therapies that show great promise for treating various cancers and hereditary diseases. While some new ATMPs have been approved, ongoing clinical trials are expected to lead to the approval of many more. However, the production of CAR-T cells presents a significant challenge due to the high costs associated with the manufacturing process, making the therapy very expensive (approx. $400,000). Furthermore, autologous CAR-T therapy is limited to a make-to-order approach, which makes scaling economical production difficult. First attempts are being made to automate this multi-step manufacturing process, which will not only directly reduce the high manufacturing costs but will also enable comprehensive data collection. AI technologies have the ability to analyze this data and convert it into knowledge and insights. In order to exploit these opportunities, this paper analyses the data potential in the automated CAR-T production process and creates a mapping to the capabilities of AI applications. The paper explores the possible use of AI in analyzing the data generated during the automated process and its capabilities to further improve the efficiency and cost-effectiveness of CAR-T cell production.

Using machine learning for process improvement in sepsis management

INTRODUCTION

In the U.S., sepsis afflicts 1.7 million adults, causing 270,000 deaths each year. Early detection of sepsis could decrease the number of deaths by 92,000 annually and decrease hospital expenditures by 1.5 billion USD. Few prior studies and reviews have presented a holistic understanding of the relationship between machine learning and existing process improvement measures. This study, in addition to discussing machine learning and existing process improvements measures, elaborates on the disadvantages and the barriers to integrating machine learning into the clinic. This article synthesizes previous studies to educate healthcare professionals on effectively managing sepsis by leveraging the benefits of machine learning.

METHODS

This study used the PubMed database. Search terms include sepsis antibiotics, sepsis process improvement, sepsis machine learning. Our search criteria included previous studies published between January 1, 2017, and February 1, 2022.

RESULTS/DISCUSSION

Although machine learning algorithms have better predictive capabilities, their effectiveness in the clinical setting is limited as studies show mixed results because the medical staff often fails to intervene. To overcome poor interventional response, clinicians need to work with the facility's IT department to ensure integration into clinical workflow and minimize alert-fatigue. Algorithms should enhance the productivity of clinical teams, not attempt to replace them entirely.

CONCLUSION

Hospitals can employ process improvement measures that effectively utilize machine learning algorithms to ensure integration into clinical workflows. Healthcare professionals can utilize workflow tools in addition to the predictive capabilities of machine learning to enhance clinical decisions in sepsis.

Distribution, Consequences, and Determinants of Time to Antibiotics in Children With Community-Onset Severe Bacterial Infection: A Secondary Analysis of a Prospective Population-Based Study

OBJECTIVES

To describe the distribution, consequences and potential determinants of time to antibiotics administration in children with community-onset severe bacterial infections (COSBIs).

DESIGN

Secondary analysis of the available data from a prospective population-based study from 2009 to 2014.

SETTING

An administrative area in western France accounting for 13% of the national pediatric population.

PATIENTS

All children from 1 month to 16 years old admitted to a PICU or who died before admission and had a COSBI.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The time to antibiotics was divided into patient interval (from first signs of COSBI to the first medical consultation) and medical interval (from the first consultation to appropriate antibiotics administration). The association between the medical interval and child outcome was studied by a multinomial logistic regression model and the potential determinants of the patient and medical intervals were by a Cox proportional-hazards model. Of the 227 children included (median age 2.1 yr), 22 died (9.7%), and 21 (9.3%) had severe sequelae at PICU discharge. Median patient and medical intervals were 7.0 hours (interquartile range [IQR], 2.0-16.5 hr) and 3.3 hours (IQR, 1.1-12.2 hr), respectively. The last quartile of medical interval was not associated with death (adjusted odds ratio [aOR], 3.7; 95% CI, 0.8-17.5) or survival with severe sequelae (aOR, 1.3; 95% CI, 0.4-4.0) versus survival without severe sequelae. Patient interval was shorter in younger children (adjusted hazard ratio [aHR], 0.95; 95% CI, 0.92-0.99), and medical interval was reduced when the first consultation was conducted in a hospital (aHR, 1.5; 95% CI, 1.1-2.0) versus outpatient medicine.

CONCLUSIONS

For children with COSBI, we found no significant association between medical interval and mortality or severe sequelae. An initial hospital referral could help reduce the time to antibiotics in COSBIs.

Integrating structured and unstructured data for timely prediction of bloodstream infection among children

BACKGROUND

Hospitalized children with central venous lines (CVLs) are at higher risk of hospital-acquired infections. Information in electronic health records (EHRs) can be employed in training deep learning models to predict the onset of these infections. We incorporated clinical notes in addition to structured EHR data to predict serious bloodstream infections, defined as positive blood culture followed by at least 4 days of new antimicrobial agent administration, among hospitalized children with CVLs.

METHODS

Structured EHR information and clinical notes were extracted for a retrospective cohort including all hospitalized patients with CVLs at a single tertiary care pediatric health system from 2013 to 2018. Deep learning models were trained to determine the added benefit of incorporating the information embedded in clinical notes in predicting serious bloodstream infection.

RESULTS

A total of 24,351 patient encounters met inclusion criteria. The best-performing model restricted to structured EHR data had a specificity of 0.951 and positive predictive value (PPV) of 0.056 when the sensitivity was set to 0.85. The addition of contextualized word embeddings improved the specificity to 0.981 and PPV to 0.113.

CONCLUSIONS

Integrating clinical notes with structured EHR data improved the prediction of serious bloodstream infections among pediatric patients with CVLs.

IMPACT

Developed an advanced infection prediction model in pediatrics that integrates the structured and unstructured EHRs. Extracted information from clinical notes to do timely prediction in a clinical setting. Developed a deep learning model framework that can be employed in predicting rare events in a complex and dynamic environment.

An Introduction to Artificial Intelligence in Developmental and Behavioral Pediatrics

ABSTRACT

Technological breakthroughs, together with the rapid growth of medical information and improved data connectivity, are creating dramatic shifts in the health care landscape, including the field of developmental and behavioral pediatrics. While medical information took an estimated 50 years to double in 1950, by 2020, it was projected to double every 73 days. Artificial intelligence (AI)-powered health technologies, once considered theoretical or research-exclusive concepts, are increasingly being granted regulatory approval and integrated into clinical care. In the United States, the Food and Drug Administration has cleared or approved over 160 health-related AI-based devices to date. These trends are only likely to accelerate as economic investment in AI health care outstrips investment in other sectors. The exponential increase in peer-reviewed AI-focused health care publications year over year highlights the speed of growth in this sector. As health care moves toward an era of intelligent technology powered by rich medical information, pediatricians will increasingly be asked to engage with tools and systems underpinned by AI. However, medical students and practicing clinicians receive insufficient training and lack preparedness for transitioning into a more AI-informed future. This article provides a brief primer on AI in health care. Underlying AI principles and key performance metrics are described, and the clinical potential of AI-driven technology together with potential pitfalls is explored within the developmental and behavioral pediatric health context.

Data harnessing to nurture the human mind for a tailored approach to the child

Big data in pediatrics is an ocean of structured and unstructured data. Big data analysis helps to dive into the ocean of data to filter out information that can guide pediatricians in their decision making, precision diagnosis, and targeted therapy. In addition, big data and its analysis have helped in the surveillance, prevention, and performance of the health system. There has been a considerable amount of work in pediatrics that we have tried to highlight in this review and some of it has been already incorporated into the health system. Work in specialties of pediatrics is still forthcoming with the creation of a common data model and amalgamation of the huge "omics" database. The physicians entrusted with the care of children must be aware of the outcome so that they can play a role to ensure that big data algorithms have a clinically relevant effect in improving the health of their patients. They will apply the outcome of big data and its analysis in patient care through clinical algorithms or with the help of embedded clinical support alerts from the electronic medical records. IMPACT: Big data in pediatrics include structured, unstructured data, waveform data, biological, and social data. Big data analytics has unraveled significant information from these databases. This is changing how pediatricians will look at the body of available evidence and translate it into their clinical practice. Data harnessed so far is implemented in certain fields while in others it is in the process of development to become a clinical adjunct to the physician. Common databases are being prepared for future work. Diagnostic and prediction models when incorporated into the health system will guide the pediatrician to a targeted approach to diagnosis and therapy.

Artificial intelligence-based clinical decision support in pediatrics

Machine learning models may be integrated into clinical decision support (CDS) systems to identify children at risk of specific diagnoses or clinical deterioration to provide evidence-based recommendations. This use of artificial intelligence models in clinical decision support (AI-CDS) may have several advantages over traditional "rule-based" CDS models in pediatric care through increased model accuracy, with fewer false alerts and missed patients. AI-CDS tools must be appropriately developed, provide insight into the rationale behind decisions, be seamlessly integrated into care pathways, be intuitive to use, answer clinically relevant questions, respect the content expertise of the healthcare provider, and be scientifically sound. While numerous machine learning models have been reported in pediatric care, their integration into AI-CDS remains incompletely realized to date. Important challenges in the application of AI models in pediatric care include the relatively lower rates of clinically significant outcomes compared to adults, and the lack of sufficiently large datasets available necessary for the development of machine learning models. In this review article, we summarize key concepts related to AI-CDS, its current application to pediatric care, and its potential benefits and risks. IMPACT: The performance of clinical decision support may be enhanced by the utilization of machine learning-based algorithms to improve the predictive performance of underlying models. Artificial intelligence-based clinical decision support (AI-CDS) uses models that are experientially improved through training and are particularly well suited toward high-dimensional data. The application of AI-CDS toward pediatric care remains limited currently but represents an important area of future research.

State-of-the-art management of the acutely unwell child

Children make up around one-fifth of all emergency department visits in the USA and UK, with an increasing trend of emergency admissions requiring intensive care. Anaesthetists play a vital role in the management of paediatric emergencies contributing to stabilisation, emergency anaesthesia, transfers and non-technical skills that optimise team performance. From neonates to adolescents, paediatric patients have diverse physiology and present with a range of congenital and acquired pathologies that often differ from the adult population. With increasing centralisation of paediatric services, staff outside these centres have less exposure to caring for children, yet are often the first responders in managing these high stakes situations. Staying abreast of the latest evidence for managing complex low frequency emergencies is a challenge. This review focuses on recent evidence and pertinent clinical updates within the field. The challenges of maintaining skills and training are explored as well as novel advancements in care.

On AI Approaches for Promoting Maternal and Neonatal Health in Low Resource Settings: A Review

A significant challenge for hospitals and medical practitioners in low- and middle-income nations is the lack of sufficient health care facilities for timely medical diagnosis of chronic and deadly diseases. Particularly, maternal and neonatal morbidity due to various non-communicable and nutrition related diseases is a serious public health issue that leads to several deaths every year. These diseases affecting either mother or child can be hospital-acquired, contracted during pregnancy or delivery, postpartum and even during child growth and development. Many of these conditions are challenging to detect at their early stages, which puts the patient at risk of developing severe conditions over time. Therefore, there is a need for early screening, detection and diagnosis, which could reduce maternal and neonatal mortality. With the advent of Artificial Intelligence (AI), digital technologies have emerged as practical assistive tools in different healthcare sectors but are still in their nascent stages when applied to maternal and neonatal health. This review article presents an in-depth examination of digital solutions proposed for maternal and neonatal healthcare in low resource settings and discusses the open problems as well as future research directions.

Standard Vocabularies to Improve Machine Learning Model Transferability With Electronic Health Record Data: Retrospective Cohort Study Using Health Care-Associated Infection

BACKGROUND

With the widespread adoption of electronic healthcare records (EHRs) by US hospitals, there is an opportunity to leverage this data for the development of predictive algorithms to improve clinical care. A key barrier in model development and implementation includes the external validation of model discrimination, which is rare and often results in worse performance. One reason why machine learning models are not externally generalizable is data heterogeneity. A potential solution to address the substantial data heterogeneity between health care systems is to use standard vocabularies to map EHR data elements. The advantage of these vocabularies is a hierarchical relationship between elements, which allows the aggregation of specific clinical features to more general grouped concepts.

OBJECTIVE

This study aimed to evaluate grouping EHR data using standard vocabularies to improve the transferability of machine learning models for the detection of postoperative health care-associated infections across institutions with different EHR systems.

METHODS

Patients who underwent surgery from the University of Utah Health and Intermountain Healthcare from July 2014 to August 2017 with complete follow-up data were included. The primary outcome was a health care-associated infection within 30 days of the procedure. EHR data from 0-30 days after the operation were mapped to standard vocabularies and grouped using the hierarchical relationships of the vocabularies. Model performance was measured using the area under the receiver operating characteristic curve (AUC) and F₁-score in internal and external validations. To evaluate model transferability, a difference-in-difference metric was defined as the difference in performance drop between internal and external validations for the baseline and grouped models.

RESULTS

A total of 5775 patients from the University of Utah and 15,434 patients from Intermountain Healthcare were included. The prevalence of selected outcomes was from 4.9% (761/15,434) to 5% (291/5775) for surgical site infections, from 0.8% (44/5775) to 1.1% (171/15,434) for pneumonia, from 2.6% (400/15,434) to 3% (175/5775) for sepsis, and from 0.8% (125/15,434) to 0.9% (50/5775) for urinary tract infections. In all outcomes, the grouping of data using standard vocabularies resulted in a reduced drop in AUC and F₁-score in external validation compared to baseline features (all P<.001, except urinary tract infection AUC: P=.002). The difference-in-difference metrics ranged from 0.005 to 0.248 for AUC and from 0.075 to 0.216 for F₁-score.

CONCLUSIONS

We demonstrated that grouping machine learning model features based on standard vocabularies improved model transferability between data sets across 2 institutions. Improving model transferability using standard vocabularies has the potential to improve the generalization of clinical prediction models across the health care system.

The Risk of Coding Racism into Pediatric Sepsis Care: The Necessity of Antiracism in Machine Learning

Machine learning holds the possibility of improving racial health inequalities by compensating for human bias and structural racism. However, unanticipated racial biases may enter during model design, training, or implementation and perpetuate or worsen racial inequalities if ignored. Pre-existing racial health inequalities could be codified into medical care by machine learning without clinicians being aware. To illustrate the importance of a commitment to antiracism at all stages of machine learning, we examine machine learning in predicting severe sepsis in Black children, focusing on the impacts of structural racism that may be perpetuated by machine learning and difficult to discover. To move toward antiracist machine learning, we recommend partnering with ethicists and experts in model development, enrolling representative samples for training, including socioeconomic inputs with proximate causal associations to racial inequalities, reporting outcomes by race, and committing to equitable models that narrow inequality gaps or at least have equal benefit.

G-Sep: A Deep Learning Algorithm for Detection of Long-Term Sepsis Using Bidirectional Gated Recurrent Unit

Sepsis is a common and deadly condition that must be treated eloquently within 19 hours. Numerous deep learning techniques, including Recurrent Neural Networks, Convolution Neural Networks, Long Short-Term Memory, and Gated Recurrent Units, have been suggested for diagnosing long-term sepsis. Regardless, a sizable portion of them are computationally risky and have precision problems. The primary issue described is that output will degrade, and resource utilization will expand proportionately as the volume of dependencies grows. To overcome these issues, we propose a G-Sep technique utilizing Bidirectional Gated Recurrent Unit Algorithm, which consumes much less resource to detect the disease and in a short time with better accuracy than the existing methods to diagnose the sepsis. AI models could assist with distinguishing potential clinical factors and give better than existing conventional low-execution models. The proposed model is implemented utilizing Conda and Tensorflow Framework using the California Inpatient Severe Sepsis (CISS) Patient Dataset. The comparative simulation of the various existing models and the proposed G-Sep model is done using Conda and Tensor frameworks. The simulation results revealed that the proposed model had outperformed other frameworks in terms of mean average precision (mAP), receiver operating characteristic curve (ROC), and Area under the ROC Curve (AUROC) metrics linearly.

Clinical applications of machine learning in the survival prediction and classification of sepsis: coagulation and heparin usage matter

BACKGROUND

Sepsis is a life-threatening syndrome eliciting highly heterogeneous host responses. Current prognostic evaluation methods used in clinical practice are characterized by an inadequate effectiveness in predicting sepsis mortality. Rapid identification of patients with high mortality risk is urgently needed. The phenotyping of patients will assistant invaluably in tailoring treatments.

METHODS

Machine learning and deep learning technology are used to characterize the patients' phenotype and determine the sepsis severity. The database used in this study is MIMIC-III and MIMIC-IV ('Medical information Mart for intensive care') which is a large, public, and freely available database. The K-means clustering is used to classify the sepsis phenotype. Convolutional neural network (CNN) was used to predict the 28-day survival rate based on 35 blood test variables of the sepsis patients, whereas a double coefficient quadratic multivariate fitting function (DCQMFF) is utilized to predict the 28-day survival rate with only 11 features of sepsis patients.

RESULTS

The patients were grouped into four clusters with a clear survival nomogram. The first cluster (C_1) was characterized by low white blood cell count, low neutrophil, and the highest lymphocyte proportion. C_2 obtained the lowest Sequential Organ Failure Assessment (SOFA) score and the highest survival rate. C_3 was characterized by significantly prolonged PTT, high SIC, and a higher proportion of patients using heparin than the patients in other clusters. The early mortality rate of patients in C_3 was high but with a better long-term survival rate than that in C_4. C_4 contained septic coagulation patients with the worst prognosis, characterized by slightly prolonged partial thromboplastin time (PTT), significantly prolonged prothrombin time (PT), and high septic coagulation disease score (SIC). The survival rate prediction accuracy of CNN and DCQMFF models reached 92% and 82%, respectively. The models were tested on an external dataset (MIMIC-IV) and achieved good performance. A DCQMFF-based application platform was established for fast prediction of the 28-day survival rate.

CONCLUSION

CNN and DCQMFF accurately predicted the sepsis patients' survival, while K-means successfully identified the phenotype groups. The distinct phenotypes associated with survival, and significant features correlated with mortality were identified. The findings suggest that sepsis patients with abnormal coagulation had poor outcomes, abnormal coagulation increase mortality during sepsis. The anticoagulation effects of appropriate heparin sodium treatment may improve extensive micro thrombosis-caused organ failure.

Machine Learning Approach to Predicting Absence of Serious Bacterial Infection at PICU Admission

BACKGROUND AND OBJECTIVES

Serious bacterial infection (SBI) is common in the PICU. Antibiotics can mitigate associated morbidity and mortality but have associated adverse effects. Our objective is to develop machine learning models able to identify SBI-negative children and reduce unnecessary antibiotics.

METHODS

We developed models to predict SBI-negative status at PICU admission using vital sign, laboratory, and demographic variables. Children 3-months to 18-years-old admitted to our PICU, between 2011 and 2020, were included if evaluated for infection within 24-hours, stratified by documented antibiotic exposure in the 48-hours prior. Area under the receiver operating characteristic curve (AUROC) was the primary model accuracy measure; secondarily, we calculated the number of SBI-negative children subsequently provided antibiotics in the PICU identified as low-risk by each model.

RESULTS

A total of 15 074 children met inclusion criteria; 4788 (32%) received antibiotics before PICU admission. Of these antibiotic-exposed patients, 2325 of 4788 (49%) had an SBI. Of the 10 286 antibiotic-unexposed patients, 2356 of 10 286 (23%) had an SBI. In antibiotic-exposed children, a radial support vector machine model had the highest AUROC (0.80) for evaluating SBI, identifying 48 of 442 (11%) SBI-negative children provided antibiotics in the PICU who could have been spared a median 3.7 (interquartile range 0.9-9.0) antibiotic-days per patient. In antibiotic-unexposed children, a random forest model performed best, but was less accurate overall (AUROC 0.76), identifying 33 of 469 (7%) SBI-negative children provided antibiotics in the PICU who could have been spared 1.1 (interquartile range 0.9-3.7) antibiotic-days per patient.

CONCLUSIONS

Among children who received antibiotics before PICU admission, machine learning models can identify children at low risk of SBI and potentially reduce antibiotic exposure.

Development and validation of machine learning-driven prediction model for serious bacterial infection among febrile children in emergency departments

Serious bacterial infection (SBI) in children, such as bacterial meningitis or sepsis, is an important condition that can lead to fatal outcomes. Therefore, since it is very important to accurately diagnose SBI, SBI prediction tools such as 'Refined Lab-score' or 'clinical prediction rule' have been developed and used. However, these tools can predict SBI only when there are values of all factors used in the tool, and if even one of them is missing, the tools become useless. Therefore, the purpose of this study was to develop and validate a machine learning-driven model to predict SBIs among febrile children, even with missing values. This was a multicenter retrospective observational study including febrile children <6 years of age who visited Emergency departments (EDs) of 3 different tertiary hospitals from 2016 to 2018. The SBI prediction model was trained with a derivation cohort (data from two hospitals) and externally tested with a validation cohort (data from a third hospital). A total of 11,973 and 2,858 patient records were included in the derivation and validation cohorts, respectively. In the derivation cohort, the area under the receiver operating characteristic curve (AUROC) of the RF model was 0.964 (95% confidence interval [CI], 0.943-0.986), and the area under the precision-recall curve (AUPRC) was 0.753 (95% CI, 0.681-0.824). The conventional LR (CLR) model showed corresponding values of 0.902 (95% CI, 0.894-0.910) and 0.573 (95% CI, 0.560-0.586), respectively. In the validation cohort, the AUROC (95% CI) of the RF model was 0.950 (95% CI, 0.945-0.956), the AUPRC was 0.605 (95% CI, 0.593-0.616), and the CLR presented corresponding values of 0.815 (95% CI, 0.789-0.841) and 0.586 (95% CI, 0.553-0.619), respectively. We developed a machine learning-driven prediction model for SBI among febrile children, which works robustly despite missing values. And it showed superior performance compared to CLR in both internal validation and external validation.

Considerations for the implementation of machine learning into acute care settings

INTRODUCTION

Management of patients in the acute care setting requires accurate diagnosis and rapid initiation of validated treatments; therefore, this setting is likely to be an environment in which cognitive augmentation of the clinician's provision of care with technology rooted in artificial intelligence, such as machine learning (ML), is likely to eventuate.

SOURCES OF DATA

PubMed and Google Scholar with search terms that included ML, intensive/critical care unit, electronic health records (EHR), anesthesia information management systems and clinical decision support were the primary sources for this report.

AREAS OF AGREEMENT

Different categories of learning of large clinical datasets, often contained in EHRs, are used for training in ML. Supervised learning uses algorithm-based models, including support vector machines, to pair patients' attributes with an expected outcome. Unsupervised learning uses clustering algorithms to define to which disease grouping a patient's attributes most closely approximates. Reinforcement learning algorithms use ongoing environmental feedback to deterministically pursue likely patient outcome.

AREAS OF CONTROVERSY

Application of ML can result in undesirable outcomes over concerns related to fairness, transparency, privacy and accountability. Whether these ML technologies irrevocably change the healthcare workforce remains unresolved.

GROWING POINTS

Well-resourced Learning Health Systems are likely to exploit ML technology to gain the fullest benefits for their patients. How these clinical advantages can be extended to patients in health systems that are neither well-endowed, nor have the necessary data gathering technologies, needs to be urgently addressed to avoid further disparities in healthcare.

Locating Youth Exposed to Parental Justice Involvement in the Electronic Health Record: Development of a Natural Language Processing Model

BACKGROUND

Parental justice involvement (eg, prison, jail, parole, or probation) is an unfortunately common and disruptive household adversity for many US youths, disproportionately affecting families of color and rural families. Data on this adversity has not been captured routinely in pediatric health care settings, and if it is, it is not discrete nor able to be readily analyzed for purposes of research.

OBJECTIVE

In this study, we outline our process training a state-of-the-art natural language processing model using unstructured clinician notes of one large pediatric health system to identify patients who have experienced a justice-involved parent.

METHODS

Using the electronic health record database of a large Midwestern pediatric hospital-based institution from 2011-2019, we located clinician notes (of any type and written by any type of provider) that were likely to contain such evidence of family justice involvement via a justice-keyword search (eg, prison and jail). To train and validate the model, we used a labeled data set of 7500 clinician notes identifying whether the patient was ever exposed to parental justice involvement. We calculated the precision and recall of the model and compared those rates to the keyword search.

RESULTS

The development of the machine learning model increased the precision (positive predictive value) of locating children affected by parental justice involvement in the electronic health record from 61% (a simple keyword search) to 92%.

CONCLUSIONS

The use of machine learning may be a feasible approach to addressing the gaps in our understanding of the health and health services of underrepresented youth who encounter childhood adversities not routinely captured-particularly for children of justice-involved parents.

The inclusion of augmented intelligence in medicine: A framework for successful implementation

Artificial intelligence (AI) algorithms are being applied across a large spectrum of everyday life activities. The implementation of AI algorithms in clinical practice has been met with some skepticism and concern, mainly because of the uneasiness that stems, in part, from a lack of understanding of how AI operates, together with the role of physicians and patients in the decision-making process; uncertainties regarding the reliability of the data and the outcomes; as well as concerns regarding the transparency, accountability, liability, handling of personal data, and monitoring and system upgrades. In this viewpoint, we take these issues into consideration and offer an integrated regulatory framework to AI developers, clinicians, researchers, and regulators, aiming to facilitate the adoption of AI that rests within the FDA’s pathway, in research, development, and clinical medicine.

•

Concerns hamper the implementation of augmented intelligence in clinical practice

•

Transparency and external validation increase trust in developed algorithms

•

A regulatory framework is needed for adoption of augmented intelligence in medicine

Pediatric sepsis screening in US hospitals

Sepsis is a major cause of morbidity and mortality in children. While adverse outcomes can be reduced through prompt initiation of sepsis protocols including fluid resuscitation and antibiotics, provision of these therapies relies on clinician recognition of sepsis. Recognition is challenging in children because early signs of shock such as tachycardia and tachypnea have low specificity while hypotension often does not occur until late in the clinical course. This narrative review highlights the important context that has led to the rapid growth of pediatric sepsis screening in the United States. In this review, we (1) describe different screening tools used in US emergency department, inpatient, and intensive care unit settings; (2) highlight details of the design, implementation, and evaluation of specific tools; (3) review the available data on the process of integrating sepsis screening into an overall sepsis quality improvement program and on the effect of these screening tools on patient outcomes; (4) discuss potential harms of sepsis screening including alarm fatigue; and (5) highlight several future directions in sepsis screening, such as novel tools that incorporate artificial intelligence and machine learning methods to augment sepsis identification with the ultimate goal of precision-based approaches to sepsis recognition and treatment. IMPACT: This narrative review highlights the context that has led to the rapid growth of pediatric sepsis screening nationally. Screening tools used in US emergency department, inpatient, and intensive care unit settings are described in terms of their design, implementation, and clinical performance. Limitations and potential harms of these tools are highlighted, as well as future directions that may lead to a more precision-based approach to sepsis recognition and treatment.

Deep Learning Model to Predict Serious Infection Among Children With Central Venous Lines

Objective: Predict the onset of presumed serious infection, defined as a positive blood culture drawn and new antibiotic course of at least 4 days (PSI*), among pediatric patients with Central Venous Lines (CVLs). Design: Retrospective cohort study. Setting: Single academic children's hospital. Patients: All hospital encounters from January 2013 to December 2018, excluding the ones without a CVL or with a length-of-stay shorter than 24 h. Measurements and Main Results: Clinical features including demographics, laboratory results, vital signs, characteristics of the CVLs and medications used were extracted retrospectively from electronic medical records. Data were aggregated across all hospitals within a single pediatric health system and used to train a deep learning model to predict the occurrence of PSI* during the next 48 h of hospitalization. The proposed model prediction was compared to prediction of PSI* by a marker of illness severity (PELOD-2). The baseline prevalence of line infections was 0.34% over all segmented 48-h time windows. Events were identified among cases using onset time. All data from admission till the onset was used for cases and among controls we used all data from admission till discharge. The benchmarks were aggregated over all 48 h time windows [N=748,380 associated with 27,137 patient encounters]. The model achieved an area under the receiver operating characteristic curve of 0.993 (95% CI = [0.990, 0.996]), the enriched positive predictive value (PPV) was 23 times greater than the base prevalence. Conversely, prediction by PELOD-2 achieved a lower PPV of 1.5% [0.9%, 2.1%] which was 5 times the baseline prevalence. Conclusion: A deep learning model that employs common clinical features in the electronic health record can help predict the onset of CLABSI in hospitalized children with central venous line 48 hours prior to the time of specimen collection.

Prolonged hospital length of stay in pediatric trauma: a model for targeted interventions

BACKGROUND

In this study, trauma-specific risk factors of prolonged length of stay (LOS) in pediatric trauma were examined. Statistical and machine learning models were used to proffer ways to improve the quality of care of patients at risk of prolonged length of stay and reduce cost.

METHODS

Data from 27 hospitals were retrieved on 81,929 hospitalizations of pediatric patients with a primary diagnosis of trauma, and for which the LOS was >24 h. Nested mixed effects model was used for simplified statistical inference, while a stochastic gradient boosting model, considering high-order statistical interactions, was built for prediction.

RESULTS

Over 18.7% of the encounters had LOS >1 week. Burns and corrosion and suspected and confirmed child abuse are the strongest drivers of prolonged LOS. Several other trauma-specific and general pediatric clinical variables were also predictors of prolonged LOS. The stochastic gradient model obtained an area under the receiver operator characteristic curve of 0.912 (0.907, 0.917).

CONCLUSIONS

The high performance of the machine learning model coupled with statistical inference from the mixed effects model provide an opportunity for targeted interventions to improve quality of care of trauma patients likely to require long length of stay.

IMPACT

Targeted interventions on high-risk patients would improve the quality of care of pediatric trauma patients and reduce the length of stay. This comprehensive study includes data from multiple hospitals analyzed with advanced statistical and machine learning models. The statistical and machine learning models provide opportunities for targeted interventions and reduction in prolonged length of stay reducing the burden of hospitalization on families.

Development of a Heart Rate Variability Risk Score to Predict Organ Dysfunction and Death in Critically Ill Children

OBJECTIVES

Determine whether the Heart Rate Variability Dysfunction score, a novel age-normalized measure of autonomic nervous system dysregulation, is associated with the development of new or progressive multiple organ dysfunction syndrome or death in critically ill children.

DESIGN, SETTING, AND PATIENTS

This was a retrospective, observational cohort study from 2012 to 2018. Patients admitted to the PICU with at least 12 hours of continuous heart rate data available from bedside monitors during the first 24 hours of admission were included in the analysis.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Heart rate variability was measured using the integer heart rate variability, which is the sd of the heart rate sampled every 1 second over 5 consecutive minutes. The Heart Rate Variability Dysfunction score was derived from age-normalized values of integer heart rate variability and transformed, so that higher scores were indicative of lower integer heart rate variability and a proxy for worsening autonomic nervous system dysregulation. Heart Rate Variability Dysfunction score performance as a predictor of new or progressive multiple organ dysfunction syndrome and 28-day mortality were determined using the area under the receiver operating characteristic curve. Of the 7,223 patients who met inclusion criteria, 346 patients (4.8%) developed new or progressive multiple organ dysfunction syndrome, and 103 (1.4%) died by day 28. For every one-point increase in the median Heart Rate Variability Dysfunction score in the first 24 hours of admission, there was a 25% increase in the odds of new or progressive multiple organ dysfunction syndrome and a 51% increase in the odds of mortality. The median Heart Rate Variability Dysfunction score in the first 24 hours had an area under the receiver operating characteristic curve to discriminate new or progressive multiple organ dysfunction syndrome of 0.67 and to discriminate mortality of 0.80. These results were reproducible in a temporal validation cohort.

CONCLUSIONS

The Heart Rate Variability Dysfunction score, an age-adjusted proxy for autonomic nervous system dysregulation derived from bedside monitor data is independently associated with new or progressive multiple organ dysfunction syndrome and mortality in PICU patients. The Heart Rate Variability Dysfunction score could potentially be used as a single continuous physiologic biomarker or as part of a multivariable prediction model to increase awareness of at-risk patients and augment clinical decision-making.

Improvement of 1st-hour bundle compliance and sepsis mortality in pediatrics after the implementation of the surviving sepsis campaign guidelines

OBJECTIVES

To study the impact of the implementation of the Pediatric Surviving Sepsis Campaign protocol on early recognition of sepsis, 1-h treatment bundle and mortality.

METHODS

Retrospective, single-center study, before and after the implementation of the sepsis protocol.

OUTCOMES

sepsis recognition, compliance with the 1-h bundle (fluid resuscitation, blood culture, antibiotics), time interval to fluid resuscitation and antibiotics administration, and mortality. Patients with febrile neutropenia were excluded. The comparisons between the periods were performed using non-parametric tests and odds ratios or relative risk were calculated.

RESULTS

We studied 84 patients before and 103 after the protocol implementation. There was an increase in sepsis recognition (OR 21.5 [95% CI: 10.1-45.7]), in the compliance with the 1-h bundle as a whole (62% x 0%), and with its three components: fluid resuscitation (OR 31.1 [95% CI: 3.9-247.2]), blood culture (OR 15.9 [95% CI: 3.9-65.2]), and antibiotics (OR 35.6 [95% CI: 8.9-143.2]). Significant reduction between sepsis recognition to fluid resuscitation (152min×12min, p<0.001) and to antibiotics administration (137min×30min) also occurred. The risk of death before protocol implementation was four times greater (RR 4.1 [95% CI: 1.2-14.4]), and the absolute death risk reduction was 9%.

CONCLUSION

Even if we considered the low precision of some estimates, the lower limits of the Confidence Intervals show that the implementation of the Pediatric Surviving Sepsis Campaign guidelines alongside a qualitive assurance initiative has led to improvements in sepsis recognition, compliance with the 1-h treatment bundle, reduction in the time interval to fluid resuscitation and antibiotics, and reduction in sepsis mortality.

Temporal Differential Expression of Physiomarkers Predicts Sepsis in Critically Ill Adults

BACKGROUND

Sepsis is a life-threatening condition with high mortality rates. Early detection and treatment are critical to improving outcomes. Our primary objective was to develop artificial intelligence capable of predicting sepsis earlier using a minimal set of streaming physiological data in real time.

METHODS AND FINDINGS

A total of 29,552 adult patients were admitted to the intensive care unit across five regional hospitals in Memphis, Tenn, over 18 months from January 2017 to July 2018. From these, 5,958 patients were selected after filtering for continuous (minute-by-minute) physiological data availability. A total of 617 (10.4%) patients were identified as sepsis cases, using the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) criteria. Physiomarkers, a set of signal processing features, were derived from five physiological data streams including heart rate, respiratory rate, and blood pressure (systolic, diastolic, and mean), captured every minute from the bedside monitors. A support vector machine classifier was used for classification. The model accurately predicted sepsis up to a mean and 95% confidence interval of 17.4 ± 0.22 h before sepsis onset, with an average test accuracy of 83.0% (average sensitivity, specificity, and area under the receiver operating characteristics curve of 0.757, 0.902, and 0.781, respectively).

CONCLUSIONS

This study demonstrates that salient physiomarkers derived from continuous bedside monitoring are temporally and differentially expressed in septic patients. Using this information, minimalistic artificial intelligence models can be developed to predict sepsis earlier in critically ill patients.

Multi-Branching Temporal Convolutional Network for Sepsis Prediction

Sepsis is among the leading causes of morbidity and mortality in modern intensive care units. Accurate sepsis prediction is of critical importance to save lives and reduce medical costs. The rapid advancements in sensing and information technology facilitate the effective monitoring of patients health conditions, generating a wealth of medical data, and provide an unprecedented opportunity for data-driven diagnosis of sepsis. However, real-world medical data are often complexly structured with a high level of uncertainty (e.g., missing values, imbalanced data). Realizing the full data potential depends on developing effective analytical models. In this paper, we propose a novel predictive framework with Multi-Branching Temporal Convolutional Network (MB-TCN) to model the complexly structured medical data for robust prediction of sepsis. The MB-TCN framework not only efficiently handles the missing value and imbalanced data issues but also effectively captures the temporal pattern and heterogeneous variable interactions. We evaluate the performance of the proposed MB-TCN in predicting sepsis using real-world medical data from PhysioNet/Computing in Cardiology Challenge 2019. Experimental results show that MB-TCN outperforms existing methods that are commonly used in current practice.

Next-Generation Implementation of Chimeric Antigen Receptor T-Cell Therapy Using Digital Health

Chimeric antigen receptor T-cell (CAR-T) therapy is a paradigm-shifting immunotherapy modality in oncology; however, unique toxicities such as cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome limit its ability to be implemented more widely in the outpatient setting or at smaller-volume centers. Three operational challenges with CAR-T therapy include the following: (1) the logistics of toxicity monitoring, ie, with frequent vital sign checks and neurologic assessments; (2) the specialized knowledge required for toxicity management, particularly with regard to CRS and immune effector cell-associated neurotoxicity syndrome; and (3) the need for high-quality symptomatic and supportive care during this intensive period. In this review, we explore potential niches for digital innovations that can improve the implementation of CAR-T therapy in each of these domains. These tools include patient-facing technologies and provider-facing platforms: for example, wearable devices and mobile health apps to screen for fevers and encephalopathy, electronic patient-reported outcome assessments-based workflows to assist with symptom management, machine learning algorithms to predict emerging CRS in real time, clinical decision support systems to assist with toxicity management, and digital coaching to help maintain wellness. Televisits, which have grown in prominence since the novel coronavirus pandemic, will continue to play a key role in the monitoring and management of CAR-T-related toxicities as well. Limitations of these strategies include the need to ensure care equity and stakeholder buy-in, both operationally and financially. Nevertheless, once developed and validated, the next-generation implementation of CAR-T therapy using these digital tools may improve both its safety and accessibility.

Aiding clinical assessment of neonatal sepsis using hematological analyzer data with machine learning techniques

INTRODUCTION

Early diagnosis and antibiotic administration are essential for reducing sepsis morbidity and mortality; however, diagnosis remains difficult due to complex pathogenesis and presentation. We created a machine learning model for bacterial sepsis identification in the neonatal intensive care unit (NICU) using hematological analyzer data.

METHODS

Hematological analyzer data were gathered from NICU patients up to 48 hours prior to clinical evaluation for bacterial sepsis. Five models, Support Vector Machine, K-nearest-neighbors, Logistic Regression, Random Forest (RF), and Extreme Gradient boosting (XGBoost), were trained on 60 hematological and nine clinical variables for 2357 cases (1692 control, 665 septic). Clinical feature only models (nine variables) were additionally trained and compared with models including hematological variables. Feature importance was used to assess relative contributions of parameters to performance.

RESULTS

The three best performing models were RF, Logistic Regression, and XGBoost. RF achieved an average accuracy of 0.74, AUC-ROC of 0.73, Sensitivity of 0.38, and Specificity of 0.88. Logistic Regression achieved an average accuracy of 0.70, AUC-ROC of 0.74, Sensitivity of 0.62, and Specificity of 0.73. XGBoost achieved an average accuracy of 0.72, AUC-ROC of 0.71, Sensitivity of 0.40, and Specificity of 0.85. All models with hematological variables had significantly stronger performance than models trained on only clinical features. Neutrophil parameters had the highest average feature importance.

CONCLUSIONS

Machine learning models using hematological analyzer data can classify NICU patients as sepsis positive or negative with stronger performance compared to clinical feature only models. Hematological analyzer variables could augment current sepsis classification machine learning algorithms.

Predicting presumed serious infection among hospitalized children on central venous lines with machine learning

BACKGROUND

Presumed serious infection (PSI) is defined as a blood culture drawn and new antibiotic course of at least 4 days among pediatric patients with Central Venous Lines (CVLs). Early PSI prediction and use of medical interventions can prevent adverse outcomes and improve the quality of care.

METHODS

Clinical features including demographics, laboratory results, vital signs, characteristics of the CVLs and medications used were extracted retrospectively from electronic medical records. Data were aggregated across all hospitals within a single pediatric health system and used to train machine learning models (XGBoost and ElasticNet) to predict the occurrence of PSI 8 h prior to clinical suspicion. Prediction for PSI was benchmarked against PRISM-III.

RESULTS

Our model achieved an area under the receiver operating characteristic curve of 0.84 (95% CI = [0.82, 0.85]), sensitivity of 0.73 [0.69, 0.74], and positive predictive value (PPV) of 0.36 [0.34, 0.36]. The PRISM-III conversely achieved a lower sensitivity of 0.19 [0.16, 0.22] and PPV of 0.30 [0.26, 0.34] at a cut-off of ≥ 10. The features with the most impact on the PSI prediction were maximum diastolic blood pressure prior to PSI prediction (mean SHAP = 3.4), height (mean SHAP = 3.2), and maximum temperature prior to PSI prediction (mean SHAP = 2.6).

CONCLUSION

A machine learning model using common features in the electronic medical records can predict the onset of serious infections in children with central venous lines at least 8 h prior to when a clinical team drew a blood culture.

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.

Clinical evaluation of an interoperable clinical decision-support system for the detection of systemic inflammatory response syndrome in critically ill children

Background

Systemic inflammatory response syndrome (SIRS) is defined as a non-specific inflammatory process in the absence of infection. SIRS increases susceptibility for organ dysfunction, and frequently affects the clinical outcome of affected patients. We evaluated a knowledge-based, interoperable clinical decision-support system (CDSS) for SIRS detection on a pediatric intensive care unit (PICU).

Methods

The CDSS developed retrieves routine data, previously transformed into an interoperable format, by using model-based queries and guideline- and knowledge-based rules. We evaluated the CDSS in a prospective diagnostic study from 08/2018–03/2019. 168 patients from a pediatric intensive care unit of a tertiary university hospital, aged 0 to 18 years, were assessed for SIRS by the CDSS and by physicians during clinical routine. Sensitivity and specificity (when compared to the reference standard) with 95% Wald confidence intervals (CI) were estimated on the level of patients and patient-days.

Results

Sensitivity and specificity was 91.7% (95% CI 85.5–95.4%) and 54.1% (95% CI 45.4–62.5%) on patient level, and 97.5% (95% CI 95.1–98.7%) and 91.5% (95% CI 89.3–93.3%) on the level of patient-days. Physicians’ SIRS recognition during clinical routine was considerably less accurate (sensitivity of 62.0% (95% CI 56.8–66.9%)/specificity of 83.3% (95% CI 80.4–85.9%)) when measurd on the level of patient-days. Evaluation revealed valuable insights for the general design of the CDSS as well as specific rule modifications. Despite a lower than expected specificity, diagnostic accuracy was higher than the one in daily routine ratings, thus, demonstrating high potentials of using our CDSS to help to detect SIRS in clinical routine.

Conclusions

We successfully evaluated an interoperable CDSS for SIRS detection in PICU. Our study demonstrated the general feasibility and potentials of the implemented algorithms but also some limitations. In the next step, the CDSS will be optimized to overcome these limitations and will be evaluated in a multi-center study.

Trial registration: NCT03661450 (ClinicalTrials.gov); registered September 7, 2018.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-021-01428-7.

Early Detection of Sepsis With Machine Learning Techniques: A Brief Clinical Perspective

Sepsis is a major cause of death worldwide. Over the past years, prediction of clinically relevant events through machine learning models has gained particular attention. In the present perspective, we provide a brief, clinician-oriented vision on the following relevant aspects concerning the use of machine learning predictive models for the early detection of sepsis in the daily practice: (i) the controversy of sepsis definition and its influence on the development of prediction models; (ii) the choice and availability of input features; (iii) the measure of the model performance, the output, and their usefulness in the clinical practice. The increasing involvement of artificial intelligence and machine learning in health care cannot be disregarded, despite important pitfalls that should be always carefully taken into consideration. In the long run, a rigorous multidisciplinary approach to enrich our understanding in the application of machine learning techniques for the early recognition of sepsis may show potential to augment medical decision-making when facing this heterogeneous and complex syndrome.

Validation of a machine learning algorithm for early severe sepsis prediction: a retrospective study predicting severe sepsis up to 48 h in advance using a diverse dataset from 461 US hospitals

BACKGROUND

Severe sepsis and septic shock are among the leading causes of death in the United States and sepsis remains one of the most expensive conditions to diagnose and treat. Accurate early diagnosis and treatment can reduce the risk of adverse patient outcomes, but the efficacy of traditional rule-based screening methods is limited. The purpose of this study was to develop and validate a machine learning algorithm (MLA) for severe sepsis prediction up to 48 h before onset using a diverse patient dataset.

METHODS

Retrospective analysis was performed on datasets composed of de-identified electronic health records collected between 2001 and 2017, including 510,497 inpatient and emergency encounters from 461 health centers collected between 2001 and 2015, and 20,647 inpatient and emergency encounters collected in 2017 from a community hospital. MLA performance was compared to commonly used disease severity scoring systems and was evaluated at 0, 4, 6, 12, 24, and 48 h prior to severe sepsis onset.

RESULTS

270,438 patients were included in analysis. At time of onset, the MLA demonstrated an AUROC of 0.931 (95% CI 0.914, 0.948) and a diagnostic odds ratio (DOR) of 53.105 on a testing dataset, exceeding MEWS (0.725, P < .001; DOR 4.358), SOFA (0.716; P < .001; DOR 3.720), and SIRS (0.655; P < .001; DOR 3.290). For prediction 48 h prior to onset, the MLA achieved an AUROC of 0.827 (95% CI 0.806, 0.848) on a testing dataset. On an external validation dataset, the MLA achieved an AUROC of 0.948 (95% CI 0.942, 0.954) at the time of onset, and 0.752 at 48 h prior to onset.

CONCLUSIONS

The MLA accurately predicts severe sepsis onset up to 48 h in advance using only readily available vital signs extracted from the existing patient electronic health records. Relevant implications for clinical practice include improved patient outcomes from early severe sepsis detection and treatment.

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.

Machine Learning Models of Acute Kidney Injury Prediction in Acute Pancreatitis Patients

Background. Acute kidney injury (AKI) has long been recognized as a common and important complication of acute pancreatitis (AP). In the study, machine learning (ML) techniques were used to establish predictive models for AKI in AP patients during hospitalization. This is a retrospective review of prospectively collected data of AP patients admitted within one week after the onset of abdominal pain to our department from January 2014 to January 2019. Eighty patients developed AKI after admission (AKI group) and 254 patients did not (non-AKI group) in the hospital. With the provision of additional information such as demographic characteristics or laboratory data, support vector machine (SVM), random forest (RF), classification and regression tree (CART), and extreme gradient boosting (XGBoost) were used to build models of AKI prediction and compared to the predictive performance of the classic model using logistic regression (LR). XGBoost performed best in predicting AKI with an AUC of 91.93% among the machine learning models. The AUC of logistic regression analysis was 87.28%. Present findings suggest that compared to the classical logistic regression model, machine learning models using features that can be easily obtained at admission had a better performance in predicting AKI in the AP patients.

Neural Networks to Predict Radiographic Brain Injury in Pediatric Patients Treated with Extracorporeal Membrane Oxygenation

Brain injury is a significant source of morbidity and mortality for pediatric patients treated with Extracorporeal Membrane Oxygenation (ECMO). Our objective was to utilize neural networks to predict radiographic evidence of brain injury in pediatric ECMO-supported patients and identify specific variables that can be explored for future research. Data from 174 ECMO-supported patients were collected up to 24 h prior to, and for the duration of, the ECMO course. Thirty-five variables were collected, including physiological data, markers of end-organ perfusion, acid-base homeostasis, vasoactive infusions, markers of coagulation, and ECMO-machine factors. The primary outcome was the presence of radiologic evidence of moderate to severe brain injury as established by brain CT or MRI. This information was analyzed by a neural network, and results were compared to a logistic regression model as well as clinician judgement. The neural network model was able to predict brain injury with an Area Under the Curve (AUC) of 0.76, 73% sensitivity, and 80% specificity. Logistic regression had 62% sensitivity and 61% specificity. Clinician judgment had 39% sensitivity and 69% specificity. Sequential feature group masking demonstrated a relatively greater contribution of physiological data and minor contribution of coagulation factors to the model's performance. These findings lay the foundation for further areas of research directions.

Machine learning versus conventional clinical methods in guiding management of heart failure patients-a systematic review

Machine learning (ML) algorithms “learn” information directly from data, and their performance improves proportionally with the number of high-quality samples. The aim of our systematic review is to present the state of the art regarding the implementation of ML techniques in the management of heart failure (HF) patients. We manually searched MEDLINE and Cochrane databases as well the reference lists of the relevant review studies and included studies. Our search retrieved 122 relevant studies. These studies mainly refer to (a) the role of ML in the classification of HF patients into distinct categories which may require a different treatment strategy, (b) discrimination of HF patients from the healthy population or other diseases, (c) prediction of HF outcomes, (d) identification of HF patients from electronic records and identification of HF patients with similar characteristics who may benefit form a similar treatment strategy, (e) supporting the extraction of important data from clinical notes, and (f) prediction of outcomes in HF populations with implantable devices (left ventricular assist device, cardiac resynchronization therapy). We concluded that ML techniques may play an important role for the efficient construction of methodologies for diagnosis, management, and prediction of outcomes in HF patients.

Augmented intelligence in pediatric anesthesia and pediatric critical care

PURPOSE OF REVIEW

Acute care technologies, including novel monitoring devices, big data, increased computing capabilities, machine-learning algorithms and automation, are converging. This enables the application of augmented intelligence for improved outcome predictions, clinical decision-making, and offers unprecedented opportunities to improve patient outcomes, reduce costs, and improve clinician workflow. This article briefly explores recent work in the areas of automation, artificial intelligence and outcome prediction models in pediatric anesthesia and pediatric critical care.

RECENT FINDINGS

Recent years have yielded little published research into pediatric physiological closed loop control (a type of automation) beyond studies focused on glycemic control for type 1 diabetes. However, there has been a greater range of research in augmented decision-making, leveraging artificial intelligence and machine-learning techniques, in particular, for pediatric ICU outcome prediction.

SUMMARY

Most studies focusing on artificial intelligence demonstrate good performance on prediction or classification, whether they use traditional statistical tools or novel machine-learning approaches. Yet the challenges of implementation, user acceptance, ethics and regulation cannot be underestimated. Areas in which there is easy access to routinely labeled data and robust outcomes, such as those collected through national networks and quality improvement programs, are likely to be at the forefront of the adoption of these advances.