Assessing clinical heterogeneity in sepsis through treatment patterns and machine learning

Prediction of post-delivery hemoglobin levels with machine learning algorithms

Predicting postpartum hemorrhage (PPH) before delivery is crucial for enhancing patient outcomes, enabling timely transfer and implementation of prophylactic therapies. We attempted to utilize machine learning (ML) using basic pre-labor clinical data and laboratory measurements to predict postpartum Hemoglobin (Hb) in non-complicated singleton pregnancies. The local databases of two academic care centers on patient delivery were incorporated into the current study. Patients with preexisting coagulopathy, traumatic cases, and allogenic blood transfusion were excluded from all analyses. The association of pre-delivery variables with 24-h post-delivery hemoglobin level was evaluated using feature selection with Elastic Net regression and Random Forest algorithms. A suite of ML algorithms was employed to predict post-delivery Hb levels. Out of 2051 pregnant women, 1974 were included in the final analysis. After data pre-processing and redundant variable removal, the top predictors selected via feature selection for predicting post-delivery Hb were parity (B: 0.09 [0.05-0.12]), gestational age, pre-delivery hemoglobin (B:0.83 [0.80-0.85]) and fibrinogen levels (B:0.01 [0.01-0.01]), and pre-labor platelet count (B*1000: 0.77 [0.30-1.23]). Among the trained algorithms, artificial neural network provided the most accurate model (Root mean squared error: 0.62), which was subsequently deployed as a web-based calculator: https://predictivecalculators.shinyapps.io/ANN-HB . The current study shows that ML models could be utilized as accurate predictors of indirect measures of PPH and can be readily incorporated into healthcare systems. Further studies with heterogenous population-based samples may further improve the generalizability of these models.

Public Discourse, User Reactions, and Conspiracy Theories on the X Platform About HIV Vaccines: Data Mining and Content Analysis

BACKGROUND

The initiation of clinical trials for messenger RNA (mRNA) HIV vaccines in early 2022 revived public discussion on HIV vaccines after 3 decades of unsuccessful research. These trials followed the success of mRNA technology in COVID-19 vaccines but unfolded amid intense vaccine debates during the COVID-19 pandemic. It is crucial to gain insights into public discourse and reactions about potential new vaccines, and social media platforms such as X (formerly known as Twitter) provide important channels.

OBJECTIVE

Drawing from infodemiology and infoveillance research, this study investigated the patterns of public discourse and message-level drivers of user reactions on X regarding HIV vaccines by analyzing posts using machine learning algorithms. We examined how users used different post types to contribute to topics and valence and how these topics and valence influenced like and repost counts. In addition, the study identified salient aspects of HIV vaccines related to COVID-19 and prominent anti-HIV vaccine conspiracy theories through manual coding.

METHODS

We collected 36,424 English-language original posts about HIV vaccines on the X platform from January 1, 2022, to December 31, 2022. We used topic modeling and sentiment analysis to uncover latent topics and valence, which were subsequently analyzed across post types in cross-tabulation analyses and integrated into linear regression models to predict user reactions, specifically likes and reposts. Furthermore, we manually coded the 1000 most engaged posts about HIV and COVID-19 to uncover salient aspects of HIV vaccines related to COVID-19 and the 1000 most engaged negative posts to identify prominent anti-HIV vaccine conspiracy theories.

RESULTS

Topic modeling revealed 3 topics: HIV and COVID-19, mRNA HIV vaccine trials, and HIV vaccine and immunity. HIV and COVID-19 underscored the connections between HIV vaccines and COVID-19 vaccines, as evidenced by subtopics about their reciprocal impact on development and various comparisons. The overall valence of the posts was marginally positive. Compared to self-composed posts initiating new conversations, there was a higher proportion of HIV and COVID-19-related and negative posts among quote posts and replies, which contribute to existing conversations. The topic of mRNA HIV vaccine trials, most evident in self-composed posts, increased repost counts. Positive valence increased like and repost counts. Prominent anti-HIV vaccine conspiracy theories often falsely linked HIV vaccines to concurrent COVID-19 and other HIV-related events.

CONCLUSIONS

The results highlight COVID-19 as a significant context for public discourse and reactions regarding HIV vaccines from both positive and negative perspectives. The success of mRNA COVID-19 vaccines shed a positive light on HIV vaccines. However, COVID-19 also situated HIV vaccines in a negative context, as observed in some anti-HIV vaccine conspiracy theories misleadingly connecting HIV vaccines with COVID-19. These findings have implications for public health communication strategies concerning HIV vaccines.

A customised down-sampling machine learning approach for sepsis prediction

OBJECTIVE

Sepsis is a life-threatening condition in the ICU and requires treatment in time. Despite the accuracy of existing sepsis prediction models, insufficient focus on reducing alarms could worsen alarm fatigue and desensitisation in ICUs, potentially compromising patient safety. In this retrospective study, we aim to develop an accurate, robust, and readily deployable method in ICUs, only based on the vital signs and laboratory tests.

METHODS

Our method consists of a customised down-sampling process and a specific dynamic sliding window and XGBoost to offer sepsis prediction. The down-sampling process was applied to the retrospective data for training the XGBoost model. During the testing stage, the dynamic sliding window and the trained XGBoost were used to predict sepsis on the retrospective datasets, PhysioNet and FHC.

RESULTS

With the filtered data from PhysioNet, our method achieved 80.74% accuracy (77.90% sensitivity and 84.42% specificity) and 83.95% (84.82% sensitivity and 82.00% specificity) on the test set of PhysioNet-A and PhysioNet-B, respectively. The AUC score was 0.89 for both datasets. On the FHC dataset, our method achieved 92.38% accuracy (88.37% sensitivity and 95.16% specificity) and 0.98 AUC score on the test set of FHC.

CONCLUSION

Our results indicate that the down-sampling process and the dynamic sliding window with XGBoost brought robust and accurate performance to give sepsis prediction under various hospital settings. The localisation and robustness of our method can assist in sepsis diagnosis in different ICU settings.

Advances and Challenges in Sepsis Management: Modern Tools and Future Directions

Sepsis, a critical condition marked by systemic inflammation, profoundly impacts both innate and adaptive immunity, often resulting in lymphopenia. This immune alteration can spare regulatory T cells (Tregs) but significantly affects other lymphocyte subsets, leading to diminished effector functions, altered cytokine profiles, and metabolic changes. The complexity of sepsis stems not only from its pathophysiology but also from the heterogeneity of patient responses, posing significant challenges in developing universally effective therapies. This review emphasizes the importance of phenotyping in sepsis to enhance patient-specific diagnostic and therapeutic strategies. Phenotyping immune cells, which categorizes patients based on clinical and immunological characteristics, is pivotal for tailoring treatment approaches. Flow cytometry emerges as a crucial tool in this endeavor, offering rapid, low cost and detailed analysis of immune cell populations and their functional states. Indeed, this technology facilitates the understanding of immune dysfunctions in sepsis and contributes to the identification of novel biomarkers. Our review underscores the potential of integrating flow cytometry with omics data, machine learning and clinical observations to refine sepsis management, highlighting the shift towards personalized medicine in critical care. This approach could lead to more precise interventions, improving outcomes in this heterogeneously affected patient population.

Training the next generation of delivery science researchers: 10-year experience of a post-doctoral research fellowship program within an integrated care system

Introduction

Learning health systems require a workforce of researchers trained in the methods of identifying and overcoming barriers to effective, evidence-based care. Most existing postdoctoral training programs, such as NIH-funded postdoctoral T32 awards, support basic and epidemiological science with very limited focus on rigorous delivery science methods for improving care. In this report, we present the 10-year experience of developing and implementing a Delivery Science postdoctoral fellowship embedded within an integrated health care delivery system.

Methods

In 2012, the Kaiser Permanente Northern California Division of Research designed and implemented a 2-year postdoctoral Delivery Science Fellowship research training program to foster research expertise in identifying and addressing barriers to evidence-based care within health care delivery systems.

Results

Since 2014, 20 fellows have completed the program. Ten fellows had PhD-level scientific training, and 10 fellows had clinical doctorates (eg, MD, RN/PhD, PharmD). Fellowship alumni have graduated to faculty research positions at academic institutions (9), and research or clinical organizations (4). Seven alumni now hold positions in Kaiser Permanente's clinical operations or medical group (7).

Conclusions

This delivery science fellowship program has succeeded in training graduates to address delivery science problems from both research and operational perspectives. In the next 10 years, additional goals of the program will be to expand its reach (eg, by developing joint research training models in collaboration with clinical fellowships) and strengthen mechanisms to support transition from fellowship to the workforce, especially for researchers from underrepresented groups.

A Prospective Cohort Protocol for the Remnant Investigation in Sepsis Study

BACKGROUND

Sepsis is a common and deadly syndrome, accounting for more than 11 million deaths annually. To mature a deeper understanding of the host and pathogen mechanisms contributing to poor outcomes in sepsis, and thereby possibly inform new therapeutic targets, sophisticated, and expensive biorepositories are typically required. We propose that remnant biospecimens are an alternative for mechanistic sepsis research, although the viability and scientific value of such remnants are unknown.

METHODS AND RESULTS

The Remnant Biospecimen Investigation in Sepsis study is a prospective cohort study of 225 adults (age ≥ 18 yr) presenting to the emergency department with community sepsis, defined as sepsis-3 criteria within 6 hours of arrival. The primary objective was to determine the scientific value of a remnant biospecimen repository in sepsis linked to clinical phenotyping in the electronic health record. We will study candidate multiomic readouts of sepsis biology, governed by a conceptual model, and determine the precision, accuracy, integrity, and comparability of proteins, small molecules, lipids, and pathogen sequencing in remnant biospecimens compared with paired biospecimens obtained according to research protocols. Paired biospecimens will include plasma from sodium-heparin, EDTA, sodium fluoride, and citrate tubes.

CONCLUSIONS

The study has received approval from the University of Pittsburgh Human Research Protection Office (Study 21120013). Recruitment began on October 25, 2022, with planned release of primary results anticipated in 2024. Results will be made available to the public, the funders, critical care societies, laboratory medicine scientists, and other researchers.

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.

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.

First Sampled High-Sensitive Cardiac Troponin T is Associated With One-Year Mortality in Sepsis Patients and 30- to 365-Day Mortality in Sepsis Survivors

BACKGROUND

Previous studies using cardiac troponin levels to investigate the relationship between myocardial injury and mortality in sepsis patients have been conflicting. Our aim was to investigate the relationship between plasma high-sensitive cardiac troponin T (hs-cTnT) level and 30-day and 1-year mortality in sepsis patients and 30- to 365-day mortality in sepsis survivors.

METHODS

Sepsis patients requiring vasopressor support and admitted to our institution between 2012 and 2021 (n = 586) were included in this retrospective cohort study. Elevated hs-cTnT values (≥15 ng/L) were divided into quartiles (Q): Q1 15-35 ng/L; Q2 36-61 ng/L; Q3 62-125 ng/L; Q4 126-8630 ng/L. Stratified Kaplan-Meier curves and multivariable Cox regression were used for survival analyses.

RESULTS

First sampled hs-cTnT was elevated in 529 (90%) patients. One-year mortality was 45% (n = 264). Increasing level of hs-cTnT was independently associated with higher adjusted hazard ratios (HR) for 1-year mortality compared with normal levels: Q1 HR 2.9 (95% confidence interval [CI], 1.03-8.1); Q2 HR 3.5 (95% CI, 1.2-9.8); Q3 HR 4.8 (95% CI, 1.7-13.4); Q4 HR 5.7 (95% CI, 2.1-16). In acute phase survivors, first sampled hs-cTnT was an independent predictor of 30- to 365-day mortality (HR 1.3; 95% CI, 1.1-1.6 per loge hs-cTnT).

CONCLUSIONS

First sampled plasma hs-cTnT in critically ill sepsis patients was independently associated with 30-day and 1-year mortality. Importantly, first sampled hs-cTnT was associated with mortality during the convalescence phase (30- to 365-day) and could be a feasible marker to identify acute phase survivors at high risk of death.

Sepsis Prediction Model for Determining Sepsis vs SIRS, qSOFA, and SOFA

Importance

The Sepsis Prediction Model (SPM) is a proprietary decision support tool created by Epic Systems; it generates a predicting sepsis score (PSS). The model has not undergone validation against existing sepsis prediction tools, such as Systemic Inflammatory Response Syndrome (SIRS), Sequential Organ Failure Assessment (SOFA), or quick Sepsis-Related Organ Failure Asessement (qSOFA).

Objective

To assess the validity and timeliness of the SPM compared with SIRS, qSOFA, and SOFA.

Design, Setting, and Participants

This retrospective cohort study included all adults admitted to 5 acute care hospitals in a single US health system between June 5, 2019, and December 31, 2020. Data analysis was conducted from March 2021 to February 2023.

Main Outcomes and Measures

A sepsis event was defined as receipt of 4 or more days of antimicrobials, blood cultures collected within ±48 hours of initial antimicrobial, and at least 1 organ dysfunction as defined by the organ dysfunction criteria optimized for the electronic health record (eSOFA). Time zero was defined as 15 minutes prior to qualifying antimicrobial or blood culture order.

Results

Of 60 507 total admissions, 1663 (2.7%) met sepsis criteria, with 1324 electronic health record-confirmed sepsis (699 [52.8%] male patients; 298 [22.5%] Black patients; 46 [3.5%] Hispanic/Latinx patients; 945 [71.4%] White patients), 339 COVID-19 sepsis (183 [54.0%] male patients; 98 [28.9%] Black patients; 36 [10.6%] Hispanic/Latinx patients; and 189 [55.8%] White patients), and 58 844 (97.3%; 26 632 [45.2%] male patients; 12 698 [21.6%] Black patients; 3367 [5.7%] Hispanic/Latinx patients; 40 491 White patients) did not meet sepsis criteria. The median (IQR) age was 63 (51 to 73) years for electronic health record-confirmed sepsis, 69 (60 to 77) years for COVID-19 sepsis, and 60 (42 to 72) years for nonsepsis admissions. Within the vendor recommended threshold PSS range of 5 to 8, PSS of 8 or greater had the highest balanced accuracy for classifying a sepsis admission at 0.79 (95% CI, 0.78 to 0.80). Change in SOFA score of 2 or more had the highest sensitivity, at 0.97 (95% CI, 0.97 to 0.98). At a PSS of 8 or greater, median (IQR) time to score positivity from time zero was 68.00 (6.75 to 605.75) minutes. For SIRS, qSOFA, and SOFA, median (IQR) time to score positivity was 7.00 (-105.00 to 08.00) minutes, 74.00 (-22.25 to 599.25) minutes, and 28.00 (-108.50 to 134.00) minutes, respectively.

Conclusions and Relevance

In this cohort study of hospital admissions, balanced accuracy of the SPM outperformed other models at higher threshold PSS; however, application of the SPM in a clinical setting was limited by poor timeliness as a sepsis screening tool as compared to SIRS and SOFA.

Machine Learning for Early Prediction of Sepsis in Intensive Care Unit (ICU) Patients

Background and Objectives: Early detection of sepsis is crucial and can save lives. However, identifying sepsis early and accurately remains a difficult task in the medical field. This study aims to investigate a new machine-learning approach. By analyzing the clinical laboratory results and vital signs of adult patients in the ICU, this approach can predict and detect the initial signs of sepsis. Materials and Methods: To examine survival rates and predict outcomes, the study utilized several models, including the proportional hazards model and data mining algorithms. We analyzed data from the BESTCare database at KAMC, with a focus on patients aged 14 and older who were admitted to the ICU between April and October 2018. We conducted a thorough analysis of the medical records of a total of 1182 patients who were diagnosed with sepsis. Results: We studied two approaches to predict sepsis in ICU patients. The regression model utilizing survival analysis showed moderate predictive ability, emphasizing the importance of only three factors-time (from sepsis to an outcome; discharge or death), lactic acid, and temperature-had a significant p-value (p = 0.000568, p = 0.01, p = 0.02, respectively). Other data mining algorithms may have limitations due to their assumptions of variable independence and linear classification nature. Conclusions: To achieve progress and accuracy in the field of sepsis prediction, it is important to continuously strive for improvement. By meticulously cleaning and selecting data attributes, we can create a strong foundation for future advancements in this area.

The Gut Microbiota Contributes to Systemic Responses and Liver Injury in Gut-Derived Sepsis

The gut microbiota, as a major source of opportunistic pathogens, poses a great threat to systemic infection, whereas the role of the gut microbiota in sepsis is underestimated. Here, we aimed to explore the effects of different gut microbiota patterns (namely, enterotypes) in cecal ligation and puncture (CLP)-induced murine sepsis. To achieve this purpose, we built four kinds of enterotypes by exposing mice to different types of antibiotics (azithromycin, amoxicillin, metronidazole, and levofloxacin). The results showed that antibiotic exposure induced different enterotypes, which, in turn, led to varying levels of systemic inflammation in septic mice, with amoxicillin-associated enterotypes exhibiting the most severe inflammation, followed by metronidazole, azithromycin, and levofloxacin. Specifically, the amoxicillin-associated enterotype was characterized by an abundance of intestinal opportunistic pathogens, including Enterobacteriaceae, Sutterellaceae, and Morganellaceae. This enterotype played a significant role in promoting the pathogenic potential of the gut microbiota, ultimately contributing to the development of severe systemic inflammation. Furthermore, the amoxicillin-associated enterotype exaggerated the sepsis-related liver injury, as evidenced by higher levels of alanine aminotransferase, aspartate transaminase, and hepatic malondialdehyde. The results of the RNA sequencing and the fecal suspension intraperitoneal injection sepsis model indicated that the amoxicillin-associated enterotype provoked acute hepatic immune responses and led to more significant metabolic compensation in the event of sepsis. Collectively, we concluded that the gut microbiota was one crucial factor for heterogeneity in sepsis, where the modulated gut microbiota likely prevented or reduced the serious consequences of sepsis, at least in gut-derived sepsis.

Safety of megadose meropenem in the empirical treatment of nosocomial sepsis: a pilot randomized clinical trial

Objective: To evaluate the safety of megadose meropenem as empirical treatment of nosocomial sepsis. Materials & methods: Critically ill patients diagnosed with sepsis received either high-dose (2 g every 8 h) or megadose (4 g every 8 h) meropenem as an intravenous infusion over 3 h. Results: A total of 23 patients with nosocomial sepsis were eligible and included in the megadose (n = 11) or high-dose (n = 12) group. No treatment-related adverse events were observed during a 14-day follow-up. Clinical response was also comparable between the groups. Conclusion: Megadose meropenem may be considered for empirical treatment of nosocomial sepsis without serious concern regarding its safety.

The Gut Microbiome Modulates Body Temperature Both in Sepsis and Health

Rationale: Among patients with sepsis, variation in temperature trajectories predicts clinical outcomes. In healthy individuals, normal body temperature is variable and has decreased consistently since the 1860s. The biologic underpinnings of this temperature variation in disease and health are unknown. Objectives: To establish and interrogate the role of the gut microbiome in calibrating body temperature. Methods: We performed a series of translational analyses and experiments to determine whether and how variation in gut microbiota explains variation in body temperature in sepsis and in health. We studied patient temperature trajectories using electronic medical record data. We characterized gut microbiota in hospitalized patients using 16S ribosomal RNA gene sequencing. We modeled sepsis using intraperitoneal LPS in mice and modulated the microbiome using antibiotics, germ-free, and gnotobiotic animals. Measurements and Main Results: Consistent with prior work, we identified four temperature trajectories in patients hospitalized with sepsis that predicted clinical outcomes. In a separate cohort of 116 hospitalized patients, we found that the composition of patients' gut microbiota at admission predicted their temperature trajectories. Compared with conventional mice, germ-free mice had reduced temperature loss during experimental sepsis. Among conventional mice, heterogeneity of temperature response in sepsis was strongly explained by variation in gut microbiota. Healthy germ-free and antibiotic-treated mice both had lower basal body temperatures compared with control animals. The Lachnospiraceae family was consistently associated with temperature trajectories in hospitalized patients, experimental sepsis, and antibiotic-treated mice. Conclusions: The gut microbiome is a key modulator of body temperature variation in both health and critical illness and is thus a major, understudied target for modulating physiologic heterogeneity in sepsis.

Biomaterial mediated simultaneous delivery of spermine and alpha ketoglutarate modulate metabolism and innate immune cell phenotype in sepsis mouse models

Although different metabolic pathways have been associated with distinct macrophage phenotypes, the field of utilizing metabolites to modulate macrophage phenotype is in a nascent stage. In this report, we developed microparticles based on polymerization of alpha-ketoglutarate (a Krebs cycle metabolite), with or without encapsulation of spermine (a polyamine metabolite), to modulate cell phenotype that are critical for resolution of inflammation. Poly (alpha-ketoglutarate) microparticles encapsulated and released spermine (spermine (encap)paKG MPs) in vitro, which was accelerated in an acidic environment. When delivered to bone marrow-derived-macrophages, spermine (encap)paKG MPs induced a complex phenotypic profile outside of the typical M1/M2 paradigm, with distinct effects in the presence or absence of the pro-inflammatory stimulus lipopolysaccharide. Of particular interest was the increase in expression of CD163, which has been linked to anti-inflammatory responses in sepsis. Therefore, we systemically administered spermine (encap)paKG MPs to two different murine models of sepsis using acute or chronic injection of LPS. Macrophages and neutrophils in the liver and spleen of animals treated with spermine (encap)paKG MPs increased expression of CD163, concomitant with normalizing of glycolysis and oxidative phosphorylation, in both models. Overall, these results show that spermine (encap)paKG MPs modulate macrophage phenotype in vitro and in vivo, with potential applications in inflammation-associated diseases.

Research frontiers and trends in the application of artificial intelligence to sepsis: A bibliometric analysis

Background

With the increasing interest of academics in the application of artificial intelligence to sepsis, thousands of papers on this field had been published in the past few decades. It is difficult for researchers to understand the themes and latest research frontiers in this field from a multi-dimensional perspective. Consequently, the purpose of this study is to analyze the relevant literature in the application of artificial intelligence to sepsis through bibliometrics software, so as to better understand the development status, study the core hotspots and future development trends of this field.

Methods

We collected relevant publications in the application of artificial intelligence to sepsis from the Web of Science Core Collection in 2000 to 2021. The type of publication was limited to articles and reviews, and language was limited to English. Research cooperation network, journals, cited references, keywords in this field were visually analyzed by using CiteSpace, VOSviewer, and COOC software.

Results

A total of 8,481 publications in the application of artificial intelligence to sepsis between 2000 and 2021 were included, involving 8,132 articles and 349 reviews. Over the past 22 years, the annual number of publications had gradually increased exponentially. The USA was the most productive country, followed by China. Harvard University, Schuetz, Philipp, and Intensive Care Medicine were the most productive institution, author, and journal, respectively. Vincent, Jl and Critical Care Medicine were the most cited author and cited journal, respectively. Several conclusions can be drawn from the analysis of the cited references, including the following: screening and identification of sepsis biomarkers, treatment and related complications of sepsis, and precise treatment of sepsis. Moreover, there were a spike in searches relating to machine learning, antibiotic resistance and accuracy based on burst detection analysis.

Conclusion

This study conducted a comprehensive and objective analysis of the publications on the application of artificial intelligence in sepsis. It can be predicted that precise treatment of sepsis through machine learning technology is still research hotspot in this field.

Extending artificial intelligence research in the clinical domain: a theoretical perspective

Academic research to the utilization of artificial intelligence (AI) has been proliferated over the past few years. While AI and its subsets are continuously evolving in the fields of marketing, social media and finance, its application in the daily practice of clinical care is insufficiently explored. In this systematic review, we aim to landscape various application areas of clinical care in terms of the utilization of machine learning to improve patient care. Through designing a specific smart literature review approach, we give a new insight into existing literature identified with AI technologies in the clinical domain. Our review approach focuses on strategies, algorithms, applications, results, qualities, and implications using the Latent Dirichlet Allocation topic modeling. A total of 305 unique articles were reviewed, with 115 articles selected using Latent Dirichlet Allocation topic modeling, meeting our inclusion criteria. The primary result of this approach incorporates a proposition for future research direction, abilities, and influence of AI technologies and displays the areas of disease management in clinics. This research concludes with disease administrative ramifications, limitations, and directions for future research.

Sequences of Events from the Electronic Medical Record and the Onset of Infection

We present a novel model of time-series analysis to learn from electronic health record (EHR) data when infection occurred in the intensive care unit (ICU) by translating methods from proteomics and Bayesian statistics. Using 48,536 patients hospitalized in an ICU, we describe each hospital course as an 'alphabet' of 23 physician actions ('events') in temporal order. We analyze these as k-mers of length 3-12 events and apply a Bayesian model of (cumulative) relative risk (RR). The log2-transformed RR (median=0.248, mean=0.226) supported the conclusion that the events selected were individually associated with increased risk of infection. Selecting from all possible cutoffs of maximum gain (MG), MG>0.0244 predicts administration of antibiotics with PPV 82.0 %, NPV 44.4 %, and AUC 0.706. Our approach holds value for retrospective analysis of other clinical syndromes for which time-of-onset is critical to analysis but poorly marked in EHRs, including delirium and decompensation.

Balanced Crystalloids versus Normal Saline in Adults with Sepsis: A Comprehensive Systematic Review and Meta-Analysis

The crystalloid fluid of choice in sepsis remains debatable. We aimed to perform a comprehensive meta-analysis to compare the effect of balanced crystalloids (BC) vs. normal saline (NS) in adults with sepsis. A systematic search of PubMed, EMBASE, and Web of Sciences databases through 22 January 2022, was performed for studies that compared BC vs. NS in adults with sepsis. Our outcomes included mortality and acute kidney injury (AKI), need for renal replacement therapy (RRT), and ICU length of stay (LOS). Pooled risk ratio (RR) and mean difference (MD) with the corresponding 95% confidence intervals (CIs) were obtained using a random-effect model. Fifteen studies involving 20,329 patients were included. Overall, BC showed a significant reduction in the overall mortality (RR 0.88, 95% CI 0.81-0.96), 28/30-day mortality (RR 0.87, 95% CI 0.79-0.95), and AKI (RR 0.85, 95% CI 0.77-0.93) but similar 90-day mortality (RR 0.96, 95% CI 0.90-1.03), need for RRT (RR 0.91, 95% CI 0.76-1.08), and ICU LOS (MD -0.25 days, 95% CI -3.44, 2.95), were observed between the two groups. However, subgroup analysis of randomized controlled trials (RCTs) showed no statistically significant differences in overall mortality (RR 0.92, 95% CI 0.82-1.02), AKI (RR 0.71, 95% CI 0.47-1.06), and need for RRT (RR 0.71, 95% CI 0.36-1.41). Our meta-analysis demonstrates that overall BC was associated with reduced mortality and AKI in sepsis compared to NS among patients with sepsis. However, subgroup analysis of RCTs showed no significant differences in both overall mortality and AKI between the groups. There was no significant difference in the need for RRT or ICU LOS between BC and NS. Pending further data, our study supports using BC over NS for fluid resuscitation in adults with sepsis. Further large-scale RCTs are necessary to validate our findings.

Association of ICU Admission and Outcomes in Sepsis and Acute Respiratory Failure

Rationale: Many decisions to admit patients to the ICU are not grounded in evidence regarding who benefits from such triage, straining ICU capacity and limiting its cost-effectiveness. Objectives: To measure the benefits of ICU admission for patients with sepsis or acute respiratory failure. Methods: At 27 United States hospitals across two health systems from 2013 to 2018, we performed a retrospective cohort study using two-stage instrumental variable quantile regression with a strong instrument (hospital capacity strain) governing ICU versus ward admission among high-acuity patients (i.e., laboratory-based acute physiology score v2 ⩾ 100) with sepsis and/or acute respiratory failure who did not require mechanical ventilation or vasopressors in the emergency department. Measurements and Main Results: Among patients with sepsis (n = 90,150), admission to the ICU was associated with a 1.32-day longer hospital length of stay (95% confidence interval [CI], 1.01-1.63; P < 0.001) (when treating deaths as equivalent to long lengths of stay) and higher in-hospital mortality (odds ratio, 1.48; 95% CI, 1.13-1.88; P = 0.004). Among patients with respiratory failure (n = 45,339), admission to the ICU was associated with a 0.82-day shorter hospital length of stay (95% CI, -1.17 to -0.46; P < 0.001) and reduced in-hospital mortality (odds ratio, 0.75; 95% CI, 0.57-0.96; P = 0.04). In sensitivity analyses of length of stay, excluding, ignoring, or censoring death, results were similar in sepsis but not in respiratory failure. In subgroup analyses, harms of ICU admission for patients with sepsis were concentrated among older patients and those with fewer comorbidities, and the benefits of ICU admission for patients with respiratory failure were concentrated among older patients, highest-acuity patients, and those with more comorbidities. Conclusions: Among high-acuity patients with sepsis who did not require life support in the emergency department, initial admission to the ward, compared with the ICU, was associated with shorter length of stay and improved survival, whereas among patients with acute respiratory failure, triage to the ICU compared with the ward was associated with improved survival.

Evaluating machine learning models for sepsis prediction: A systematic review of methodologies

Studies for sepsis prediction using machine learning are developing rapidly in medical science recently. In this review, we propose a set of new evaluation criteria and reporting standards to assess 21 qualified machine learning models for quality analysis based on PRISMA. Our assessment shows that (1.) the definition of sepsis is not consistent among the studies; (2.) data sources and data preprocessing methods, machine learning models, feature engineering, and inclusion types vary widely among the studies; (3.) the closer to the onset of sepsis, the higher the value of AUROC is; (4.) the improvement in AUROC is primarily due to using machine learning as a feature engineering tool; (5.) deep neural networks coupled with Sepsis-3 diagnostic criteria tend to yield better results on the time series data collected from patients with sepsis. The new evaluation criteria and reporting standards will facilitate the development of improved machine learning models for clinical applications.

•

New evaluation/reporting standard for sepsis prediction machine learning models

•

Major limitations in the current models for sepsis prediction have been identified

•

We strongly suggest using machine learning as a feature engineering tool

•

Recommending multilayer neural networks and Sepsis 3.0 for yield better result

Evaluation of Incident 7-Day Infection and Sepsis Hospitalizations in an Integrated Health System

RATIONALE

Pre-hospital opportunities to predict infection and sepsis hospitalization may exist, but little is known about their incidence following common healthcare encounters.

OBJECTIVES

To evaluate the incidence and timing of infection and sepsis hospitalization within 7 days of living hospital discharge, emergency department discharge, and ambulatory visit settings.

METHODS

In each setting, we identified patients in clinical strata based on the presence of infection and severity of illness. We estimated number needed to evaluate values with hypothetical predictive model operating characteristics.

RESULTS

We identified 97,614,228 encounters including 1,117,702 (1.1 %) hospital discharges, 4,635,517 (4.7%) emergency department discharges, and 91,861,009 (94.1 %) ambulatory visits between 2012 and 2017. The incidence of 7-day infection hospitalization varied from 37,140 (3.3%) following inpatient discharge, 50,315 (1.1%) following emergency department discharge, and 277,034 (0.3%) following ambulatory visits. The incidence of 7-day infection hospitalization was increased for inpatient discharges with high readmission risk (10.0%), emergency department discharges with increased acute or chronic severity of illness (3.5% and 4.7%, respectively), and ambulatory visits with acute infection (0.7%). The timing of 7-day infection and sepsis hospitalizations differed across settings with an early rise following ambulatory visits, a later peak following emergency department discharges, and a delayed peak following inpatient discharge. Theoretical number needed to evaluate values varied by strata, but following hospital and emergency department discharge, were as low as 15 to 25.

CONCLUSIONS

Incident 7-day infection and sepsis hospitalizations following encounters in routine healthcare settings were surprisingly common and may be amenable to clinical predictive models.

Comparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study

BACKGROUND

Machine learning (ML) algorithms are now increasingly used in infectious disease epidemiology. Epidemiologists should understand how ML algorithms behave within the context of outbreak data where missingness of data is almost ubiquitous.

METHODS

Using simulated data, we use a ML algorithmic framework to evaluate data imputation performance and the resulting case fatality ratio (CFR) estimates, focusing on the scale and type of data missingness (i.e., missing completely at random-MCAR, missing at random-MAR, or missing not at random-MNAR).

RESULTS

Across ML methods, dataset sizes and proportions of training data used, the area under the receiver operating characteristic curve decreased by 7% (median, range: 1%-16%) when missingness was increased from 10% to 40%. Overall reduction in CFR bias for MAR across methods, proportion of missingness, outbreak size and proportion of training data was 0.5% (median, range: 0%-11%).

CONCLUSION

ML methods could reduce bias and increase the precision in CFR estimates at low levels of missingness. However, no method is robust to high percentages of missingness. Thus, a datacentric approach is recommended in outbreak settings-patient survival outcome data should be prioritised for collection and random-sample follow-ups should be implemented to ascertain missing outcomes.

Sepsis Subclasses: A Framework for Development and Interpretation

Sepsis is defined as a dysregulated host response to infection that leads to life-threatening acute organ dysfunction. It afflicts approximately 50 million people worldwide annually and is often deadly, even when evidence-based guidelines are applied promptly. Many randomized trials tested therapies for sepsis over the past 2 decades, but most have not proven beneficial. This may be because sepsis is a heterogeneous syndrome, characterized by a vast set of clinical and biologic features. Combinations of these features, however, may identify previously unrecognized groups, or "subclasses" with different risks of outcome and response to a given treatment. As efforts to identify sepsis subclasses become more common, many unanswered questions and challenges arise. These include: 1) the semantic underpinning of sepsis subclasses, 2) the conceptual goal of subclasses, 3) considerations about study design, data sources, and statistical methods, 4) the role of emerging data types, and 5) how to determine whether subclasses represent "truth." We discuss these challenges and present a framework for the broader study of sepsis subclasses. This framework is intended to aid in the understanding and interpretation of sepsis subclasses, provide a mechanism for explaining subclasses generated by different methodologic approaches, and guide clinicians in how to consider subclasses in bedside care.

Simulation-derived best practices for clustering clinical data

INTRODUCTION

Clustering analyses in clinical contexts hold promise to improve the understanding of patient phenotype and disease course in chronic and acute clinical medicine. However, work remains to ensure that solutions are rigorous, valid, and reproducible. In this paper, we evaluate best practices for dissimilarity matrix calculation and clustering on mixed-type, clinical data.

METHODS

We simulate clinical data to represent problems in clinical trials, cohort studies, and EHR data, including single-type datasets (binary, continuous, categorical) and 4 data mixtures. We test 5 single distance metrics (Jaccard, Hamming, Gower, Manhattan, Euclidean) and 3 mixed distance metrics (DAISY, Supersom, and Mercator) with 3 clustering algorithms (hierarchical (HC), k-medoids, self-organizing maps (SOM)). We quantitatively and visually validate by Adjusted Rand Index (ARI) and silhouette width (SW). We applied our best methods to two real-world data sets: (1) 21 features collected on 247 patients with chronic lymphocytic leukemia, and (2) 40 features collected on 6000 patients admitted to an intensive care unit.

RESULTS

HC outperformed k-medoids and SOM by ARI across data types. DAISY produced the highest mean ARI for mixed data types for all mixtures except unbalanced mixtures dominated by continuous data. Compared to other methods, DAISY with HC uncovered superior, separable clusters in both real-world data sets.

DISCUSSION

Selecting an appropriate mixed-type metric allows the investigator to obtain optimal separation of patient clusters and get maximum use of their data. Superior metrics for mixed-type data handle multiple data types using multiple, type-focused distances. Better subclassification of disease opens avenues for targeted treatments, precision medicine, clinical decision support, and improved patient outcomes.

A novel model to label delirium in an intensive care unit from clinician actions

BACKGROUND

In the intensive care unit (ICU), delirium is a common, acute, confusional state associated with high risk for short- and long-term morbidity and mortality. Machine learning (ML) has promise to address research priorities and improve delirium outcomes. However, due to clinical and billing conventions, delirium is often inconsistently or incompletely labeled in electronic health record (EHR) datasets. Here, we identify clinical actions abstracted from clinical guidelines in electronic health records (EHR) data that indicate risk of delirium among intensive care unit (ICU) patients. We develop a novel prediction model to label patients with delirium based on a large data set and assess model performance.

METHODS

EHR data on 48,451 admissions from 2001 to 2012, available through Medical Information Mart for Intensive Care-III database (MIMIC-III), was used to identify features to develop our prediction models. Five binary ML classification models (Logistic Regression; Classification and Regression Trees; Random Forests; Naïve Bayes; and Support Vector Machines) were fit and ranked by Area Under the Curve (AUC) scores. We compared our best model with two models previously proposed in the literature for goodness of fit, precision, and through biological validation.

RESULTS

Our best performing model with threshold reclassification for predicting delirium was based on a multiple logistic regression using the 31 clinical actions (AUC 0.83). Our model out performed other proposed models by biological validation on clinically meaningful, delirium-associated outcomes.

CONCLUSIONS

Hurdles in identifying accurate labels in large-scale datasets limit clinical applications of ML in delirium. We developed a novel labeling model for delirium in the ICU using a large, public data set. By using guideline-directed clinical actions independent from risk factors, treatments, and outcomes as model predictors, our classifier could be used as a delirium label for future clinically targeted models.

The Presentation, Pace, and Profile of Infection and Sepsis Patients Hospitalized Through the Emergency Department: An Exploratory Analysis

To characterize the signs and symptoms of sepsis, compare them with those from simple infection and other emergent conditions and evaluate their association with hospital outcomes.

Design Setting Participants and INTERVENTION

A multicenter, retrospective cohort study of 408,377 patients hospitalized through the emergency department from 2012 to 2017 with sepsis, suspected infection, heart failure, or stroke. Infected patients were identified based on Sepsis-3 criteria, whereas noninfected patients were identified through diagnosis codes.

MEASUREMENTS AND MAIN RESULTS

Signs and symptoms were identified within physician clinical documentation in the first 24 hours of hospitalization using natural language processing. The time of sign and symptom onset prior to presentation was quantified, and sign and symptom prevalence was assessed. Using multivariable logistic regression, the association of each sign and symptom with four outcomes was evaluated: sepsis versus suspected infection diagnosis, hospital mortality, ICU admission, and time of first antibiotics (> 3 vs ≤ 3 hr from presentation). A total of 10,825 signs and symptoms were identified in 6,148,348 clinical documentation fragments. The most common symptoms overall were as follows: dyspnea (35.2%), weakness (27.2%), altered mental status (24.3%), pain (23.9%), cough (19.7%), edema (17.8%), nausea (16.9%), hypertension (15.6%), fever (13.9%), and chest pain (12.1%). Compared with predominant signs and symptoms in heart failure and stroke, those present in infection were heterogeneous. Signs and symptoms indicative of neurologic dysfunction, significant respiratory conditions, and hypotension were strongly associated with sepsis diagnosis, hospital mortality, and intensive care. Fever, present in only a minority of patients, was associated with improved mortality (odds ratio, 0.67, 95% CI, 0.64-0.70; p < 0.001). For common symptoms, the peak time of symptom onset before sepsis was 2 days, except for altered mental status, which peaked at 1 day prior to presentation.

Conclusions

The clinical presentation of sepsis was heterogeneous and occurred with rapid onset prior to hospital presentation. These findings have important implications for improving public education, clinical treatment, and quality measures of sepsis care.

Prediction of Sepsis in COVID-19 Using Laboratory Indicators

Background

The outbreak of coronavirus disease 2019 (COVID-19) has become a global public health concern. Many inpatients with COVID-19 have shown clinical symptoms related to sepsis, which will aggravate the deterioration of patients’ condition. We aim to diagnose Viral Sepsis Caused by SARS-CoV-2 by analyzing laboratory test data of patients with COVID-19 and establish an early predictive model for sepsis risk among patients with COVID-19.

Methods

This study retrospectively investigated laboratory test data of 2,453 patients with COVID-19 from electronic health records. Extreme gradient boosting (XGBoost) was employed to build four models with different feature subsets of a total of 69 collected indicators. Meanwhile, the explainable Shapley Additive ePlanation (SHAP) method was adopted to interpret predictive results and to analyze the feature importance of risk factors.

Findings

The model for classifying COVID-19 viral sepsis with seven coagulation function indicators achieved the area under the receiver operating characteristic curve (AUC) 0.9213 (95% CI, 89.94–94.31%), sensitivity 97.17% (95% CI, 94.97–98.46%), and specificity 82.05% (95% CI, 77.24–86.06%). The model for identifying COVID-19 coagulation disorders with eight features provided an average of 3.68 (±) 4.60 days in advance for early warning prediction with 0.9298 AUC (95% CI, 86.91–99.04%), 82.22% sensitivity (95% CI, 67.41–91.49%), and 84.00% specificity (95% CI, 63.08–94.75%).

Interpretation

We found that an abnormality of the coagulation function was related to the occurrence of sepsis and the other routine laboratory test represented by inflammatory factors had a moderate predictive value on coagulopathy, which indicated that early warning of sepsis in COVID-19 patients could be achieved by our established model to improve the patient’s prognosis and to reduce mortality.

Nonstationary multivariate Gaussian processes for electronic health records

Advances in the modeling and analysis of electronic health records (EHR) have the potential to improve patient risk stratification, leading to better patient outcomes. The modeling of complex temporal relations across the multiple clinical variables inherent in EHR data is largely unexplored. Existing approaches to modeling EHR data often lack the flexibility to handle time-varying correlations across multiple clinical variables, or they are too complex for clinical interpretation. Therefore, we propose a novel nonstationary multivariate Gaussian process model for EHR data to address the aforementioned drawbacks of existing methodologies. Our proposed model is able to capture time-varying scale, correlation and smoothness across multiple clinical variables. We also provide details on two inference approaches: Maximum a posteriori and Hamilton Monte Carlo. Our model is validated on synthetic data and then we demonstrate its effectiveness on EHR data from Kaiser Permanente Division of Research (KPDOR). Finally, we use the KPDOR EHR data to investigate the relationships between a clinical patient risk metric and the latent processes of our proposed model and demonstrate statistically significant correlations between these entities.

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.

Identifying subpopulations of septic patients: A temporal data-driven approach

Sepsis is one of the deadliest diseases in North America and in spite of the vast amount of research on this topic there is still uncertainty in the outcome of sepsis treatments. This study aimed at investigating the informativeness of temporal electronic health records (EHR) in stratifying septic patients and identifying subpopulations of septic patients with similar trajectories and clinical needs. We performed hierarchical clustering and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) analyses using data from septic patients in the MIMIC III intensive care unit database. The t-Distributed Stochastic Neighbor Embedding (t-SNE) method was utilized to map patients to a two-dimensional space. We utilized silhouette index and cluster-wise stability assessment by resampling to investigate the validity of the clusters. The hierarchical clustering with Euclidean metric identified twelve clinically recognizable subgroups that demonstrated different characteristics in spite of sharing common conditions. Our results demonstrated that data-driven approaches can help in customizing care platforms for septic patients by identifying similar clinically relevant groups.

Sepsis in the critically ill patient: current and emerging management strategies

Introduction: Sepsis, a dysregulated host response to infection, is a major cause of morbidity and mortality worldwide. Early identification and evidence-based treatment of sepsis are associated with improved outcomes.Areas covered: This narrative review was undertaken following a PubMed search for English language reports published before July 2020 using the terms 'sepsis,' 'septic shock,' 'fluids,' 'fluid therapy,' 'albumin,' 'corticosteroids,' 'vasopressor.' Emerging management strategies were identified following a search of the ClinicalTrails.gov database using the term 'sepsis.' Additional reports were identified by examining the reference lists of selected articles and based on personnel knowledge of the field of sepsis.Expert opinion: The core treatment of sepsis relies on source control, early antibiotics, and organ support. The main emerging strategies focus on immunomodulation, artificial intelligence, and on multi-omics approaches for a personalized therapy.

Machine Learning and Statistical Models to Predict Postpartum Hemorrhage

OBJECTIVE

To predict a woman's risk of postpartum hemorrhage at labor admission using machine learning and statistical models.

METHODS

Predictive models were constructed and compared using data from 10 of 12 sites in the U.S. Consortium for Safe Labor Study (2002-2008) that consistently reported estimated blood loss at delivery. The outcome was postpartum hemorrhage, defined as an estimated blood loss at least 1,000 mL. Fifty-five candidate risk factors routinely available on labor admission were considered. We used logistic regression with and without lasso regularization (lasso regression) as the two statistical models, and random forest and extreme gradient boosting as the two machine learning models to predict postpartum hemorrhage. Model performance was measured by C statistics (ie, concordance index), calibration, and decision curves. Models were constructed from the first phase (2002-2006) and externally validated (ie, temporally) in the second phase (2007-2008). Further validation was performed combining both temporal and site-specific validation.

RESULTS

Of the 152,279 assessed births, 7,279 (4.8%, 95% CI 4.7-4.9) had postpartum hemorrhage. All models had good-to-excellent discrimination. The extreme gradient boosting model had the best discriminative ability to predict postpartum hemorrhage (C statistic: 0.93; 95% CI 0.92-0.93), followed by random forest (C statistic: 0.92; 95% CI 0.91-0.92). The lasso regression model (C statistic: 0.87; 95% CI 0.86-0.88) and logistic regression (C statistic: 0.87; 95% CI 0.86-0.87) had lower-but-good discriminative ability. The above results held with validation across both time and sites. Decision curve analysis demonstrated that, although all models provided superior net benefit when clinical decision thresholds were between 0% and 80% predicted risk, the extreme gradient boosting model provided the greatest net benefit.

CONCLUSION

Postpartum hemorrhage on labor admission can be predicted with excellent discriminative ability using machine learning and statistical models. Further clinical application is needed, which may assist health care providers to be prepared and triage at-risk women.

Unsupervised machine learning and prognostic factors of survival in chronic lymphocytic leukemia

OBJECTIVE

Unsupervised machine learning approaches hold promise for large-scale clinical data. However, the heterogeneity of clinical data raises new methodological challenges in feature selection, choosing a distance metric that captures biological meaning, and visualization. We hypothesized that clustering could discover prognostic groups from patients with chronic lymphocytic leukemia, a disease that provides biological validation through well-understood outcomes.

METHODS

To address this challenge, we applied k-medoids clustering with 10 distance metrics to 2 experiments ("A" and "B") with mixed clinical features collapsed to binary vectors and visualized with both multidimensional scaling and t-stochastic neighbor embedding. To assess prognostic utility, we performed survival analysis using a Cox proportional hazard model, log-rank test, and Kaplan-Meier curves.

RESULTS

In both experiments, survival analysis revealed a statistically significant association between clusters and survival outcomes (A: overall survival, P = .0164; B: time from diagnosis to treatment, P = .0039). Multidimensional scaling separated clusters along a gradient mirroring the order of overall survival. Longer survival was associated with mutated immunoglobulin heavy-chain variable region gene (IGHV) status, absent Zap 70 expression, female sex, and younger age.

CONCLUSIONS

This approach to mixed-type data handling and selection of distance metric captured well-understood, binary, prognostic markers in chronic lymphocytic leukemia (sex, IGHV mutation status, ZAP70 expression status) with high fidelity.

Early Detection of Heart Failure With Reduced Ejection Fraction Using Perioperative Data Among Noncardiac Surgical Patients: A Machine-Learning Approach

BACKGROUND

Heart failure with reduced ejection fraction (HFrEF) is a condition imposing significant health care burden. Given its syndromic nature and often insidious onset, the diagnosis may not be made until clinical manifestations prompt further evaluation. Detecting HFrEF in precursor stages could allow for early initiation of treatments to modify disease progression. Granular data collected during the perioperative period may represent an underutilized method for improving the diagnosis of HFrEF. We hypothesized that patients ultimately diagnosed with HFrEF following surgery can be identified via machine-learning approaches using pre- and intraoperative data.

METHODS

Perioperative data were reviewed from adult patients undergoing general anesthesia for major surgical procedures at an academic quaternary care center between 2010 and 2016. Patients with known HFrEF, heart failure with preserved ejection fraction, preoperative critical illness, or undergoing cardiac, cardiology, or electrophysiologic procedures were excluded. Patients were classified as healthy controls or undiagnosed HFrEF. Undiagnosed HFrEF was defined as lacking a HFrEF diagnosis preoperatively but establishing a diagnosis within 730 days postoperatively. Undiagnosed HFrEF patients were adjudicated by expert clinician review, excluding cases for which HFrEF was secondary to a perioperative triggering event, or any event not associated with HFrEF natural disease progression. Machine-learning models, including L1 regularized logistic regression, random forest, and extreme gradient boosting were developed to detect undiagnosed HFrEF, using perioperative data including 628 preoperative and 1195 intraoperative features. Training/validation and test datasets were used with parameter tuning. Test set model performance was evaluated using area under the receiver operating characteristic curve (AUROC), positive predictive value, and other standard metrics.

RESULTS

Among 67,697 cases analyzed, 279 (0.41%) patients had undiagnosed HFrEF. The AUROC for the logistic regression model was 0.869 (95% confidence interval, 0.829-0.911), 0.872 (0.836-0.909) for the random forest model, and 0.873 (0.833-0.913) for the extreme gradient boosting model. The corresponding positive predictive values were 1.69% (1.06%-2.32%), 1.42% (0.85%-1.98%), and 1.78% (1.15%-2.40%), respectively.

CONCLUSIONS

Machine-learning models leveraging perioperative data can detect undiagnosed HFrEF with good performance. However, the low prevalence of the disease results in a low positive predictive value, and for clinically meaningful sensitivity thresholds to be actionable, confirmatory testing with high specificity (eg, echocardiography or cardiac biomarkers) would be required following model detection. Future studies are necessary to externally validate algorithm performance at additional centers and explore the feasibility of embedding algorithms into the perioperative electronic health record for clinician use in real time.

Machine learning in infection management using routine electronic health records: tools, techniques, and reporting of future technologies

BACKGROUND

Machine learning (ML) is increasingly being used in many areas of health care. Its use in infection management is catching up as identified in a recent review in this journal. We present here a complementary review to this work.

OBJECTIVES

To support clinicians and researchers in navigating through the methodological aspects of ML approaches in the field of infection management.

SOURCES

A Medline search was performed with the keywords artificial intelligence, machine learning, infection∗, and infectious disease∗ for the years 2014-2019. Studies using routinely available electronic hospital record data from an inpatient setting with a focus on bacterial and fungal infections were included.

CONTENT

Fifty-two studies were included and divided into six groups based on their focus. These studies covered detection/prediction of sepsis (n = 19), hospital-acquired infections (n = 11), surgical site infections and other postoperative infections (n = 11), microbiological test results (n = 4), infections in general (n = 2), musculoskeletal infections (n = 2), and other topics (urinary tract infections, deep fungal infections, antimicrobial prescriptions; n = 1 each). In total, 35 different ML techniques were used. Logistic regression was applied in 18 studies followed by random forest, support vector machines, and artificial neural networks in 18, 12, and seven studies, respectively. Overall, the studies were very heterogeneous in their approach and their reporting. Detailed information on data handling and software code was often missing. Validation on new datasets and/or in other institutions was rarely done. Clinical studies on the impact of ML in infection management were lacking.

IMPLICATIONS

Promising approaches for ML use in infectious diseases were identified. But building trust in these new technologies will require improved reporting. Explainability and interpretability of the models used were rarely addressed and should be further explored. Independent model validation and clinical studies evaluating the added value of ML approaches are needed.