Predicting sepsis using deep learning across international sites: a retrospective development and validation study.

Background: When sepsis is detected, organ damage may have progressed to irreversible stages, leading to poor prognosis. The use of machine learning for predicting sepsis early has shown promise, however international validations are missing. Methods: This was a retrospective, observational, multi-centre cohort study. We developed and externally validated a deep learning system for the prediction of sepsis in the intensive care unit (ICU). Our analysis represents the first international, multi-centre in-ICU cohort study for sepsis prediction using deep learning to our knowledge. Our dataset contains 136,478 unique ICU admissions, representing a refined and harmonised subset of four large ICU databases comprising data collected from ICUs in the US, the Netherlands, and Switzerland between 2001 and 2016. Using the international consensus definition Sepsis-3, we derived hourly-resolved sepsis annotations, amounting to 25,694 (18.8%) patient stays with sepsis. We compared our approach to clinical baselines as well as machine learning baselines and performed an extensive internal and external statistical validation within and across databases, reporting area under the receiver-operating-characteristic curve (AUC). Findings: Averaged over sites, our model was able to predict sepsis with an AUC of 0.846 (95% confidence interval [CI], 0.841-0.852) on a held-out validation cohort internal to each site, and an AUC of 0.761 (95% CI, 0.746-0.770) when validating externally across sites. Given access to a small fine-tuning set (10% per site), the transfer to target sites was improved to an AUC of 0.807 (95% CI, 0.801-0.813). Our model raised 1.4 false alerts per true alert and detected 80% of the septic patients 3.7 h (95% CI, 3.0-4.3) prior to the onset of sepsis, opening a vital window for intervention. Interpretation: By monitoring clinical and laboratory measurements in a retrospective simulation of a real-time prediction scenario, a deep learning system for the detection of sepsis generalised to previously unseen ICU cohorts, internationally. Funding: This study was funded by the Personalized Health and Related Technologies (PHRT) strategic focus area of the ETH domain.

Early Prediction of Sepsis in the ICU Using Machine Learning: A Systematic Review.

Background: Sepsis is among the leading causes of death in intensive care units (ICUs) worldwide and its recognition, particularly in the early stages of the disease, remains a medical challenge. The advent of an affluence of available digital health data has created a setting in which machine learning can be used for digital biomarker discovery, with the ultimate goal to advance the early recognition of sepsis. Objective: To systematically review and evaluate studies employing machine learning for the prediction of sepsis in the ICU. Data Sources: Using Embase, Google Scholar, PubMed/Medline, Scopus, and Web of Science, we systematically searched the existing literature for machine learning-driven sepsis onset prediction for patients in the ICU. Study Eligibility Criteria: All peer-reviewed articles using machine learning for the prediction of sepsis onset in adult ICU patients were included. Studies focusing on patient populations outside the ICU were excluded. Study Appraisal and Synthesis Methods: A systematic review was performed according to the PRISMA guidelines. Moreover, a quality assessment of all eligible studies was performed. Results: Out of 974 identified articles, 22 and 21 met the criteria to be included in the systematic review and quality assessment, respectively. A multitude of machine learning algorithms were applied to refine the early prediction of sepsis. The quality of the studies ranged from "poor" (satisfying ≤ 40% of the quality criteria) to "very good" (satisfying ≥ 90% of the quality criteria). The majority of the studies (n = 19, 86.4%) employed an offline training scenario combined with a horizon evaluation, while two studies implemented an online scenario (n = 2, 9.1%). The massive inter-study heterogeneity in terms of model development, sepsis definition, prediction time windows, and outcomes precluded a meta-analysis. Last, only two studies provided publicly accessible source code and data sources fostering reproducibility. Limitations: Articles were only eligible for inclusion when employing machine learning algorithms for the prediction of sepsis onset in the ICU. This restriction led to the exclusion of studies focusing on the prediction of septic shock, sepsis-related mortality, and patient populations outside the ICU. Conclusions and Key Findings: A growing number of studies employs machine learning to optimize the early prediction of sepsis through digital biomarker discovery. This review, however, highlights several shortcomings of the current approaches, including low comparability and reproducibility. Finally, we gather recommendations how these challenges can be addressed before deploying these models in prospective analyses. Systematic Review Registration Number: CRD42020200133.

Topological and kernel-based microbial phenotype prediction from MALDI-TOF mass spectra.

MOTIVATION: Microbial species identification based on matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) has become a standard tool in clinical microbiology. The resulting MALDI-TOF mass spectra also harbour the potential to deliver prediction results for other phenotypes, such as antibiotic resistance. However, the development of machine learning algorithms specifically tailored to MALDI-TOF MS-based phenotype prediction is still in its infancy. Moreover, current spectral pre-processing typically involves a parameter-heavy chain of operations without analyzing their influence on the prediction results. In addition, classification algorithms lack quantification of uncertainty, which is indispensable for predictions potentially influencing patient treatment. RESULTS: We present a novel prediction method for antimicrobial resistance based on MALDI-TOF mass spectra. First, we compare the complex conventional pre-processing to a new approach that exploits topological information and requires only a single parameter, namely the number of peaks of a spectrum to keep. Second, we introduce PIKE, the peak information kernel, a similarity measure specifically tailored to MALDI-TOF mass spectra which, combined with a Gaussian process classifier, provides well-calibrated uncertainty estimates about predictions. We demonstrate the utility of our approach by predicting antibiotic resistance of three clinically highly relevant bacterial species. Our method consistently outperforms competitor approaches, while demonstrating improved performance and security by rejecting out-of-distribution samples, such as bacterial species that are not represented in the training data. Ultimately, our method could contribute to an earlier and precise antimicrobial treatment in clinical patient care. AVAILABILITY AND IMPLEMENTATION: We make our code publicly available as an easy-to-use Python package under https://github.com/BorgwardtLab/maldi_PIKE.

Early prediction of circulatory failure in the intensive care unit using machine learning.

Intensive-care clinicians are presented with large quantities of measurements from multiple monitoring systems. The limited ability of humans to process complex information hinders early recognition of patient deterioration, and high numbers of monitoring alarms lead to alarm fatigue. We used machine learning to develop an early-warning system that integrates measurements from multiple organ systems using a high-resolution database with 240 patient-years of data. It predicts 90% of circulatory-failure events in the test set, with 82% identified more than 2 h in advance, resulting in an area under the receiver operating characteristic curve of 0.94 and an area under the precision-recall curve of 0.63. On average, the system raises 0.05 alarms per patient and hour. The model was externally validated in an independent patient cohort. Our model provides early identification of patients at risk for circulatory failure with a much lower false-alarm rate than conventional threshold-based systems.

Post-Antibiotic Gut Mucosal Microbiome Reconstitution Is Impaired by Probiotics and Improved by Autologous FMT.

Probiotics are widely prescribed for prevention of antibiotics-associated dysbiosis and related adverse effects. However, probiotic impact on post-antibiotic reconstitution of the gut mucosal host-microbiome niche remains elusive. We invasively examined the effects of multi-strain probiotics or autologous fecal microbiome transplantation (aFMT) on post-antibiotic reconstitution of the murine and human mucosal microbiome niche. Contrary to homeostasis, antibiotic perturbation enhanced probiotics colonization in the human mucosa but only mildly improved colonization in mice. Compared to spontaneous post-antibiotic recovery, probiotics induced a markedly delayed and persistently incomplete indigenous stool/mucosal microbiome reconstitution and host transcriptome recovery toward homeostatic configuration, while aFMT induced a rapid and near-complete recovery within days of administration. In vitro, Lactobacillus-secreted soluble factors contributed to probiotics-induced microbiome inhibition. Collectively, potential post-antibiotic probiotic benefits may be offset by a compromised gut mucosal recovery, highlighting a need of developing aFMT or personalized probiotic approaches achieving mucosal protection without compromising microbiome recolonization in the antibiotics-perturbed host.

Antidepressants Rescue Stress-Induced Disruption of Synaptic Plasticity via Serotonin Transporter-Independent Inhibition of L-Type Calcium Channels.

BACKGROUND: Long-term synaptic plasticity is a basic ability of the brain to dynamically adapt to external stimuli and regulate synaptic strength and ultimately network function. It is dysregulated by behavioral stress in animal models of depression and in humans with major depressive disorder. Antidepressants have been shown to restore disrupted synaptic plasticity in both animal models and humans; however, the underlying mechanism is unclear. METHODS: We examined modulation of synaptic plasticity by selective serotonin reuptake inhibitors (SSRIs) in hippocampal brain slices from wild-type rats and serotonin transporter (SERT) knockout mice. Recombinant voltage-gated calcium (Ca2+) channels in heterologous expression systems were used to determine the modulation of Ca2+ channels by SSRIs. We tested the behavioral effects of SSRIs in the chronic behavioral despair model of depression both in the presence and in the absence of SERT. RESULTS: SSRIs selectively inhibited hippocampal long-term depression. The inhibition of long-term depression by SSRIs was mediated by a direct block of voltage-activated L-type Ca2+ channels and was independent of SERT. Furthermore, SSRIs protected both wild-type and SERT knockout mice from behavioral despair induced by chronic stress. Finally, long-term depression was facilitated in animals subjected to the behavioral despair model, which was prevented by SSRI treatment. CONCLUSIONS: These results showed that antidepressants protected synaptic plasticity and neuronal circuitry from the effects of stress via a modulation of Ca2+ channels and synaptic plasticity independent of SERT. Thus, L-type Ca2+ channels might constitute an important signaling hub for stress response and for pathophysiology and treatment of depression.

TRAPP webserver: predicting protein binding site flexibility and detecting transient binding pockets.

The TRAnsient Pockets in Proteins (TRAPP) webserver provides an automated workflow that allows users to explore the dynamics of a protein binding site and to detect pockets or sub-pockets that may transiently open due to protein internal motion. These transient or cryptic sub-pockets may be of interest in the design and optimization of small molecular inhibitors for a protein target of interest. The TRAPP workflow consists of the following three modules: (i) TRAPP structure- generation of an ensemble of structures using one or more of four possible molecular simulation methods; (ii) TRAPP analysis-superposition and clustering of the binding site conformations either in an ensemble of structures generated in step (i) or in PDB structures or trajectories uploaded by the user; and (iii) TRAPP pocket-detection, analysis, and visualization of the binding pocket dynamics and characteristics, such as volume, solvent-exposed area or properties of surrounding residues. A standard sequence conservation score per residue or a differential score per residue, for comparing on- and off-targets, can be calculated and displayed on the binding pocket for an uploaded multiple sequence alignment file, and known protein sequence annotations can be displayed simultaneously. The TRAPP webserver is freely available at http://trapp.h-its.org.

Backbone circularization of Bacillus subtilis family 11 xylanase increases its thermostability and its resistance against aggregation.

The activity of proteins is dictated by their three-dimensional structure, the native state, and is influenced by their ability to remain in or return to the folded native state under physiological conditions. Backbone circularization is thought to increase protein stability by decreasing the conformational entropy in the unfolded state. A positive effect of circularization on stability has been shown for several proteins. Here, we report the development of a cloning standard that facilitates implementing the SICLOPPS technology to circularize proteins of interest using split inteins. To exemplify the usage of the cloning standard we constructed two circularization vectors based on the Npu DnaE and gp41-1 split inteins, respectively. We use these vectors to overexpress in Escherichia coli circular forms of the Bacillus subtilis enzyme family 11 xylanase that differ in the identity and number of additional amino acids used for circularization (exteins). We found that the variant circularized with only one additional serine has increased thermostability of 7 °C compared to native xylanase. The variant circularized with six additional amino acids has only a mild increase in thermostability compared to the corresponding exteins-bearing linear xylanase, but is less stable than native xylanase. However, this circular xylanase retains more than 50% of its activity after heat shock at elevated temperatures, while native xylanase and the corresponding exteins-bearing linear xylanase are largely inactivated. We correlate this residual activity to the fewer protein aggregates found in the test tubes of circular xylanase after heat shock, suggesting that circularization protects the protein from aggregation under these conditions. Taken together, these data indicate that backbone circularization has a positive effect on xylanase and can lead to increased thermostability, provided the appropriate exteins are selected. We believe that our cloning standard and circularization vectors will facilitate testing the effects of circularization on other proteins.

Slowing Amazon deforestation through public policy and interventions in beef and soy supply chains.

The recent 70% decline in deforestation in the Brazilian Amazon suggests that it is possible to manage the advance of a vast agricultural frontier. Enforcement of laws, interventions in soy and beef supply chains, restrictions on access to credit, and expansion of protected areas appear to have contributed to this decline, as did a decline in the demand for new deforestation. The supply chain interventions that fed into this deceleration are precariously dependent on corporate risk management, and public policies have relied excessively on punitive measures. Systems for delivering positive incentives for farmers to forgo deforestation have been designed but not fully implemented. Territorial approaches to deforestation have been effective and could consolidate progress in slowing deforestation while providing a framework for addressing other important dimensions of sustainable development.