Ensemble of deep learning language models to support the creation of living systematic reviews for the COVID-19 literature.

BACKGROUND: The COVID-19 pandemic has led to an unprecedented amount of scientific publications, growing at a pace never seen before. Multiple living systematic reviews have been developed to assist professionals with up-to-date and trustworthy health information, but it is increasingly challenging for systematic reviewers to keep up with the evidence in electronic databases. We aimed to investigate deep learning-based machine learning algorithms to classify COVID-19-related publications to help scale up the epidemiological curation process. METHODS: In this retrospective study, five different pre-trained deep learning-based language models were fine-tuned on a dataset of 6365 publications manually classified into two classes, three subclasses, and 22 sub-subclasses relevant for epidemiological triage purposes. In a k-fold cross-validation setting, each standalone model was assessed on a classification task and compared against an ensemble, which takes the standalone model predictions as input and uses different strategies to infer the optimal article class. A ranking task was also considered, in which the model outputs a ranked list of sub-subclasses associated with the article. RESULTS: The ensemble model significantly outperformed the standalone classifiers, achieving a F1-score of 89.2 at the class level of the classification task. The difference between the standalone and ensemble models increases at the sub-subclass level, where the ensemble reaches a micro F1-score of 70% against 67% for the best-performing standalone model. For the ranking task, the ensemble obtained the highest recall@3, with a performance of 89%. Using an unanimity voting rule, the ensemble can provide predictions with higher confidence on a subset of the data, achieving detection of original papers with a F1-score up to 97% on a subset of 80% of the collection instead of 93% on the whole dataset. CONCLUSION: This study shows the potential of using deep learning language models to perform triage of COVID-19 references efficiently and support epidemiological curation and review. The ensemble consistently and significantly outperforms any standalone model. Fine-tuning the voting strategy thresholds is an interesting alternative to annotate a subset with higher predictive confidence.

Deep learning-based risk prediction for interventional clinical trials based on protocol design: A retrospective study.

Success rate of clinical trials (CTs) is low, with the protocol design itself being considered a major risk factor. We aimed to investigate the use of deep learning methods to predict the risk of CTs based on their protocols. Considering protocol changes and their final status, a retrospective risk assignment method was proposed to label CTs according to low, medium, and high risk levels. Then, transformer and graph neural networks were designed and combined in an ensemble model to learn to infer the ternary risk categories. The ensemble model achieved robust performance (area under the receiving operator characteristic curve [AUROC] of 0.8453 [95% confidence interval: 0.8409-0.8495]), similar to the individual architectures but significantly outperforming a baseline based on bag-of-words features (0.7548 [0.7493-0.7603] AUROC). We demonstrate the potential of deep learning in predicting the risk of CTs from their protocols, paving the way for customized risk mitigation strategies during protocol design.

Reducing systematic review burden using Deduklick: a novel, automated, reliable, and explainable deduplication algorithm to foster medical research.

BACKGROUND: Identifying and removing reference duplicates when conducting systematic reviews (SRs) remain a major, time-consuming issue for authors who manually check for duplicates using built-in features in citation managers. To address issues related to manual deduplication, we developed an automated, efficient, and rapid artificial intelligence-based algorithm named Deduklick. Deduklick combines natural language processing algorithms with a set of rules created by expert information specialists. METHODS: Deduklick's deduplication uses a multistep algorithm of data normalization, calculates a similarity score, and identifies unique and duplicate references based on metadata fields, such as title, authors, journal, DOI, year, issue, volume, and page number range. We measured and compared Deduklick's capacity to accurately detect duplicates with the information specialists' standard, manual duplicate removal process using EndNote on eight existing heterogeneous datasets. Using a sensitivity analysis, we manually cross-compared the efficiency and noise of both methods. DISCUSSION: Deduklick achieved average recall of 99.51%, average precision of 100.00%, and average F1 score of 99.75%. In contrast, the manual deduplication process achieved average recall of 88.65%, average precision of 99.95%, and average F1 score of 91.98%. Deduklick achieved equal to higher expert-level performance on duplicate removal. It also preserved high metadata quality and drastically reduced time spent on analysis. Deduklick represents an efficient, transparent, ergonomic, and time-saving solution for identifying and removing duplicates in SRs searches. Deduklick could therefore simplify SRs production and represent important advantages for scientists, including saving time, increasing accuracy, reducing costs, and contributing to quality SRs.

Vaccine Development in the Time of COVID-19: The Relevance of the Risklick AI to Assist in Risk Assessment and Optimize Performance.

The 2019 coronavirus (COVID-19) pandemic revealed the urgent need for the acceleration of vaccine development worldwide. Rapid vaccine development poses numerous risks for each category of vaccine technology. By using the Risklick artificial intelligence (AI), we estimated the risks associated with all types of COVID-19 vaccine during the early phase of vaccine development. We then performed a postmortem analysis of the probability and the impact matrix calculations by comparing the 2020 prognosis to the contemporary situation. We used the Risklick AI to evaluate the risks and their incidence associated with vaccine development in the early stage of the COVID-19 pandemic. Our analysis revealed the diversity of risks among vaccine technologies currently used by pharmaceutical companies providing vaccines. This analysis highlighted the current and future potential pitfalls connected to vaccine production during the COVID-19 pandemic. Hence, the Risklick AI appears as an essential tool in vaccine development for the treatment of COVID-19 in order to formally anticipate the risks, and increases the overall performance from the production to the distribution of the vaccines. The Risklick AI could, therefore, be extended to other fields of research and development and represent a novel opportunity in the calculation of production-associated risks.

Information Retrieval in an Infodemic: The Case of COVID-19 Publications.

BACKGROUND: The COVID-19 global health crisis has led to an exponential surge in published scientific literature. In an attempt to tackle the pandemic, extremely large COVID-19-related corpora are being created, sometimes with inaccurate information, which is no longer at scale of human analyses. OBJECTIVE: In the context of searching for scientific evidence in the deluge of COVID-19-related literature, we present an information retrieval methodology for effective identification of relevant sources to answer biomedical queries posed using natural language. METHODS: Our multistage retrieval methodology combines probabilistic weighting models and reranking algorithms based on deep neural architectures to boost the ranking of relevant documents. Similarity of COVID-19 queries is compared to documents, and a series of postprocessing methods is applied to the initial ranking list to improve the match between the query and the biomedical information source and boost the position of relevant documents. RESULTS: The methodology was evaluated in the context of the TREC-COVID challenge, achieving competitive results with the top-ranking teams participating in the competition. Particularly, the combination of bag-of-words and deep neural language models significantly outperformed an Okapi Best Match 25-based baseline, retrieving on average, 83% of relevant documents in the top 20. CONCLUSIONS: These results indicate that multistage retrieval supported by deep learning could enhance identification of literature for COVID-19-related questions posed using natural language.

Utilizing Artificial Intelligence to Manage COVID-19 Scientific Evidence Torrent with Risklick AI: A Critical Tool for Pharmacology and Therapy Development.

INTRODUCTION: The SARS-CoV-2 pandemic has led to one of the most critical and boundless waves of publications in the history of modern science. The necessity to find and pursue relevant information and quantify its quality is broadly acknowledged. Modern information retrieval techniques combined with artificial intelligence (AI) appear as one of the key strategies for COVID-19 living evidence management. Nevertheless, most AI projects that retrieve COVID-19 literature still require manual tasks. METHODS: In this context, we pre-sent a novel, automated search platform, called Risklick AI, which aims to automatically gather COVID-19 scientific evidence and enables scientists, policy makers, and healthcare professionals to find the most relevant information tailored to their question of interest in real time. RESULTS: Here, we compare the capacity of Risklick AI to find COVID-19-related clinical trials and scientific publications in comparison with clinicaltrials.gov and PubMed in the field of pharmacology and clinical intervention. DISCUSSION: The results demonstrate that Risklick AI is able to find COVID-19 references more effectively, both in terms of precision and recall, compared to the baseline platforms. Hence, Risklick AI could become a useful alternative assistant to scientists fighting the COVID-19 pandemic.

Evolocumab for Early Reduction of LDL Cholesterol Levels in Patients With Acute Coronary Syndromes (EVOPACS).

BACKGROUND: Although guidelines recommend in-hospital initiation of high-intensity statin therapy in patients with acute coronary syndromes (ACS), low-density lipoprotein cholesterol (LDL-C) target levels are frequently not attained. Evolocumab, a rapidly acting, potent LDL-C-lowering drug, has not been studied in the acute phase of ACS. OBJECTIVES: The purpose of this study was to assess the feasibility, safety, and LDL-C-lowering efficacy of evolocumab initiated during the in-hospital phase of ACS. METHODS: The authors conducted an investigator-initiated, randomized, double-blind, placebo-controlled trial involving 308 patients hospitalized for ACS with elevated LDL-C levels (≥1.8 mmol/l on high-intensity statin for at least 4 weeks; ≥2.3 mmol/l on low- or moderate-intensity statin; or ≥3.2 mmol/l on no stable dose of statin). Patients were randomly assigned 1:1 to receive subcutaneous evolocumab 420 mg or matching placebo, administered in-hospital and after 4 weeks, on top of atorvastatin 40 mg. The primary endpoint was percentage change in calculated LDL-C from baseline to 8 weeks. RESULTS: Most patients (78.2%) had not been on previous statin treatment. Mean LDL-C levels decreased from 3.61 to 0.79 mmol/l at week 8 in the evolocumab group, and from 3.42 to 2.06 mmol/l in the placebo group; the difference in mean percentage change from baseline was -40.7% (95% confidence interval: -45.2 to -36.2; p < 0.001). LDL-C levels <1.8 mmol/l were achieved at week 8 by 95.7% of patients in the evolocumab group versus 37.6% in the placebo group. Adverse events and centrally adjudicated cardiovascular events were similar in both groups. CONCLUSIONS: In this first randomized trial assessing a PCSK9 antibody in the very high-risk setting of ACS, evolocumab added to high-intensity statin therapy was well tolerated and resulted in substantial reduction in LDL-C levels, rendering >95% of patients within currently recommended target levels. (EVOlocumab for Early Reduction of LDL-cholesterol Levels in Patients With Acute Coronary Syndromes [EVOPACS]; NCT03287609).

To NET or not to NET:current opinions and state of the science regarding the formation of neutrophil extracellular traps.

Since the discovery and definition of neutrophil extracellular traps (NETs) 14 years ago, numerous characteristics and physiological functions of NETs have been uncovered. Nowadays, the field continues to expand and novel mechanisms that orchestrate formation of NETs, their previously unknown properties, and novel implications in disease continue to emerge. The abundance of available data has also led to some confusion in the NET research community due to contradictory results and divergent scientific concepts, such as pro- and anti-inflammatory roles in pathologic conditions, demarcation from other forms of cell death, or the origin of the DNA that forms the NET scaffold. Here, we present prevailing concepts and state of the science in NET-related research and elaborate on open questions and areas of dispute.

Neutrophil extracellular trap formation requires OPA1-dependent glycolytic ATP production.

Optic atrophy 1 (OPA1) is a mitochondrial inner membrane protein that has an important role in mitochondrial fusion and structural integrity. Dysfunctional OPA1 mutations cause atrophy of the optic nerve leading to blindness. Here, we show that OPA1 has an important role in the innate immune system. Using conditional knockout mice lacking Opa1 in neutrophils (Opa1N∆), we report that lack of OPA1 reduces the activity of mitochondrial electron transport complex I in neutrophils. This then causes a decline in adenosine-triphosphate (ATP) production through glycolysis due to lowered NAD+ availability. Additionally, we show that OPA1-dependent ATP production in these cells is required for microtubule network assembly and for the formation of neutrophil extracellular traps. Finally, we show that Opa1N∆ mice exhibit a reduced antibacterial defense capability against Pseudomonas aeruginosa.

Targeting anticoagulant protein S to improve hemostasis in hemophilia.

Improved treatments are needed for hemophilia A and B, bleeding disorders affecting 400 000 people worldwide. We investigated whether targeting protein S could promote hemostasis in hemophilia by rebalancing coagulation. Protein S (PS) is an anticoagulant acting as cofactor for activated protein C and tissue factor pathway inhibitor (TFPI). This dual role makes PS a key regulator of thrombin generation. Here, we report that targeting PS rebalances coagulation in hemophilia. PS gene targeting in hemophilic mice protected them against bleeding, especially when intra-articular. Mechanistically, these mice displayed increased thrombin generation, resistance to activated protein C and TFPI, and improved fibrin network. Blocking PS in plasma of hemophilia patients normalized in vitro thrombin generation. Both PS and TFPIα were detected in hemophilic mice joints. PS and TFPI expression was stronger in the joints of hemophilia A patients than in those of hemophilia B patients when receiving on-demand therapy, for example, during a bleeding episode. In contrast, PS and TFPI expression was decreased in hemophilia A patients receiving prophylaxis with coagulation factor concentrates, comparable to osteoarthritis patients. These results establish PS inhibition as both controller of coagulation and potential therapeutic target in hemophilia. The murine PS silencing RNA approach that we successfully used in hemophilic mice might constitute a new therapeutic concept for hemophilic patients.

ROS and glutathionylation balance cytoskeletal dynamics in neutrophil extracellular trap formation.

The antimicrobial defense activity of neutrophils partly depends on their ability to form neutrophil extracellular traps (NETs), but the underlying mechanism controlling NET formation remains unclear. We demonstrate that inhibiting cytoskeletal dynamics with pharmacological agents or by genetic manipulation prevents the degranulation of neutrophils and mitochondrial DNA release required for NET formation. Wiskott-Aldrich syndrome protein-deficient neutrophils are unable to polymerize actin and exhibit a block in both degranulation and DNA release. Similarly, neutrophils with a genetic defect in NADPH oxidase fail to induce either actin and tubulin polymerization or NET formation on activation. Moreover, neutrophils deficient in glutaredoxin 1 (Grx1), an enzyme required for deglutathionylation of actin and tubulin, are unable to polymerize either cytoskeletal network and fail to degranulate or release DNA. Collectively, cytoskeletal dynamics are achieved as a balance between reactive oxygen species-regulated effects on polymerization and glutathionylation on the one hand and the Grx1-mediated deglutathionylation that is required for NET formation on the other.

NET formation can occur independently of RIPK3 and MLKL signaling.

The importance of neutrophil extracellular traps (NETs) in innate immunity is well established but the molecular mechanisms responsible for their formation are still a matter of scientific dispute. Here, we aim to characterize a possible role of the receptor-interacting protein kinase 3 (RIPK3) and the mixed lineage kinase domain-like (MLKL) signaling pathway, which are known to cause necroptosis, in NET formation. Using genetic and pharmacological approaches, we investigated whether this programmed form of necrosis is a prerequisite for NET formation. NETs have been defined as extracellular DNA scaffolds associated with the neutrophil granule protein elastase that are capable of killing bacteria. Neither Ripk3-deficient mouse neutrophils nor human neutrophils in which MLKL had been pharmacologically inactivated, exhibited abnormalities in NET formation upon physiological activation or exposure to low concentrations of PMA. These data indicate that NET formation occurs independently of both RIPK3 and MLKL signaling.