scholarly journals InferBERT: A Transformer-Based Causal Inference Framework for Enhancing Pharmacovigilance

2021 ◽  
Vol 4 ◽  
Author(s):  
Xingqiao Wang ◽  
Xiaowei Xu ◽  
Weida Tong ◽  
Ruth Roberts ◽  
Zhichao Liu
Keyword(s):  
Liver Failure ◽  
Acute Liver Failure ◽  
Causal Inference ◽  
Language Processing ◽  
Adverse Event Reporting System ◽  
Adverse Event Reporting ◽  
Language Models ◽  
Data Sets ◽  
Clinical Events ◽  
Wide Range

Background: T ransformer-based language models have delivered clear improvements in a wide range of natural language processing (NLP) tasks. However, those models have a significant limitation; specifically, they cannot infer causality, a prerequisite for deployment in pharmacovigilance, and health care. Therefore, these transformer-based language models should be developed to infer causality to address the key question of the cause of a clinical outcome.Results: In this study, we propose an innovative causal inference model–InferBERT, by integrating the A Lite Bidirectional Encoder Representations from Transformers (ALBERT) and Judea Pearl’s Do-calculus to establish potential causality in pharmacovigilance. Two FDA Adverse Event Reporting System case studies, including Analgesics-related acute liver failure and Tramadol-related mortalities, were employed to evaluate the proposed InferBERT model. The InferBERT model yielded accuracies of 0.78 and 0.95 for identifying Analgesics-related acute liver failure and Tramadol-related death cases, respectively. Meanwhile, the inferred causes of the two clinical outcomes, (i.e. acute liver failure and death) were highly consistent with clinical knowledge. Furthermore, inferred causes were organized into a causal tree using the proposed recursive do-calculus algorithm to improve the model’s understanding of causality. Moreover, the high reproducibility of the proposed InferBERT model was demonstrated by a robustness assessment.Conclusion: The empirical results demonstrated that the proposed InferBERT approach is able to both predict clinical events and to infer their causes. Overall, the proposed InferBERT model is a promising approach to establish causal effects behind text-based observational data to enhance our understanding of intrinsic causality.Availability and implementation: The InferBERT model and preprocessed FAERS data sets are available on GitHub at https://github.com/XingqiaoWang/DeepCausalPV-master.

scholarly journals Systemic quinolones and risk of acute liver failure I: Analysis of data from the US FDA adverse event reporting system

JGH Open ◽  
2021 ◽  
Author(s):  
Mohamed Kadry Taher ◽  
Abdallah Alami ◽  
Christopher A. Gravel ◽  
Derek Tsui ◽  
Lise M. Bjerre ◽  
...  
Keyword(s):  
Adverse Event ◽  
Liver Failure ◽  
Acute Liver Failure ◽  
Reporting System ◽  
Adverse Event Reporting System ◽  
Adverse Event Reporting ◽  
Event Reporting ◽  
The Us ◽  
Us Fda

scholarly journals Crossroad of infection and autoimmunity in acute liver failure: a case report

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Akash Roy ◽  
Sunil Taneja ◽  
Arka De ◽  
Ashim Das ◽  
Ajay K. Duseja ◽  
...  
Keyword(s):  
Liver Failure ◽  
Acute Liver Failure ◽  
Autoimmune Hepatitis ◽  
De Novo ◽  
Dramatic Improvement ◽  
Case Presentation ◽  
Wide Range ◽  
Syndromic Diagnosis ◽  
Initial Improvement ◽  
Tropical Infections

Abstract Background Acute liver failure (ALF) is a syndromic diagnosis, consisting of jaundice, coagulopathy, and any degree of encephalopathy in a patient without pre-existing liver disease within 26 weeks of the onset of symptoms. Autoimmune hepatitis has a wide range of presentations and can rarely present as ALF, which frequently tends to be autoantibody negative. Tropical infections like dengue, malaria, and leptospirosis, which present as mimickers of ALF, always remain a differential diagnosis of ALF and mandate an etiology specific management. In rare cases, such infections themselves act as a trigger for autoimmunity. We report a case of diagnostic crossroads of infection and autoimmunity, presenting as acute liver failure and describe the challenges in management. Case presentation A 25-year-old female presented with a syndromic diagnosis of acute liver failure with possibility of tropical illness-related ALF mimic on account of positive Leptospira serology. After initial improvement, there was a rebound worsening of liver functions which prompted a liver biopsy. Biopsy narrowed the differential to Leptospira-associated hepatitis and severe auto-immune hepatitis. Trial of low dose steroid was given which led to dramatic improvement. Conclusion Tropical infections can present as ALF mimics and can themselves act as triggers for autoimmunity. Distinguishing such cases from de-novo acute severe autoimmune hepatitis is difficult and requires therapeutic brinksmanship. An early trial of steroid is mandated but comes with its own challenges.

scholarly journals Neural Language Models Capture Some, But Not All, Agreement Attraction Effects

2020 ◽  
Author(s):  
Suhas Arehalli ◽  
Tal Linzen
Keyword(s):  
Real Time ◽  
Language Processing ◽  
Language Production ◽  
Prediction Models ◽  
Language Model ◽  
Cognitive Model ◽  
Language Models ◽  
Word Prediction ◽  
Wide Range ◽  
Attraction Effects

The number of the subject in English must match the number of the corresponding verb (dog runs but dogs run). Yet in real-time language production and comprehension, speakers often mistakenly compute agreement between the verb and a grammatically irrelevant non-subject noun phrase instead. This phenomenon, referred to as agreement attraction, is modulated by a wide range of factors; any complete computational model of grammatical planning and comprehension would be expected to derive this rich empirical picture. Recent developments in Natural Language Processing have shown that neural networks trained only on word-prediction over large corpora are capable of capturing subject-verb agreement dependencies to a significant extent, but with occasional errors. The goal of this paper is to evaluate the potential of such neural word prediction models as a foundation for a cognitive model of real-time grammatical processing. We simulate six experiments taken from the agreement attraction literature with LSTMs, one common type of neural language model. The LSTMs captured the critical human behavior in three of them, indicating that (1) some agreement attraction phenomena can be captured by a generic sequence processing model, but (2) capturing the other phenomena may require models with more language-specific mechanisms

scholarly journals Inducing Relational Knowledge from BERT

2020 ◽  
Vol 34 (05) ◽  
pp. 7456-7463 ◽  
Author(s):  
Zied Bouraoui ◽  
Jose Camacho-Collados ◽  
Steven Schockaert
Keyword(s):  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Language Model ◽  
Language Models ◽  
Word Embeddings ◽  
Relational Knowledge ◽  
Wide Range ◽  
Fine Tune ◽  
Standard Word

One of the most remarkable properties of word embeddings is the fact that they capture certain types of semantic and syntactic relationships. Recently, pre-trained language models such as BERT have achieved groundbreaking results across a wide range of Natural Language Processing tasks. However, it is unclear to what extent such models capture relational knowledge beyond what is already captured by standard word embeddings. To explore this question, we propose a methodology for distilling relational knowledge from a pre-trained language model. Starting from a few seed instances of a given relation, we first use a large text corpus to find sentences that are likely to express this relation. We then use a subset of these extracted sentences as templates. Finally, we fine-tune a language model to predict whether a given word pair is likely to be an instance of some relation, when given an instantiated template for that relation as input.

scholarly journals Reverse translation of adverse event reports paves the way for de-risking preclinical off-targets

eLife ◽  
2017 ◽  
Vol 6 ◽  
Author(s):  
Mateusz Maciejewski ◽  
Eugen Lounkine ◽  
Steven Whitebread ◽  
Pierre Farmer ◽  
William DuMouchel ◽  
...  
Keyword(s):  
Adverse Event ◽  
Language Processing ◽  
Molecular Mechanisms ◽  
Adverse Event Reporting System ◽  
Primary Source ◽  
Adverse Event Reporting ◽  
Vegf Receptor ◽  
Chemical Structures ◽  
Post Marketing ◽  
Underlying Mechanisms

The Food and Drug Administration Adverse Event Reporting System (FAERS) remains the primary source for post-marketing pharmacovigilance. The system is largely un-curated, unstandardized, and lacks a method for linking drugs to the chemical structures of their active ingredients, increasing noise and artefactual trends. To address these problems, we mapped drugs to their ingredients and used natural language processing to classify and correlate drug events. Our analysis exposed key idiosyncrasies in FAERS, for example reports of thalidomide causing a deadly ADR when used against myeloma, a likely result of the disease itself; multiplications of the same report, unjustifiably increasing its importance; correlation of reported ADRs with public events, regulatory announcements, and with publications. Comparing the pharmacological, pharmacokinetic, and clinical ADR profiles of methylphenidate, aripiprazole, and risperidone, and of kinase drugs targeting the VEGF receptor, demonstrates how underlying molecular mechanisms can emerge from ADR co-analysis. The precautions and methods we describe may enable investigators to avoid confounding chemistry-based associations and reporting biases in FAERS, and illustrate how comparative analysis of ADRs can reveal underlying mechanisms.

scholarly journals Deep Transfer Learning & Beyond: Transformer Language Models in Information Systems Research

2022 ◽  
Author(s):  
Ross Gruetzemacher ◽  
David Paradice
Keyword(s):  
Text Mining ◽  
Language Processing ◽  
User Interfaces ◽  
Recent Progress ◽  
Language Models ◽  
Is Research ◽  
Systems Research ◽  
Information Systems Research ◽  
Wide Range ◽  
Language User

AI is widely thought to be poised to transform business, yet current perceptions of the scope of this transformation may be myopic. Recent progress in natural language processing involving transformer language models (TLMs) offers a potential avenue for AI-driven business and societal transformation that is beyond the scope of what most currently foresee. We review this recent progress as well as recent literature utilizing text mining in top IS journals to develop an outline for how future IS research can benefit from these new techniques. Our review of existing IS literature reveals that suboptimal text mining techniques are prevalent and that the more advanced TLMs could be applied to enhance and increase IS research involving text data, and to enable new IS research topics, thus creating more value for the research community. This is possible because these techniques make it easier to develop very powerful custom systems and their performance is superior to existing methods for a wide range of tasks and applications. Further, multilingual language models make possible higher quality text analytics for research in multiple languages. We also identify new avenues for IS research, like language user interfaces, that may offer even greater potential for future IS research.

scholarly journals Analysis of Pharmacovigilance Databases for Dolutegravir Safety in Pregnancy

2019 ◽  
Vol 70 (12) ◽  
pp. 2599-2606 ◽  
Author(s):  
Nikolien S van De Ven ◽  
Anton L Pozniak ◽  
Jacob A Levi ◽  
Polly Clayden ◽  
Anna Garratt ◽  
...  
Keyword(s):  
Adverse Event Reporting System ◽  
Adverse Event Reporting ◽  
Antiretroviral Drugs ◽  
Data Availability ◽  
New Drugs ◽  
World Health ◽  
European Medicines Agency ◽  
System Organ Class ◽  
Online Data ◽  
Wide Range

Abstract Background The Botswana Tsepamo study reported neural tube defects (NTDs) in 4 of 426 (0.94%) infants of women receiving preconception dolutegravir (DTG) antiretroviral therapy (ART) vs 14 of 11 300 (0.12%) receiving preconception non-DTG ART. Data are needed to investigate this potential safety signal. Clinicians, patients, and pharmaceutical companies can report adverse drug reactions (ADRs) to pharmacovigilance databases. Data from ADRs reported to various pharmacovigilance databases were searched for NTDs. Methods Four pharmacovigilance databases (World Health Organization [WHO] VigiAccess; United Kingdom Medicines Health Regulatory Authority [UK MHRA]; European Medicines Agency [EMA] EudraVigilance; US Food and Drug Administration Adverse Event Reporting System [FAERS]) with online data availability were analyzed for NTD reports for 4 integrase inhibitors (DTG, raltegravir, elvitegravir, bictegravir), 2 protease inhibitors (darunavir, atazanavir), and 2 nonnucleoside reverse transcriptase inhibitors (nevirapine, efavirenz). Reports in the system organ class “congenital or familial disorders” were searched for NTDs. Results NTDs have been reported among infants born from women taking a wide range of antiretrovirals in 4 pharmacovigilance databases (WHO VigiAccess, 116 reactions; UK MHRA, 8 cases; EMA EudraVigilance, 20 cases; FAERS, 44 cases). Six NTDs were identified for DTG across the pharmacovigilance databases. Cases were very hard to interpret, given the lack of clear denominators. Conclusions Pharmacovigilance databases have many limitations, most importantly lack of a clear denominator for patients exposed to the drug of interest and duplicate cases that are difficult to identify. Given widespread use of new antiretroviral drugs worldwide and anticipated use of new drugs, prospective follow-up of pregnant women and birth surveillance studies such as Tsepamo are critically needed.

scholarly journals Proton-pump inhibitor use is associated with a broad spectrum of neurological adverse events including impaired hearing, vision, and memory

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Tigran Makunts ◽  
Sama Alpatty ◽  
Kelly C. Lee ◽  
Rabia S. Atayee ◽  
Ruben Abagyan
Keyword(s):  
Proton Pump ◽  
Adverse Event Reporting System ◽  
Adverse Event Reporting ◽  
Control Group ◽  
Impaired Hearing ◽  
Alzheimer Dementia ◽  
Stomach Acid ◽  
Wide Range ◽  
United States Food ◽  
Gastric Parietal Cells

AbstractProton-pump inhibitors, PPIs, are considered effective therapy for stomach acid suppression due to their irreversible inhibition of the hydrogen/potassium pump in the gastric parietal cells. They are widely prescribed and are considered safe for over-the-counter use. Recent studies have shown an association between PPI use and Alzheimer dementia, while others have disputed that connection. We analyzed over ten million United States Food and Drug Administration Adverse Event Reporting System reports, including over forty thousand reports containing PPIs, and provided evidence of increased propensity for memory impairment among PPI reports when compared to histamine-2 receptor antagonist control group. Furthermore, we found significant associations of PPI use with a wide range of neurological adverse reactions including, migraine, several peripheral neuropathies, and visual and auditory neurosensory abnormalities.

scholarly journals A New Criterion for Fluoroquinolone-Associated Disability Diagnosis: Functional Gastrointestinal Disorders

Medicina ◽  
2021 ◽  
Vol 57 (12) ◽  
pp. 1371
Author(s):  
Deanna N. Cannizzaro ◽  
Lydia F. Naughton ◽  
Maya Z. Freeman ◽  
Linda Martin ◽  
Charles L. Bennett ◽  
...  
Keyword(s):  
Bacterial Infections ◽  
Neural Control ◽  
Adverse Event Reporting System ◽  
Functional Gastrointestinal Disorders ◽  
Antimicrobial Properties ◽  
Adverse Event Reporting ◽  
Gastrointestinal Disorders ◽  
Gi Symptoms ◽  
Wide Range ◽  
Survey Responses

Background and Objectives: Fluoroquinolones (FQs) are a broad-spectrum class of antibiotics routinely prescribed for common bacterial infections despite recent recommendations to use them only for life-threatening cases. In addition to their antimicrobial properties, FQs act in the central nervous system as GABAA receptor inhibitors, which could potentially affect functionality of the vagus nerve at the forefront of gastrointestinal (GI) tract function. Alterations in neural control of digestion have been shown to be linked to Functional Gastrointestinal Disorders (FGIDs), which are usually diagnosed based on self-reported symptoms. The aim of this study was to assess the incidence of FGIDs following FQ use. Materials and Methods: Self-reports from the FDA Adverse Event Reporting System were analyzed together with ~300 survey responses from a social network derived sample to the Bowel Disease Questionnaire. Results: The results of this study suggested that six different FQs are associated with a wide range of GI symptoms not currently reported in the drugs’ labels. The responses from the survey suggested that ~70% of FQ users scored positive for FGID, with no positive correlation between drug type, duration of administration, dosage and frequency of administration. Conclusions: This study showed that GI disorders other than nausea, vomiting and diarrhea are more common than currently reported on the drug labels, and that FGIDs are possibly a common consequence of FQ use even after single use.

scholarly journals Acute Liver Failure in Rats Induced Vasogenic Edema and Nitric Oxide Synthase-Nitric Oxide-cGMP Pathway after Differential Regulation of NMDA Receptors in Cerebral Cortex and Cerebellum

2022 ◽  
Author(s):  
SANTOSH SINGH ◽  
Arghya Mukherjee ◽  
Deepika Jeswani
Keyword(s):  
Nitric Oxide ◽  
Cerebral Cortex ◽  
Nitric Oxide Synthase ◽  
Liver Failure ◽  
Acute Liver Failure ◽  
Vasogenic Edema ◽  
Differential Regulation ◽  
Guanosine Monophosphate ◽  
Wide Range ◽  
Oxide Synthase

Abstract Acute liver failure (ALF) is a complication of severe liver dysfunction resulting from a wide range of factors including alcoholism, drug-abuse, improper medication, viral hepatitis etc., and present with high mortality rate among the human population. ALF led hyperammonemia (HA) induced cerebral dysfunction is considered to be the main cause of death in patients, however, the precise molecular mechanism is not completely understood. The aim of this study was to investigate the status of brain edema and modulation of N-methyl D-aspartate receptors (NMDAR)- Nitric oxide synthase (NOS)- Nitric oxide (NO)- cyclic guanosine monophosphate (cGMP) axis in the cerebral cortex and cerebellum of ALF rats. ALF was induced by intraperitoneal (IP) injection of thioacetamide (TAA). We observed significantly increased brain water content in ALF rats but absence of astrocytes swelling suggested induction of vasogenic edema. Except constant NR2B, down regulation of NR2A, 2C and 2D subunits containing NMDAR genes in cerebral cortex, however, constant NR2A-C but up-regulation of NR2D subunit in cerebellum suggested brain regions specific differential regulation of NMDAR in ALF rats. Significantly increased nNOS gene and protein level were found to be accompanied by the significantly increased level of NO and cGMP in both brain tissues; however, increased eNOS expression in cortex but increased iNOS expression and activity in cerebellum were observed in ALF rats. Together these findings suggested that ALF in rats may trigger differential regulation of NR2A-D subunits containing NMDAR, induction of NOS-NO-cGMP axis and vasogenic edema in cerebral cortex and cerebellum.

Sign in / Sign up

Export Citation Format

Share Document