Detecting early safety signals of infliximab using machine learning algorithms in the Korea adverse event reporting system

Lee, Jeong Eun; Kim, Woo Jung; Bae, Jihwan; Song, Inmyung; Shin, Ju‐Young

doi:10.1038/s41598-022-18522-z

Cited by 4 publications

(3 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 5 presents the relevant information of the selected storms for the SYM-H index. Particularly, the number assigned to the storm, the start and end dates of the storm (the selected time is from 00:00 of the start date until the start of the end date at 00:00), the SYM-H peak, the category of the storm according to the SYM-H peak following Collado-Villaverde et al (2023a) criteria, the solar cycle of the storm and whether the disturbance was caused by a CME or HSS, according to the CME catalog in Larrodera and Temmer (2023).…”

Section: Storm Sets For Sym-h Indexmentioning

confidence: 99%

“…This approach to partitioning ensured that each set contained a diverse representation of storms, thereby facilitating the comprehensive evaluation of forecasting models across a wide range of geomagnetic storm scenarios. Stratified sampling has been widely used in ML projects when randomly separating the available data can result in sets to not have enough representative samples (Fernandes et al., 2020; Lee et al., 2022). Although the category of each storm is already considered, the range within each category is notably wide.…”

Section: Storms Setsmentioning

confidence: 99%

See 1 more Smart Citation

A Framework for Evaluating Geomagnetic Indices Forecasting Models

Collado‐Villaverde,

Muñoz,

Cid

2024

Space Weather

View full text Add to dashboard Cite

The use of Deep Learning models to forecast geomagnetic storms is achieving great results. However, the evaluation of these models is mainly supported on generic regression metrics (such as the Root Mean Squared Error or the Coefficient of Determination), which are not able to properly capture the specific particularities of geomagnetic storms forecasting. Particularly, they do not provide insights during the high activity periods. To overcome this issue, we introduce the Binned Forecasting Error to provide a more accurate assessment of models' performance across the different intensity levels of a geomagnetic storm. This metric facilitates a robust comparison of different forecasting models, presenting a true representation of a model's predictive capabilities while being resilient to different storms duration. In this direction, for enabling fair comparison among models, it is important to standardize the sets of geomagnetic storms for model training, validation and testing. To do this, we have started from the current sets used in the literature for forecasting the SYM‐H, enriching them with newer storms not considered previously, focusing not only on disturbances caused by Coronal Mass Ejections but also addressing High‐Speed Streams. To operationalize the evaluation framework, a comparative study is conducted between a baseline neural network model and a persistence model, showcasing the effectiveness of the new metric in evaluating forecasting performance during intense geomagnetic storms. Finally, we propose the use of preliminary measurements from ACE to evaluate the model performance in settings closer to an operational real‐time scenario, where the forecasting models are expected to operate.

show abstract

Section: Storm Sets For Sym-h Indexmentioning

confidence: 99%

Section: Storms Setsmentioning

confidence: 99%

A Framework for Evaluating Geomagnetic Indices Forecasting Models

Collado‐Villaverde,

Muñoz,

Cid

2024

Space Weather

View full text Add to dashboard Cite

show abstract

“…Moreover, the emergence of additional data sources such as biochemical databases, electronic health records (EHRs), insurance claims or other "Real-World Data" (RWD) and social media (Hussain, 2021; Knowledge Base workgroup of the Observational Health Data Sciences and Informatics (OHDSI) collaborative, 2017) have led to relevant research initiatives (Natsiavas et al, 2019b;Ball et al, 2022). To this end, Machine Learning (ML) algorithms are also under investigation (Lee et al, 2022;Imran et al, 2022), including the use of Natural Language Processing (NLP) which is elaborated to identify ADR mentions in EHRs/clinical notes or other free text/unstructured data. Other Knowledge Engineering approaches (e.g., the use of Semantic Web technologies, ontologies and "reasoning" upon Knowledge Graphs etc.)…”

Section: Introductionmentioning

confidence: 99%

An open-source platform integrating emerging data sources to support multi-modal active pharmacovigilance

Dimitriadis

Dimitsaki

Chytas

et al. 2023

Front. Drug Saf. Regul.

View full text Add to dashboard Cite

As new data sources have emerged, the data space which Pharmacovigilance (PV) processes can use has significantly expanded. However, still, the currently available tools do not widely exploit data sources beyond Spontaneous Report Systems built to collect Individual Case Safety Reports (ICSRs). This article presents an open-source platform enabling the integration of heterogeneous data sources to support the analysis of drug safety related information. Furthermore, the results of a comparative study as part of the project’s pilot phase are also presented. Data sources were integrated in the form of four “workspaces”: (a) Individual Case Safety Reports—obtained from OpenFDA, (b) Real-World Data (RWD) —using the OMOP-CDM data model, (c) social media data—collected via Twitter, and (d) scientific literature—retrieved from PubMed. Data intensive analytics are built for each workspace (e.g., disproportionality analysis metrics are used for OpenFDA data, descriptive statistics for OMOP-CDM data and twitter data streams etc.). Upon these workspaces, the end-user sets up “investigation scenarios” defined by Drug-Event Combinations (DEC). Specialized features like detailed reporting which could be used to support reports for regulatory purposes and also “quick views” are provided to facilitate use where detailed statistics might not be needed and a qualitative overview of the available information might be enough (e.g., clinical environment). The platform’s technical features are presented as Supplementary Material via a walkthrough of an example “investigation scenario”. The presented platform is evaluated via a comparative study against the EVDAS system, conducted by PV professionals. Results from the comparative study, show that there is indeed a need for relevant technical tools and the ability to draw recent data from heterogeneous data sources is appreciated. However, a reluctance by end-users is also outlined as they feel technical improvements and systematic training are required before the potential adoption of the presented software. As a whole, it is concluded that integrating such a platform in real-world setting is far from trivial, requiring significant effort on training and usability aspects.

show abstract

A real-world pharmacovigilance study on cardiovascular adverse events of tisagenlecleucel using machine learning approach

Jung,

Kim,

Bae

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Chimeric antigen receptor T-cell (CAR-T) therapies are a paradigm-shifting therapeutic in patients with hematological malignancies. However, some concerns remain that they may cause serious cardiovascular adverse events (AEs), for which data are scarce. In this study, gradient boosting machine algorithm-based model was fitted to identify safety signals of serious cardiovascular AEs reported for tisagenlecleucel in the World Health Organization Vigibase up until February 2024. Input dataset, comprised of positive and negative controls of tisagenlecleucel based on its labeling information and literature search, was used to train the model. Then, we implemented the model to calculate the predicted probability of serious cardiovascular AEs defined by preferred terms included in the important medical event list from European Medicine Agency. There were 467 distinct AEs from 3,280 safety cases reports for tisagenlecleucel, of which 363 (77.7%) were classified as positive controls, 66 (14.2%) as negative controls, and 37 (7.9%) as unknown AEs. The prediction model had area under the receiver operating characteristic curve of 0.76 in the test dataset application. Of the unknown AEs, six cardiovascular AEs were predicted as the safety signals: bradycardia (predicted probability 0.99), pleural effusion (0.98), pulseless electrical activity (0.89), cardiotoxicity (0.83), cardio-respiratory arrest (0.69), and acute myocardial infarction (0.58). Our findings underscore vigilant monitoring of acute cardiotoxicities with tisagenlecleucel therapy.

show abstract

Detecting early safety signals of infliximab using machine learning algorithms in the Korea adverse event reporting system

Cited by 4 publications

References 40 publications

A Framework for Evaluating Geomagnetic Indices Forecasting Models

A Framework for Evaluating Geomagnetic Indices Forecasting Models

An open-source platform integrating emerging data sources to support multi-modal active pharmacovigilance

A real-world pharmacovigilance study on cardiovascular adverse events of tisagenlecleucel using machine learning approach

Contact Info

Product

Resources

About