Evidence-based toxicology最新文献

{"title":"Proof-of-concept for using machine learning to facilitate data extraction for human health chemical assessments: a study protocol.","authors":"Michelle Angrish, Kristina A Thayer, Brittany Schulz, Artur Nowak, Amanda Persad, Allison L Phillips, Glenn Rice, Teresa Shannon, A Amina Wilkins, Krista Christensen, Elizabeth G Radke, Andrew Shapiro, Michele M Taylor, Vickie R Walker, Andrew A Rooney, Sean M Watford","doi":"10.1080/2833373x.2024.2421192","DOIUrl":"10.1080/2833373x.2024.2421192","url":null,"abstract":"<p><strong>Background: </strong>Systematic review (SR) methods are relied upon to develop transparent, unbiased, and standardized human health chemical assessments. The expectation is that these assessments will have discovered and evaluated all of the available information in a trackable, transparent, and reproducible manner inherent to SR principles. The challenge is that chemical assessment development relies on mostly literature-based data using manual approaches that are not scalable. Various SR tools have increased the efficiency of assessment development by implementing semi-automated approaches (human in the loop) for data discovery (literature search and screening) and enhanced data repositories with standardized data collection and curation frameworks. Yet filling these repositories with data extractions has remained a manual process and connecting the various tools together in one interoperable workflow remains challenging.</p><p><strong>Objectives: </strong>The objective of this protocol is to explore incorporation of a semi-automated data extraction tool (Dextr) into a chemical assessment workflow and understand if the new tool improves overall user experience.</p><p><strong>Methods: </strong>The workflow will use template systematic evidence map (SEM) methods developed by the Environmental Protection Agency for the identification of included studies. The methods described focus on the data extraction component of the workflow using a fully manual or a semi-automated (human in the loop) data extraction approach. Both the manual and semi-automated data extractions will occur in Dextr. The new data extraction tool will be evaluated for user experience and whether the data extracted using the automated approach meets or exceeds metrics (precision, recall, and F1 score) for a fully manual data extraction.</p><p><strong>Discussion: </strong>Artificial intelligence (AI) and machine learning (ML) methods have rapidly advanced and show promise in achieving operational efficiencies in chemical assessment workflows by supporting automated or semi-automated SR methods, possibly improving the user experience. Yet incorporating advances into sustainable workflows has remained a challenge. Whether using a tool like Dextr improves operational efficiencies and the user experience remains to be determined.</p>","PeriodicalId":520510,"journal":{"name":"Evidence-based toxicology","volume":"2 1","pages":"2421192"},"PeriodicalIF":0.0,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12004489/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144054740","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comprehensive item bank of internal validity issues of relevance to in vitro toxicology studies.","authors":"Gunn E Vist, Heather M R Ames, Gro H Mathisen, Trine Husøy, Camilla Svendsen, Anna Beronius, Emma Di Consiglio, Ingrid L Druwe, Thomas Hartung, Sebastian Hoffmann, Carlijn R Hooijmans, Kyriaki Machera, Pilar Prieto, Joshua F Robinson, Erwin Roggen, Andrew A Rooney, Nicolas Roth, Eliana Spilioti, Anastasia Spyropoulou, Olga Tcheremenskaia, Emanuela Testai, Mathieu Vinken, Paul Whaley","doi":"10.1080/2833373X.2024.2418045","DOIUrl":"10.1080/2833373X.2024.2418045","url":null,"abstract":"<p><strong>Context: </strong><i>In vitro</i> toxicology studies are increasingly being included as evidence in systematic reviews and chemical risk assessments. INVITES-IN, a tool for assessing the internal validity of <i>in vitro</i> studies, is currently under development. The first step in developing INVITES-IN involves the creation of an \"item bank,\" an overview of study assessment concepts that may be relevant to evaluating the internal validity of <i>in vitro</i> toxicology studies. The item bank and methodology for its creation presented in this manuscript are intended to be a general resource for supporting the development of appraisal tools for <i>in vitro</i> toxicology studies and potentially other study designs.</p><p><strong>Methods: </strong>We derived the item bank from seven literature sources (one existing item bank created from a systematic review of assessment criteria for <i>in vitro</i> studies, and six purposively sampled study appraisal tools) and the transcripts of three focus groups. Assessment criteria plausibly relating to internal validity were abstracted from the literature sources and focus group transcripts, disaggregated into individual criteria, then normalised to express in the simplest achievable language the core issue in each criterion - an \"item bank\" of assessment concepts. The items were then mapped onto a set of bias domains. We conducted simple descriptive statistical analyses and visualisations to describe patterns in the dataset and developed recommendations for the use and development of the item bank.</p><p><strong>Results: </strong>The item bank contains 405 items of potential relevance to evaluating the internal validity of <i>in vitro</i> toxicology studies.</p><p><strong>Discussion: </strong>To our knowledge, this is the second item bank of any kind to have been created for toxicology studies, and the first to use focus groups as a data source alongside literature analysis. The large number of items contributed by focus group discussions suggests this is an efficient method for capturing internal validity issues that are not easily identifiable in the literature. We believe our item bank and methodology for its creation will be a useful resource for supporting the development of appraisal tools. Due to the broad applicability of many items in the item bank, it may be informative for study designs beyond the <i>in vitro</i> domain.</p>","PeriodicalId":520510,"journal":{"name":"Evidence-based toxicology","volume":"2 1","pages":"2418045"},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12180937/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144478502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}