Webinar's content

Join us for the webinar titled "AI-Powered Tools for Advancing 3R Alternatives in Research" featuring insights from two distinguished speakers:

Mariana Neves (BfR) presenting SMAFIRA Title: SMAFIRA Web Tool for Searching the Scientific Literature for Alternatives

In many countries, a careful search of the scientific literature for alternative methods is necessary before requesting permission to perform an animal experiment. However, finding an alternative method for a particular research goal is a difficult and time-consuming task, for which few tools are currently available.

We present the SMAFIRA Web tool that includes methods to support researchers in two tasks: (i) an automatic classification of the abstracts' models in eight labels (e.g., in vivo, tissue/biopsy) using machine learning, and (ii) ranking the results according to the similarity of their research goal. The tool integrates with PubMed, and given input article (PubMed identifier), it performs the search on the corresponding list of similar articles. SMAFIRA provides many additional interesting features, e.g., customization of the results based on users’ feedback and the possibility to save the session URL for a later analysis of the results.

Currently, two improvements are being carried out for the tool: (i) an expansion of the labels and the training data for the classification of the abstracts' models, and (ii) a shared task (https://smafira-bf3r.github.io/smafira-st/) to collect annotations from the community for the re-ranking procedure, for which we invite the contribution of researchers worldwide.

Wynand Alkema (TenWise) presenting 3Ranker Title: 3Ranker – A Free Tool to Search for Non-Animal Alternatives

The search for existing non-animal alternative methods for use in experiments is currently challenging because of the lack of both comprehensive structured databases and balanced keyword-based search strategies to mine unstructured textual databases. We will present 3Ranker, which is a fast, keyword-independent algorithm for finding non-animal alternative methods for use in biomedical research. The 3Ranker algorithm was created by using a machine learning approach, consisting of a Random Forest model built on a dataset of 35 million abstracts, followed by iterative model improvement with expert curated data. Trials showed that the AI-based classifier was able to identify articles that describe potential alternatives to animal use, among the thousands of articles returned by generic PubMed queries on dermatitis and Parkinson's disease. The 3Ranker algorithm is freely available at www.open3r.org; the future goal is to expand this framework to cover multiple research domains and to enable its broad use by researchers, policymakers, funders and ethical review boards, in order to promote the replacement of animal use in research wherever possible.

Speakers

LARA-NEVES Mariana
Researcher at BfR

Mariana Neves is a researcher in the area of biomedical natural language processing at the German Federal Institute for Risk Assessment (BfR) in Berlin since 2017. Previously, she was a post-doc researcher at the Hasso-Plattner Institute in Potsdam and at the Humboldt-University zu Berlin. She obtained her doctoral studies in Computer Science at the Universidad Complutense de Madrid in Spain, and her Master in Computer Science as well as an Engineering degree at the Universidade Federal de Pernambuco in Brazil.

Linkedin profile

ALKEMA Wynand
Professor at Hanze University
CSO at TenWise

Wynand Alkema is a professor of data science at the Hanze University of Applied Sciences. He is also CSO of TenWise, a company specialized in analysing large scale data sets for pharmaceutical and food companies. Throughout his career he has worked at the Karolinska Institute, Merck and NIZO, always on projects that involved data science, text mining, artificial intelligence, biomarker and drug development and food science.

Linkedin profile

Back