BlogPhoto

SaferWorldbyDesign: Trusted reproducible evidence supporting data-driven decisions

This blog post presents a collection of perspectives that elaborate our vision of reliable and reproducible science supporting product and risk assessment as presented by us this year at the Precision Health conference co-organized by MCBIOS and MAQC.

This blog post presents a collection of perspectives that elaborate our vision of reliable and reproducible science supporting product and risk assessment as presented by us this year at the Precision Health conference co-organized by MCBIOS and MAQC.


The described best practices of data science combined with modelling, machine learning, and AI guides our development of trusted SaferWorldbyDesign knowledge infrastructure and solutions development providing an ecosystem and toolset oriented towards enabling our customers to make the best decisions they can with all available evidence. Scientific reproducibility requires that data analyses and, more generally, scientific claims and regulatory evidence be published with data and software code so that others may verify the findings and build upon them. We are pursuing approaches to implement trusted reproducible in silico workflows and enhancing their acceptance, including community and collaboration approaches supported by SaferWorldbyDesign. Our goal is to establish best practices for tracking and reporting modern in silico data analyses in a trusted reproducible manner and obtaining community convergence on emerging best practices for the generation, evaluation, and acceptance of evidence. We hope these developments will contribute to solutions to resolving the current challenges in reproducibility in science broadly but also for the practical preparation and submission of evidence by industry practitioners for a regulatory science evaluation purpose. A trusted tamper-free framework for evidence generation and communication should help industries follow the guided trusted paths they wish to stay on with regulators and society over the longer term.

SaferSkin: principles we are following in building a trusted reproducible app for skin safety

Presenter: Barry Hardy (Edelweiss Connect)

https://youtu.be/2wACjemsu0E

Trusted Interpretation of Machine Learning Methods 

Presenter: Dawer Jamshead (Vertex Labs) 

https://youtu.be/dx1tXXKxmUU

Blockchain in the evaluation of new products including food, drugs, medical devices, and consumer goods

Presenter: Daniel Burgwinkel (KRM Competence Center, Switzerland)

https://youtu.be/M5pn1iWN52E

Data reproducibility and in silico modeling: the impact of data curation on machine learning models to reduce animal testing

Presenter: Vinicius M. Alves, University of North Carolina

https://whova.com/portal/mcbab_202104/videos/1QTMzATMxgDM/