• Project number: F2566
• Institution: Federal Institute for Occupational Safety and Health (BAuA)
• Status: Ongoing Project
• Planned end: 2024-12-31

Description:

Since biopersistence can currently only be determined by animal experiments and there is such a huge diversity of fibre materials, the detection of critical fibre morphologies in material and aerosol samples is the first important step in the identification of properties that are hazardous to health at the earliest possible stage. It is therefore necessary to collect relevant, verified material data in a database management system (DBMS) and make those data readily accessible to material developers, thus enabling them to design substances safely.

This project will involve the creation of an image database for fibre materials to support automatic image evaluation powered by neural networks. In addition, the database will be used to manage the research data produced by Unit 4.5 of the Federal Institute for Occupational Safety and Health (Bundesanstalt für Arbeitsschutz und Arbeitsmedizin, BAuA) and make this valuable information accessible to third parties via a web server. The content provided will include meta-information about the images generated by scanning electron microscopy, the relevant material and morphological properties of the materials analysed, and the details of the learned neural networks. In addition, software interfaces are to be developed to serve as links between the database and pre-existing software and to ensure the data feed can be managed effectively.