|Engineering Sector:||Industrial Design, Applied Ergonomics and Design|
|Community preferences:||3D scanning
Overall FAIRness visualisation:
In this case study, the research focus is on ergonomics, helping ensure that design processes are optimised for use by the human body. This specific group of researchers work with 3D scanning of the human body so to facilitate the design process. The resulting data is both 3d-scans of individuals and anonymised, statistical measurements derived and collated from the scans.
The researchers highlight the contrast between the 3D scans and the statistical data. The former is much more complex to manage. The scanning process creates raw data (typically 45GB per scan). This is then converted to the PLY format, commonly used for processing 3D scans. This can be visualised via CAD software; the research group uses specific code in Matlab to analyse it in detail.
There are data protection issues with the scans, as they could be used to identify specific individuals. Sometimes, children are scanned too. The raw data, therefore, is highly sensitive. However, it is still important to document this data so it can be managed and resued within the group. It is a lot of effort to do this (there is currently no best practice guidelines), and the group noted it would be useful to have automated tools to help with metadata creation, indexing and retrieval. In any case, sharing the data is tricky.
The group will sometimes make use of third party data, licencing others’ scans of the human body
Dealing with the anonymised statistical data is much more straightforward. It is common practice to publish and a large collection of anthropometric data has already been published under the name of the DINED dataset collection. As this data has been processed, refined and anonymised its single components are of less use to researchers (the original 3d scans would be required for this). However, the data can be used to provide benchmark data and also to verify the group’s published results.
It’s loads of effort to index and document data being used during the research process, particularly raw and processed data
It is common practice, and makes good sense, to make the statistical data available for reuse.