Valuable medical and biological images languish for want of support

Experts are working globally on standards for sharing medical and biological image data, much of which cannot be reused because they are not publicly available or well organized. To tap this resource for discovery, a group of scientists is calling for long-term funding commitments for adequate data storage at openly accessible sites. 

Australia, Japan, and Europe are already collaborating to enable global image data sharing, but more countries should invest in open data resources, or an immense amount of valuable data will languish, a team of concerned scientists recommended in a Nature Methods opinion column. 

"This will allow us to harvest the enormous potential of existing image data, preventing substantial loss of unrealized value from past investments in imaging acquisition infrastructure," wrote the group, which included Kedar Narayan of the Frederick National Laboratory for Cancer Research and 23 other scientists from government, industry, academic, and nonprofit institutions around the world. 

"What researchers most urgently need are openly accessible resources for image data storage that are operated under long-term commitments by their funders," the scientists wrote. Biological and medical image data are among the fastest growing data resource in the life sciences, they said. 

These images are complex and rich in biological information that can be reused for similar or unrelated projects. And as computer technology advances, these unique stores of imaging data become even more valuable for types of analyses that are only now becoming possible, the scientists said. But many research institutions do not have the technical resources to support this kind of open data sharing. Biological imaging is stored in large files, often reaching 10 terabytes or more in size. And to share them requires standardized labeling, annotation, and storage over the long term. 

To reap the full benefit from present and future image stores, the concerned scientists encouraged funders to develop open-access image data repositories and imaging infrastructures along with computing resources for interacting with the data, specialized staff to manage open-data sharing, and relevant training and education. 

Narayan is one of 46 scientists who collaborated in proposing Recommended Metadata for Biological Images (REMBI), a standard for light and electron microscopy and volume electron microscopy. Their recommendation was published in the May 21, 2021, issue of Nature Methods.  

These authors urge institutions, funders, and the scientific community to focus on implementing these standards worldwide, while ensuring that the image data is findable, accessible, interoperable, and reusable (FAIR).   

In this editorial the scientists argued, “Funders need to recognize that they currently lose substantial value from their previous investments into imaging acquisition infrastructure unless they provide the essential resources for storing and making the resulting image data accessible.” 

“It is essential that the imaging community and funders join forces to establish a sustainable and profitable image data ecosystem ensuring optimal output for all stakeholders,” they said. “We are looking forward to an exciting and fruitful dialog with national funders on the suggested action items presented here.”