Digital pathology storage: strategies and considerations

Executive summary

Technology innovations are keeping pace with and, in some cases, driving the digital transformation of pathology image analysis.

Healthcare professionals have noted the significant benefits of this new way of practicing pathology — including faster access to important information, accelerated collaboration, and greater opportunities for consultation.

Organizations that have already adopted digital pathology have reduced their handling of glass slides and enhanced their workflows with algorithm-based assistance. Those with advanced implementations have also been able to monetize data that was once trapped in large sets of stored glass slides. With each stage of digital pathology adoption, the potential to support research and education grows.

But there’s one major challenge. How and where will all the data be stored?

The answer isn’t simple, and there’s no one-size-fits-all approach. However, some facts do apply to every digital pathology practice:

• Healthcare organizations need storage models with enough scalable capacity to manage growing volumes of data.
• The data must be available as needed without too much complexity in the storage system’s pricing or access.
• The storage environment must be secure to control access and protect personal health information.

Unlock the value from your data

Once in digital form, digital pathology image data can be used to train artificial intelligence models, and identifying specific patterns for analysis, verification, and patient care collaboration.

Smaller sets of images, even individual images, can be made available anonymized with consent to other entities. In these instances of approved data sharing, there is a monetization opportunity.

Compensation can be monetary, or it can take other forms. For example, researchers may trade access to data sets to further discovery, and data can be used to enhance education.

Storage strategies emerging now range from on-premises to hybrid to the cloud. Here, we investigate these solution options and identify key considerations for digital pathology storage.

Sizing up the storage issue

Industry experts estimate a typical digital pathology practice will need to consume at least a petabyte of storage per year. To fathom a petabyte, imagine 20 million filing cabinets filled with 500 billion pages of information. It’s enough storage to hold 223,101 DVD quality movies.

Pathologists prefer to work with high-resolution whole slide images, which are between 2 and 4 gigabytes each, approximately the file size of one full-length feature film in high definition (HD). Many hospitals and labs will produce 700 to 1,000 slides per day, generating an average of 3,000 gigabytes, or 3 terabytes, of new data each day. This adds up as 1,000 terabytes amounts to 1 petabyte.

It’s clear why the storage of digital image files leads the list of concerns with the massive amount of data. The information also needs to be stored in a way that is secure and accessible. The burden is tremendous, especially for an organization whose primary focus is healthcare.