Storage: The 800-Pound Gorilla in Digital Pathology

Dr. Rajendra Singh

by Rajendra Singh MD, Co-Founder, PathPresenter
Director of Dermatopathology and Digital Pathology,
Summit Health

The 800 pound gorilla in the room

Digital pathology is transforming how pathology departments operate, offering unprecedented opportunities for research, diagnosis, and education. However, the field faces a formidable challenge: data storage. The scale of the problem is vast and growing, posing one of the most significant barriers to widespread adoption of digital pathology solutions.

The Scale of the Storage Challenge

Pathology departments generate approximately one billion glass slides annually. Each digitized whole-slide image (WSI) ranges from 1 to 10 GB in size, resulting in a staggering accumulation of thousands of petabytes and even zetabytes of data every year. Moreover, the College of American Pathologists (CAP) guidelines require glass slides to be stored for at least ten years. When combined with decades of retrospective data, the sheer volume of storage needed becomes daunting.

Adding to the complexity is the requirement for redundancy in cloud storage. These measures, while critical for data integrity and disaster recovery, exponentially increase storage demands. Moreover, viewing and analyzing these massive files requires high bandwidth, significantly driving up costs. These escalating expenses often make digital pathology appear non-viable for many institutions.

Lessons Learned in Storage Solutions for Digital Pathology

Implementing digital pathology solutions, whether on-premises, cloud-based, or hybrid, has provided valuable insights. Below are some key considerations:

  1. Understand the Institution’s Risk Profile

Cloud Aversion vs. Cloud Adoption: Some institutions are highly risk-averse, preferring on-premises storage solutions to avoid cloud-related risks. Others embrace cloud storage for its scalability and adaptability to rapid technological advances. Increasingly, institutions are opting for a hybrid approach, leveraging the benefits of both models.

Critical Insight: Aligning storage strategies with the institution’s risk tolerance is essential for a successful implementation.

  1. Define the Use Case

Digital pathology use cases vary significantly. Some institutions pursue comprehensive implementation, while others take a phased approach—starting with specific areas like consultations, education, research, or primary diagnosis.

Critical Insight: Tailoring the storage solution to the institution’s immediate and long-term goals ensures resource optimization.

  1. Understand the True Costs of Storage

The cost of storing pathology slides extends beyond the per-terabyte price. Each slide involves millions of write operations for storage and read operations for retrieval, both of which incur separate charges.

Hot vs. Cold Storage: Retrieval speed plays a critical role in cost. Hot storage, which allows instant access, is significantly more expensive than cold or archival (glacial) storage. Developing a strategy to determine what data resides in which tier is essential.

  1. Implement Data Tiering and Flagging Mechanisms

Data tiering ensures efficient and cost-effective storage. For instance:

Hot Storage: Critical cancer cases can remain in hot storage for immediate access.

Cold or Archival Storage: Benign diagnoses and re-excision cases can be moved to lower-cost options or on-premises storage.

Deletion: Data with no foreseeable use can be safely deleted according to retention policies.

Critical Insight: Automated flagging mechanisms streamline this process, ensuring data is appropriately tiered based on clinical and institutional needs.

  1. Establish Comprehensive Policies

Storage strategies must include clear policies on data retention, transfer, disaster recovery, compliance audits, and adherence to regulatory frameworks. Regular reviews ensure policies remain aligned with evolving standards and technologies.

The Path Forward

While storage remains the proverbial 800-pound gorilla in the room, addressing the challenge head-on with thoughtful planning and strategic implementation can make digital pathology a viable and transformative solution. Institutions that successfully navigate the storage conundrum will be well-positioned to leverage the full potential of digital pathology, ultimately improving patient outcomes and advancing medical science.

About the Author

Dr. Rajendra Singh is Director of Dermatopathology at Summit Health and co-founder of PathPresenter. He serves as a member of the Digital and Computational Pathology Committee of the CAP, Editorial Board of the WHO for Classification of tumors, 5th Edition and the Board of Digital Pathology Association.