Digital pathology is the clinical system for creating, managing, viewing, and interpreting digital slide images. Whole slide imaging is the scanning step inside that system. Neither term means AI. They describe the infrastructure that has to exist before software can analyze a pathology slide.

That distinction explains much of the gap between pathology research and clinical deployment. A model cannot analyze a glass slide that was never digitized, and a scanner alone does not create a validated diagnostic workflow.

Whole Slide Imaging Is One Component

A whole slide imaging system converts a prepared glass slide into a high-resolution digital image. The FDA's product classification describes a system that includes an image management system, scanner and software, and display monitor. It also says the WSI system itself does not include automated image-analysis functions.

The first system authorized in the United States for primary diagnosis was the Philips IntelliSite Pathology Solution in 2017. The FDA reviewed it through the De Novo process after a study of roughly 2,000 surgical pathology cases found diagnoses made from the digital slides comparable to diagnoses made from glass slides.

Digital Pathology Is The Larger Workflow

The FDA defines digital pathology around acquisition, visualization, analysis, transfer, storage, and interpretation. In practice that means slide preparation and scanning, image quality control, storage, case assignment, a diagnostic viewer, display hardware, access controls, and a way to keep the digital case tied to the laboratory record.

Each part can affect the final image. Focus, tissue folds, staining, scanner settings, compression, display conditions, and network performance all become workflow variables. The relevant question is not only whether a scanner produces a beautiful image. It is whether the complete system preserves the information a pathologist needs for the intended specimen types and use.

Validation Comes Before Routine Diagnosis

The College of American Pathologists maintains guidance for validating whole slide imaging for diagnostic purposes. The principle is practical: a laboratory validates the system for its intended clinical use before it replaces or supplements the microscope in routine diagnosis.

Local validation matters because the deployed system includes local specimens, preparation methods, scanners, displays, users, and workflows. An FDA marketing authorization for the product does not perform that implementation work for the laboratory.

Where AI Enters The System

Once slides are digital, software can support quality control, detect or segment tissue, count cells, quantify biomarkers, flag suspicious regions, or generate a score for a narrow intended use. Those functions require their own evidence. A validated WSI workflow proves that digital slides can support the intended diagnostic work. It does not validate every model later applied to those images.

For the difference between digitization and an authorized algorithm, see why pathology has so few cleared AI devices. The concise definitions remain in digital pathology and whole slide imaging.

The Implementation Questions To Ask

  • Which specimen types and clinical uses were included in validation?
  • How are scan failures, focus problems, and missing tissue detected and resolved?
  • What display, storage, and network conditions are required for routine reading?
  • Does an AI function analyze the same image format and preparation workflow used locally?
  • Who reviews the output, and what happens when the algorithm is unavailable or wrong?

Primary Sources

See the FDA's first WSI authorization, Digital Pathology Program, and CAP validation guideline.