What Is A Whole Slide Image? Lets Peel Through The “Layers” of a Pyramid Image

In the era of digital pathology, Whole Slide Images (WSIs) have revolutionized how we analyse tissue and cytology samples. But what exactly is a WSI, and how does a glass slide transform into a dynamic, zoomable digital universe? Let’s peel through the layers—literally—of this transformative technology.

From Pixels to Hex: How Whole Slide Images Are Digitally Built

At its core, a WSI is a massive mosaic of 1s and 0s. Here’s the journey:

1. Scanning the Slide: A high-resolution scanner captures thousands of tiny image “tiles” from the glass slide. Each tile is a grid of pixels, where every pixel’s colour (RGB or grayscale) is represented by binary code.

   
   

2. Binary Code: An example of the first 16 values of a binary tile can be represented as in the image below. This is the machine-readable format that is interpretable by computers. Both the tile data and metadata are stored in this format.

   
   

3. Binary to Hexadecimal: Binary data is cumbersome to process, so in some use cases, these values are converted into hexadecimal values (base-16). For example, the binary sequence 10101111 becomes the hex value AF. This compact format streamlines storage and processing. The above binary values can be represented in Hexadecimal values as shown below in Table A

   
   

4. Building the Image: These hex values define everything—pixel colours, tile coordinates, and even compression algorithms (like JPEG or PNG). Each tile is stitched together like a puzzle to recreate the full slide that is shown to the user. (See Table A)

   
   

   

   Table A - From Pixels to Hex: How Whole Slide Images Are Digitally Built

   
   

The Hidden Architect: Metadata

A WSI isn’t just pixels—it’s a structured package. Alongside the image tiles, a metadata file (often XML or JSON) acts as the blueprint. It includes:

• Technical specs: Tile size, resolution levels, scanner settings.

• Layer information: How zoom levels (e.g., 1x, 10x, 40x) are organized.

• Patient data: De-identified IDs, staining protocols, or clinical notes.

• Z-stack details: Multiple focal planes for cytology (more on this later).

  
  

Without metadata, your WSI viewer wouldn’t know how to assemble the tiles or display the right layer at the right zoom.

Pyramid Layers: Why Your WSI Is Like Google Maps

WSIs are multi-resolution images, which means they contain several layers, each representing the image at a different magnification. This design allows for smooth zooming and panning, without the need to load the full-resolution image every time. Think of it like Google Maps: when you first open the app, you see the whole country. As you zoom in, you see states, then cities, then streets, and finally individual buildings. Each zoom level loads only the relevant layer to save bandwidth and improve performance. WSIs work the same way.

This layered approach is not just about zoom; it enables efficient analysis, storage, and transmission, which are all vital in medical and research settings.

Imagine if Google Maps downloaded the street level view of the entire world onto your phone when you downloaded the app. It would take up immensely more storage than any one phone can handle.

Similarly, if a cloud-based slide viewer required one to download the entire WSI before opening, it would take hours of additional download time and storage issues on any one computer.                

Ultimately, these apps load that portion of the slide / map which the user is viewing on their screen at that moment in time. Additionally, these images are stored in layers to download the specific resolution required.

This pyramid-like structure ensures smooth navigation through gigapixel-sized images. (See Table B)

Table B: A Rendering of the Structure of Two Different Pyramid Images

Z-Stacking: Adding Depth to the Flat World

In cytology, cells often cluster in 3D, making single-focus imaging insufficient. Z-stacking solves this by capturing multiple focal planes. Each “Z-layer” represents a different depth, allowing pathologists to:

• Scroll through layers to find optimal focus.

• Reconstruct 3D cell clusters for accurate diagnosis.

Without Z-stacks, critical details in Pap smears or fine-needle aspirates could be missed.

From Data to Diagnosis: The Power of a Unified Viewer

WSIs come in many formats—SVS, TIFF, NDPI, MRXS—each with unique metadata structures. This is where compatibility matters. At identify.bio, our viewer is engineered to:

• Decode any WSI format, no matter how complex the metadata.

• Render Z-stacks and multi-resolution layers seamlessly.

• Integrate with your workflow, whether for research, education, or clinical diagnostics.

  
  

Conclusion: The Future Is Layered

Whole Slide Imaging is more than a digital replica of a glass slide—it’s a meticulously engineered universe of data, layers, and metadata. By understanding how binary code becomes hex, how tiles form images, and why layers and Z-stacks matter, we unlock the true potential of computational pathology.

Curious how identify.bio can simplify your WSI workflow? Let’s talk! Whether you’re battling compatibility issues or optimizing Z-stack analysis, we’re here to help you navigate the digital frontier.

Happy to get on a call to discuss how identify.bio’s WSI Viewer is the most optimal, secure and cost effective way of going digital.