Distributed Edge Streaming

Images stream. Start reading in seconds.

Sirona streams diagnostic images instead of downloading the whole study first. Studies are pre-cached from your worklist and hanging protocols, so the first image is up while the rest is still arriving.

See Sirona in Action

How Sirona is Different

Radiologists see images in seconds, not minutes

Traditional PACS clients typically transfer a study before display. A chest CT with 500 slices can mean waiting for all 500 to arrive before you see the first one. On a slow link, that is minutes of dead time per study. Sirona streams instead: the image in front of you is fetched first, the rest of the stack is prefetched behind it, and studies are pre-cached from your worklist and hanging protocols before you open them. What you read is the diagnostic image, not a thumbnail. Radiologists start reading while the study is still arriving.

How the streaming architecture works

A global CDN, priority-scheduled delivery, worklist-driven prefetching, and GPU rendering in the browser: the infrastructure behind fast first-image delivery.

AWS CloudFront (700+ Edge Locations)

Images are delivered over AWS CloudFront's global edge network — 700+ edge locations worldwide. Studies route to the nearest edge node, minimizing latency regardless of where the radiologist sits.

Priority-Scheduled Delivery

The image in front of you is always fetched first; the rest of the stack is prefetched behind it.

Progressive Image Loading

The first image is up in seconds, and the rest of the study keeps arriving while you read. You never wait for a whole-study download.

Worklist-Driven Prefetching

Studies are pre-cached from your worklist, and the active hanging protocol decides which series and priors load next, so they are ready before you navigate to them.

WebGL GPU Rendering

Images render using your device's GPU via WebGL: hardware-accelerated windowing, zoom, pan, and MPR with no Java, ActiveX, or desktop client. Just Chrome and the Sirona RadOS Manager extension.

Relevant Priors, Pre-Loaded

Relevant priors are identified and pre-loaded before you open the exam, so the comparison is ready when you are.

Streaming at every layer of the stack

Edge Delivery

AWS's global edge network between the archive and the viewer

Studies live in Sirona's cloud archive on AWS and are delivered through AWS CloudFront's global edge network (700+ locations), so data reaches the reader from infrastructure close to them rather than from a server room hundreds of miles away. The CDN carries the read load, so adding radiologists who read at the same time does not mean adding servers at your sites. Netflix is the familiar analogy: playback starts from a nearby edge, not from the studio's data center.

AWS CloudFront: 700+ edge locations globally

Delivery from infrastructure close to the reader

No whole-study download before the first image

The CDN carries read load, so more radiologists don't require more servers

Delivery Order

Priority-scheduled delivery for a fast first image

Sirona delivers a study progressively rather than making the viewer wait for a complete download. That is what produces the 'start reading in seconds' experience: not a thumbnail, but the study arriving in a usable order.

The image in front of you is always fetched first

First image within seconds on a business-class connection

The rest of the stack is prefetched behind the active image

What you read is the diagnostic image, not a thumbnail

Intelligence

Prefetching driven by your worklist and hanging protocol

The viewer doesn't wait for you to ask. Prefetch is driven by your worklist and by the active hanging protocol: open a chest CT comparison and the prior's matching series is already loading. That is what makes streamed reading feel local.

Hanging protocol-aware prefetching — priors and comparison series preloaded

Worklist-driven pre-caching before the study is opened

Relevant priors identified and pre-loaded before you open the exam

Browser-side cache eliminates re-fetching of recently viewed images

Rendering

GPU-accelerated viewing without a desktop client

Once images arrive at the browser, WebGL renders them using the device's GPU: hardware-accelerated windowing, zoom, pan, scroll, MPR reconstruction, and MIP generation. No Java plugin. No ActiveX control. No desktop application. Sirona runs in Google Chrome with the Sirona RadOS Manager extension, and the same rendering pipeline runs on a reading-room workstation and a laptop at home. Window/level and zoom run on the GPU. MPR reconstruction happens in the browser. Legacy PACS needs a dedicated workstation and proprietary software to do the same.

WebGL GPU rendering, hardware-accelerated in Chrome

No Java, no ActiveX, no desktop client

Full diagnostic tools: window/level, zoom, MPR, MIP, 3D

Same rendering pipeline on every supported workstation

Streaming performance

Seconds

to a first image, while the rest of the study arrives

700+

CloudFront CDN edge locations worldwide

>50 Mbps

recommended business-class connection

Images that appear in seconds

Read the partnership announcement

“This partnership marks the beginning of the cloud-native era in radiology software globally. Sirona has delivered an architecture that is fundamentally different from anything else in the market today — Sirona is unified, and cloud-native from the ground up. For teleradiology, that architectural distinction is existential, not incremental.”

Andy Donaldson

CTO, Everlight Radiology

Built Different: the architecture behind RadOS

Sirona's technical advisor Jeff Queisser (Co-Founder of Box, NYSE: BOX) on how RadOS was architected for reliability and performance at scale.

Read the ERA case study

FAQs

What is progressive image streaming for radiology?

Instead of downloading an entire study before anything appears, Sirona delivers images progressively from cloud infrastructure close to the reader, and pre-caches studies based on your worklist and hanging protocols. The first image is up while the rest arrives.

How fast do images load?

The first image appears within seconds on a business-class connection, and the rest of the study keeps arriving while you read. Studies pre-cached from your worklist open faster still.

What bandwidth does Sirona require?

Sirona runs in Google Chrome over a business-class internet connection; we recommend 50 Mbps or better. More bandwidth means the rest of the study arrives sooner.

How does edge delivery improve performance?

AWS CloudFront's global edge network (700+ locations) delivers study data from infrastructure close to the reader, keeping delivery fast wherever your radiologists sit.

Is diagnostic quality maintained with streaming?

Yes. Streaming changes the order in which a study arrives, not what arrives. The image you read is rendered from the full DICOM pixel data, the same as with a whole-study download. It just gets in front of you sooner.

Does this work with all modalities?

Yes. The streaming pipeline handles DICOM modalities including CT, MR, CR, US, MG, PT, and NM. Multi-frame studies such as fluoroscopy and cine MR stream frame by frame.

How does prefetching work?

Prefetching is driven by your worklist and by the active hanging protocol, not by guesswork. Studies are pre-cached from the worklist; prior comparison series and protocol-defined series load before you navigate to them. That is what makes reading feel local even though images are streaming from the cloud.