Use Case Intelligence RSS Search use cases Search
NVIDIA Healthcare Australia

UNSW Sydney

Real-time Analytics

UNSW Sydney and the NSW Telestroke Service needed fast, quantitative stroke imaging for regional hospitals, but manual hemorrhage and infarct segmentation took too long and made it hard to guide treatment decisions in real time.

Read the original story Explore all use cases
20 minutes to seconds
23 regional hospitals statewide

Reported impact

Business outcome

The platform reduced hemorrhage-volume analysis from 20 minutes to seconds, automated lesion segmentation across 23 regional hospitals, and gave clinicians shared quantitative imaging during live stroke consultations instead of after-the-fact manual review.

  • Productivity Improvement
  • Risk Reduction
  • Scalability

Solution

UNSW Sydney and the NSW Telestroke Service moved stroke quantification from a slow research workflow into routine care by wiring MONAI Label and Auto3DSeg into a secure AWS eHealth cloud pipeline. CT and MRI studies from the statewide image repository and Telestroke PACS are pushed into XNAT, which triggers GPU-accelerated inference to segment hemorrhage and ischemic lesions, calculate volumes, and generate DICOM segmentations, overlays, and PDF reports. Those outputs flow back to clinicians for immediate review through the statewide repository, while tools such as OHIF and 3D Slicer support visualization and correction. Corrected segmentations are fed back for active learning and retraining, so the same platform can continuously improve while serving 23 regional hospitals with real-time imaging support during live consultations.

Data flow

CT and MRI scans move from the statewide repository and Telestroke PACS into XNAT, which triggers MONAI Label and Auto3DSeg on NVIDIA GPUs in AWS. The system returns segmentations, lesion volumes, overlays, and PDF reports for consultant review, and corrected annotations are fed back into the active-learning loop for retraining.

Solution architecture

4 components · 3 layers
  1. Compute
    • Auto3DSeg Automates the lesion-segmentation workflow that the Telestroke team uses for hemorrhage and infarct analysis.
    • NVIDIA GPUs on AWS Provide the accelerated cloud compute that makes near-real-time medical-image inference practical across the statewide network.
  2. Storage
    • XNAT Stores imaging studies and orchestrates the handoff that triggers MONAI inference and returns structured outputs to care teams.
  3. Serving
    • MONAI Label Runs the interactive inference service that segments stroke lesions and returns quantitative results during consultations.

Architecture clues

  • Approved models are published through the MONAI Model Zoo
  • Generates DICOM segmentation objects, bounding boxes, PNG overlays, and PDF reports
  • Runs in a secure AWS enterprise eHealth cloud environment
  • Uses active learning to correct, retrain, and redeploy models inside the clinical workflow

Evidence from the source

Images from the statewide repository, including the Telestroke PACS, are pushed to XNAT, automatically triggering MONAI Label for inference.
Powered by NVIDIA GPU-accelerated MONAI Label with Auto3DSeg, NSW Telestroke Service automated lesion segmentation across 23 regional hospitals statewide.
Thanks to MONAI and NVIDIA, we can now generate accurate hemorrhage volumes in seconds-something that used to take up to 20 minutes.
Using MONAI Label and NVIDIA GPUs, NSW Telestroke Service reduced hemorrhage volume analysis time from 20 minutes to seconds.
Transparency Evidence & methodology 90/100 reference value · 89% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value90/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact15/25
    1 metric describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence89%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    4 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. NVIDIA

    UNSW Sydney customer story

    First-party
    https://www.nvidia.com/en-us/case-studies/unsw-sydney/
    Source type
    Vendor-published customer story
    Published
    Feb 2, 2026
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page Feb 2, 2026 · NVIDIA · First-party
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Related

Similar use cases

See all like this
NVIDIA Apr 23, 2026 Biofy Healthcare Global Hospitals needed pathogen identification and antibiotic-resistance analysis much faster than five-day culture and antibiogram workflows, but CPU-bound genomic processing and limited AI expertise made it hard to turn… 100,000+ synthetic DNA sequences generated by the biology model Real-time Analytics Abby · H100 Tensor Core GPUs · NVIDIA CUDA-accelerated libraries NVIDIA Archived · Feb 9, 2026 Clinomic Healthcare Germany Clinomic needed to consolidate ICU patient data, reduce documentation burden, and run bedside analytics and telemedicine functions without relying on cloud connectivity that could introduce latency or operational risk… 143% increase in patient-care capacity Document ProcessingReal-time Analytics NVIDIA Jetson AGX Xavier · NVIDIA TensorRT NVIDIA Jan 17, 2025 The Netherlands Cancer Institute Healthcare Netherlands The Netherlands Cancer Institute needed faster, higher-resolution cone-beam CT reconstruction so clinicians could adapt radiotherapy to a patient's current anatomy, but memory-heavy 3D reconstruction models and CPU-era… 14 minutes on the bare-metal RTX 8000 machine Real-time Analytics NVIDIA A100 80GB GPUs · NVIDIA AI Enterprise · VMware vSphere