🇦🇺 Australia · Annalise.ai — AI Radiology Reporting

Status: 🟩 COMPLETE 🟦 LIVING Section: 10 — AI and LLMs


Vendor	Annalise.ai (a Harrison.ai company)
Country/origin	🇦🇺 Australia (Sydney; founded 2019)
Recommended for AUS?	✅ Strongly yes — Australian medical AI; TGA-approved; deployed across Australian radiology
Privacy summary	Australian data residency; TGA Class IIb medical device; HIPAA capable; clinical-grade data handling; integrates with hospital PACS systems
Free tier	❌ Enterprise medical only
Paid tiers	Hospital/radiology practice contracts; not publicly priced
First released	Founded 2019; CXR product launched 2020; CTB product launched 2022
Last reviewed	June 2026
Official site	https://annalise.ai

What it is

Annalise.ai is an Australian medical AI company that builds AI tools for radiologists — software that analyses medical imaging (chest X-rays, CT brain scans, mammography) and flags findings for radiologist review. It’s one of Australia’s most significant medical AI deployments and is used in radiology departments across Australia and internationally.

Founded by: Brothers Aengus Tran (radiologist) and Dimitry Tran (technologist), with Harrison.ai as the parent company. The naming reference is to “Annalise” — meaning “to analyse.” Harrison.ai is named after John Harrison, the 18th-century clockmaker who solved the longitude problem (a metaphor for solving difficult precision problems with technology).

What the AI does:

Annalise CXR (Chest X-Ray): Detects 124+ findings on chest X-rays — pneumothorax (collapsed lung), pneumonia, masses, fractures, foreign bodies, and many more
Annalise CTB (CT Brain): Detects 130+ findings on CT brain scans — bleeds, strokes, tumours, fractures, and more
Annalise Triage: Prioritises urgent cases for faster reporting

The AI is a “decision support” tool — it flags possible findings; the radiologist makes the diagnosis and final clinical decision.

Why this matters for Australian healthcare

This is one of the genuinely high-impact AI deployments in Australia:

Radiologist shortage: Australia faces a chronic shortage of radiologists. Demand for imaging keeps rising; trained radiologists don’t scale at the same pace.
Speed: AI analysis takes seconds; flagging urgent cases (intracranial haemorrhage on CT brain, pneumothorax on chest X-ray) for priority reading saves lives.
Comprehensive review: AI checks for 124 things on a chest X-ray simultaneously; a tired radiologist at the end of a shift might miss subtle findings the AI catches.
Quality and safety: Used as a “second pair of eyes” to reduce missed findings.
Coverage: Annalise’s products are deployed across major Australian hospitals, private radiology providers (I-MED, Lumus Imaging, Sonic Imaging), and internationally.

This isn’t future technology — it’s actively in use across Australian radiology right now.

The TGA (Therapeutic Goods Administration) angle

Critically important context: Annalise’s products are TGA-registered as Class IIb medical devices in Australia. This means:

They’ve been formally assessed for safety and efficacy
They have specific approved use cases (intended use statements)
Their performance has been clinically validated
They’re subject to ongoing TGA oversight

This is a higher bar than consumer AI tools or non-clinical AI. It’s appropriate for clinical AI affecting patient outcomes.

How it integrates into radiology workflow

Annalise integrates with the PACS (Picture Archiving and Communication System) that radiologists use:

Image acquired at the imaging machine (X-ray, CT scanner)
PACS receives the image
Annalise AI analyses the image automatically — takes seconds
Findings flagged as overlays or notifications in the radiologist’s workflow
Radiologist reviews the image and AI findings
Radiologist makes the diagnosis and creates the report

The radiologist remains the decision-maker. The AI is a tool that supports their work.

How it compares to international alternatives

Tool	Country	Focus	TGA-approved?
Annalise CXR/CTB	🇦🇺	Multi-finding radiology	✅
Aidoc	🇮🇱	Multi-modality emergency findings	✅
Lunit	🇰🇷	Chest X-ray, mammography	✅
Qure.ai	🇮🇳	Chest X-ray, head CT	✅
Viz.ai	🇺🇸	Stroke detection workflow	✅
HeartFlow	🇺🇸	Cardiac CT specifically	✅

Annalise’s competitive advantage: the breadth of findings detected per image (124+ on chest X-ray vs many competitors with narrower scope). For Australian healthcare specifically: Australian product with Australian regulatory engagement and Australian clinical research partners.

Annalise’s research approach

Distinctive feature: Annalise’s clinical validation involves Australian academic radiologists and clinical trials. Published research includes:

Multi-finding detection performance studies
Comparison studies against human radiologist performance
Workflow integration impact studies
International multi-site validation

This evidence base supports clinical adoption and regulatory approval.

Privacy and data sovereignty

Australian data residency for Australian deployments
TGA Class IIb medical device regulatory status
HIPAA capable for international deployments
Hospital-grade integration — works within existing data security frameworks
Patient information handled per hospital policies — Annalise integrates into clinical systems rather than replacing them

For Australian Privacy Act compliance:

Health information is “sensitive information” with heightened protections
Annalise’s clinical integration generally satisfies APP requirements
Hospitals retain responsibility for patient information governance

Who can use Annalise

Annalise is enterprise medical software — not accessible to individual patients or non-medical users:

Hospital radiology departments
Private radiology practices
Tele-radiology providers
Research institutions

Access is through hospital/practice procurement processes, not individual signup.

For Australian patients: if your imaging is done at a hospital or radiology practice that uses Annalise, the AI may have analysed your scan as part of the radiologist’s workflow — generally invisible to you as the patient.

Annalise’s parent: Harrison.ai

Annalise.ai is part of Harrison.ai, a broader medical AI company founded by the Tran brothers in 2018. Harrison.ai’s other ventures include:

Franklin.ai — pathology AI (deep learning for diagnostic pathology)
Edison.ai — additional medical AI applications

Harrison.ai has raised substantial international investment and represents one of Australia’s most prominent medical AI exports.

See harrison-ai-company for the parent company entry.

Australian medical AI context

Annalise is part of a growing Australian medical AI ecosystem including:

Annalise.ai (radiology)
Harrison.ai (diagnostic AI)
Franklin.ai (pathology)
Heidi Health (clinical documentation)
Lyrebird Health (clinical documentation alternative)
CSIRO Data61 (medical AI research)
Pulse (newer Australian clinical AI)

For Australian healthcare, choosing Australian medical AI provides:

TGA regulatory familiarity
Data sovereignty advantages
Australian clinical context understanding
Local support and ongoing development

Gotchas

AI is decision support, not diagnosis. The radiologist makes the diagnosis. AI flags possible findings.
Not consumer-accessible. Patients don’t interact with Annalise directly.
False positives and false negatives both happen. AI is imperfect. Radiologist judgment is essential.
AI assistance ≠ AI replacement. Annalise augments radiologist workflow; it doesn’t eliminate the need for trained radiologists.
Regulatory approval is specific to intended use. AI approved for chest X-rays isn’t approved for other imaging types without separate approval.
Integration matters. Hospital procurement of clinical AI involves IT, clinical, governance, and regulatory considerations beyond software functionality.

Sources

Annalise.ai official: annalise.ai
Harrison.ai company information
TGA Class IIb medical device register
Published clinical research from Annalise team
Australian Radiology AI adoption coverage (RANZCR, Inside Imaging)
Healthcare IT News and STAT News on Annalise (2021-2026)
I-MED Radiology, Lumus Imaging, Sonic Imaging adoption announcements

Tech & AI, Explained

Explorer

annalise-ai