Experience, on every shift.

AI that helps every doctor read emergency head CT with the confidence of a senior radiologist.

Emergency head CT. A project in development, not in clinical use.

Become a partner hospital See how it works

Axial non-contrast head CT with a hemorrhage in the left basal ganglia outlined — Interface illustration. CT image: Dahlqvist et al., CC BY 2.0, via Wikimedia Commons. Not ApiVision data.

At 3 a.m., one doctor faces a queue of head CTs.

Night shifts often fall to junior doctors.
Reading experience takes years to build, and the emergency cases arrive anyway.
A critical case can sit in the queue.
A bleed may wait behind several routine scans.
Reports take time.
Under pressure, a complete and structured report is hard to write.

Two tools for the emergency head CT shift

In development

ApiVision Head CT

Flags suspected intracranial hemorrhage on non-contrast head CT and moves those cases to the top of the reading list.

For: Emergency doctors, on-call radiologists.

In development

ApiVision Report Assistant

Drafts a structured head CT report from a standard template and reminds you of the items that are easy to miss.

For: Radiologists, residents.

Roadmap

Chest X-ray AI

Planned after the two tools above.

Findings we are building for

Target findings in development. Not validated for clinical use.

Intraparenchymal hemorrhage
Subarachnoid hemorrhage
Subdural hematoma
Epidural hematoma
Intraventricular hemorrhage
Midline shift
Mass effect

Early ischemic signs
Skull fracture

Fits into the workflow you already have

1. ScanCT is sent to PACS as usual.
2. AnalyzeApiVision reads the scan in the background.
3. PrioritizeSuspected bleeds move to the top of the list and are highlighted.
4. ReportThe doctor reviews, decides and finalizes a drafted report.

The doctor always makes the final call.

Reading list

01:52CT HeadED
02:07CT HeadNeurology
02:15CT HeadED
02:26CT HeadICU
02:33CT HeadED
02:41CT HeadEDNewPriority

Axial 16/34

Coronal

Priority RoutineReview soonUrgent

Draft report

Findings

Hyperdense focus in the left basal ganglia. Ventricles and midline: for the doctor to confirm.

Impression

Suspected intraparenchymal hemorrhage.

Check: midline shift

Draft. Not yet reviewed by a doctor.

Interface illustration. The reading list and the draft report are mock data. The CT is a 34-slice scan of a normal brain, CC0: Uppsala University Hospital, via Wikimedia Commons. The hyperdense focus is simulated and was drawn in for illustration.

Built for hospitals in Vietnam

These are design goals for tools that are still in development.

Data stays in your hospital
Designed for an on-premise option inside the hospital network, in line with Vietnam's Law on Personal Data Protection.
Works with your PACS
Designed to work alongside the PACS your hospital already uses.
Vietnamese and English reports
Drafts follow the report templates your department knows.
Local support
Talk to us by Zalo, phone or in person.

We're looking for our first partner hospitals.

What you get

Early access to the tools, free during the pilot
Co-authorship on the research and papers we do together
A direct say in how the product is built

What we ask

A radiology or emergency team interested in research
Feedback on real workflows
Ethics approval and anonymized data, following hospital rules

Become a partner hospital

Built by a radiologist in training

Albert Pham

Founder. Radiology Resident, University of Medicine and Pharmacy at Ho Chi Minh City

I've seen how much a few years of experience change a night shift. ApiVision is my way of giving that to every young doctor.

Interested in joining? We're looking for AI engineers and clinical collaborators.

Let's build it together.

Leave your details and we will get back to you. You can also write directly to [email protected].

Experience, on every shift.

At 3 a.m., one doctor faces a queue of head CTs.

Night shifts often fall to junior doctors.

A critical case can sit in the queue.

Reports take time.