Healthcare / NeuroVerify
For hospitals, telehealth platforms and public health programmes that register patients remotely. NeuroVerify checks the selfie or video you already capture for deepfakes and manipulation, so a synthetic face does not open a patient record or unlock a benefit.
It fits the selfie or video step your registration already runs: one API call, no new step for the patient, one unified risk score back.
NeuroVerify does not verify identity documents, link records or approve enrolment. It adds authenticity evidence; your identity proofing and clinical or benefit policy keep the final decision.
The gap
Healthcare is moving registration online. Patients create a health ID, sign up for a telehealth service, enrol in an insurance or public benefit scheme, or check in for a remote consultation. Most of these flows end in a selfie or a short video to show that a real patient is present.
The records at stake are valuable. A patient record holds clinical history, prescriptions and insurance links. A benefit entitlement is worth real money. Both are reached through a face check that was designed for physical spoofs such as a printed photo or a screen replay.
Document and registry checks confirm that a record exists. Face match confirms that two images look alike. Basic liveness confirms that a capture followed a prompt such as blink or turn. None of them asks whether the media was generated, swapped or injected into the camera feed. A face swap or a virtual-camera stream can pass all three.
Patient enters details → Identity data is checked against a record → Selfie or video is captured → Liveness prompt passes → Face match passes → Record or entitlement is created
The authenticity of the selfie itself is never examined. This is the step NeuroVerify adds evidence to.
Related reading: synthetic media controls for clinical authorization and identity and media checks for telehealth visits.
Threats
These are attack techniques, not claims about any specific provider or programme.
An ineligible person registers with another person's details and presents a swapped or generated face at the selfie step. The underlying record is real; the person is not.
An attacker completes self-service registration for someone who is absent, then uses the new identity across providers, labs and pharmacies.
A registrant uploads an AI-generated portrait that matches a stolen name, then books consultations and collects prescriptions under it.
A real-time swap runs over the camera feed during onboarding or re-verification, so the face on screen is not the patient whose record is open.
At camps, kiosks and registration desks, a helper submits many registrations and substitutes faces. Volume hides the pattern from manual review.
A previously recorded clip is injected into the capture step to defeat liveness prompts on a returning patient's account.
Healthcare identity programmes differ by country, from national health IDs to insurer portals. The pattern is the same everywhere: the check confirms a record, and the media is trusted by default.
Privacy matters here. Patient data is sensitive, so see GDPR, biometrics and authenticity analysis and NeuroVerify data handling before you design the flow.
Where NeuroVerify fits
NeuroVerify sits between media capture and your registration policy. Your app sends the selfie, video or audio it already collects. NeuroVerify runs six signals in one request and returns a unified risk score with signal-level evidence.
Your policy then decides: register the patient, ask for a second factor, send to staff review, or decline. Every existing identity and consent control stays in place.
Pass the face image or video from your patient app, kiosk or telehealth sign-up to NeuroVerify with one API call.
Get one score and the evidence from each of the six signals.
Approve, step up with another factor, route to staff review or decline using thresholds you set.
Store the score and signal evidence with the registration event so a later review can see why it was approved.
What you get
Catch synthetic or swapped faces before a patient record, health ID or entitlement exists.
Patients take the same selfie or video. The check runs behind it.
Signal-level detail tells a reviewer why a case was flagged instead of showing a bare pass or fail.
Choose cloud, private cloud, on-premises or air-gapped deployment to match patient-data requirements.
Mobile and backend SDKs and REST mean a pilot can be live in under 24 hours.
Questions
No. NeuroVerify fits the selfie or video step your journey already runs. You send the media you already capture in one API call. There is no new step for the end user.
Integration uses native mobile SDKs (Android/Kotlin, iOS/Swift, Flutter), backend SDKs (Python, TypeScript/Node.js, Go, Java) or REST. Two lines of code is the product claim; live in under 24 hours is the target.
Not as a replacement. Liveness asks whether a live person was in front of the camera. NeuroVerify asks whether the media itself is authentic, and it includes a Live Presence Verification signal alongside five others, so it checks beyond liveness. See deepfake detection vs liveness detection.
You can keep your current liveness vendor and send the same selfie or video to NeuroVerify as well.
No. NeuroVerify checks face images, video and audio. It does not analyse identity documents or clinical records, and it does not link or create patient records.
It protects the selfie or video step of registration. Your existing document and registry checks continue unchanged.
Patient media is sensitive, so deployment choice matters. NeuroVerify can run in your own VPC, in a secure container, on-premises or air-gapped, so media can stay inside your boundary. See media data handling and our data handling page.
Neural Defend holds ISO/IEC 27001, ISO/IEC 27701, ISO/IEC 42001 and SOC 2 Type II.
NeuroVerify returns in under 200 ms p95 on eligible image requests under defined test conditions. The test methodology is available on request, so you can compare it with your own traffic and network path.
Image checks run inside the capture step, so the end user does not see a separate wait. Video and audio take longer because they carry more media; ask us for numbers that match your clip lengths.
NeuroVerify runs as a managed cloud API, in a private cloud or your own VPC, as a secure container, on-premises, or air-gapped. The right option depends on your data-residency and security rules.
The same six signals and the same unified risk score are available in each model. Neural Defend holds ISO/IEC 27001, ISO/IEC 27701, ISO/IEC 42001 and SOC 2 Type II; see the Trust Center for details.
NeuroVerify returns one unified risk score plus signal-level evidence, so a reviewer can see which of the six signals raised concern. It does not make the final decision. Your own policy decides whether to approve, step up, route to manual review or decline.
Because thresholds are yours, you can tune them to your own tolerance for false positives and false negatives. See deepfake detection errors.
Yes. We recommend evaluating on your own media: your real capture conditions, devices and attack samples. Start with a small test set, compare the unified score and signal evidence against your current checks, then set thresholds with your own policy team.
NeuroVerify ranked #1 of 36 systems on VendorBench-100 (detection score MCC 0.876 versus 0.646 for the next-best vendor), but a public benchmark is a starting point. Your own media is the real test. Contact us to set one up.
Evaluate NeuroVerify
Send us representative selfies and video from your registration or telehealth flow. We will show the unified score and signal evidence so your team can judge fit.
