AI · Medical · Ledger
先端知能基盤
The intelligence layer
for critical systems.
Xvian runs diagnostic-grade models inside hospitals, banks and research networks — every inference signed, versioned and anchored on-chain.
99.2%
Detection AUC
2.4B
Studies indexed
18
Clinical partners
◈xvian://core.monitorlive
throughput18,402 req/min
gpu pool74% · healthy
drift alert1 open
◈ledger.feedanchored
✓ 0x7a1f… model:thorax-v4.2 weights anchored
✓ 0x3c92… dataset:retina-2026Q2 hash
✓ 0xb14e… inference batch 4,182 signed
✓ 0x9dd0… clinician override logged · SG-04
✓ 0x2f77… audit pack exported · TGA
01 / Divisions
View all →
Four units, one ledger
UNIT_01
Medical AI医療知能
Imaging, pathology and waveform models cleared for clinical deployment.
UNIT_02
Core Models基盤模型
Multimodal foundation stack, fine-tuned per site without data egress.
UNIT_03
Ledger信頼台帳
On-chain provenance for weights, datasets and every prediction served.
UNIT_04
Capital資本部門
We back the infrastructure beneath applied AI — silicon, rails, deployment.
02 / Registry
updated 21:40 AEST
Model registry — public slice
| Model | Domain | AUC | p95 | Anchor | Status |
|---|---|---|---|---|---|
| thorax-v4.2 | Chest CT | 0.992 | 41ms | 0x7a1f… | CLEARED |
| retina-v3.8 | Fundus | 0.981 | 28ms | 0x3c92… | CLEARED |
| histo-v2.1 | Pathology | 0.964 | 112ms | 0xb14e… | REVIEW |
| cardio-sig-v1.6 | ECG stream | 0.958 | 9ms | 0x9dd0… | CLEARED |
| onco-fusion-v0.9 | Multimodal | 0.941 | 210ms | 0x2f77… | TRIAL |
03 / Deployment
Inside the perimeter
[ 1400×900 ]Server hall / GPU rack, blue-lit
shallow depth, cool grade
shallow depth, cool grade
No data egress. Models ship to the site, run on local silicon, and report back only signed metrics.
- A1Air-gapped appliancesHardware attestation on every boot.
- A2Federated fine-tuningWeights move; patient data never does.
- A3Drift monitoringClinician overrides logged and counted.
- A4Evidence packsRegulator-ready exports generated automatically.
04 / Access
Deploy Xvian on your infrastructure
Evaluation programs open quarterly for hospital networks, labs and institutional partners.
Request an evaluation →