Operate Amplinth Radiant · Healthcare · Radiology

AI-assisted triage for abdomen-pelvis CT.

Radiant runs inference on each CT study, sorts the radiologist's worklist by clinical urgency, shows where the model looked and how its findings agree with the patient's clinical picture, then syncs a structured result into your PACS and EHR. The radiologist stays accountable — this is a triage decision-support aid, not a diagnostic device.

● Deployed & running in the cloudUrgency-tiered worklist · explainable findingsSaaS · any cloud · your cloud / on-prem appliance
The problem

Flat worklists, rising volume, findings divorced from context.

In acute care, a late or missed time-critical abdominal finding is measured in patient harm and liability — yet radiologists read against an unprioritised queue, and a bare probability tells them little about how urgent a study really is.

01
Worklists are unorderedStudies arrive first-in-first-out, so a ruptured AAA can sit behind routine reads. Urgency is unknown until someone opens the study.
02
Opaque AI is untrusted AIRadiologists won't act on a black-box probability. Without visible evidence — where the model looked, and why — an AI score is just noise on the screen.
03
Findings ignore clinical contextA probability without the patient's lactate, WBC, or pregnancy status is hard to act on. Detection alone doesn't say whether the picture fits.
04
Manual re-entry into PACS/EHRRe-keying results into reporting and record systems is slow and error-prone, and it breaks the chain of evidence behind a decision.
How Radiant works

One disciplined loop, from study to a prioritised, explained read.

The Amplinth loop, tuned to radiology triage. A deterministic rules engine — not the language model — owns every clinical decision, so each result is auditable and defensible.

01

Infer

Six-stage pipeline reconstructs the CT and runs Pillar-0 detection — 20 calibrated finding probabilities plus attention.

02

Tier

A deterministic rules engine assigns an urgency tier (RED / ORANGE / YELLOW / GREEN) and re-orders the worklist.

03

Explain

Attention overlays and key slices show where the model looked; probabilities are calibrated to real prevalence.

04

Check & Act

Findings checked against vitals, labs & history for concordance; the radiologist confirms, disagrees or amends.

05

Prove

Standards-based output to PACS/EHR; value framed as triage throughput & turnaround.

Measured via EVA™
Capabilities

An inference-and-review application, built for governance.

Radiant is a focused clinical-AI product for one modality and body region — abdomen-pelvis CT in the acute setting. It complements PACS and reporting systems rather than replacing them.

Six-stage inference pipeline

Preprocess → Pillar-0 detection → MedSAM2 localization → MedGemma grounding → deterministic rules engine → LLM narrative. Each stage persists before the next, so the pipeline is resumable and every step is recorded.

Urgency-tier worklist triage

The worklist returns studies sorted by urgency tier by default, with live Server-Sent-Events updates and per-study pipeline visibility — the most time-critical studies surface first.

Explainability that actually overlays

Per-finding Score-CAM / Grad-CAM attention heatmaps and key-slice ranking, rendered directly in an OHIF / Cornerstone3D CT viewer — no PNG export, no viewer swap. Probabilities are isotonically calibrated to observed prevalence.

Deterministic clinical concordance

For each positive finding, the rules engine compares the AI call to the patient's vitals, labs and history and returns concordant / discordant / unassessable with auditable, non-LLM reasoning. Never-miss findings use lowered thresholds for sensitivity.

Three-panel study review

Primary finding with anatomy, measurements and evidence; related sub-threshold findings; a collapsible list of cleared differentials; and a recommended communication tier — beside the CT viewer with attention and segmentation overlays.

Radiologist workflow actions

Mark reviewed, disagree, amend or escalate. Amendments propagate automatically into every export; escalations push a live worklist update. A full per-study review history feeds clinical-quality reporting.

Standards-native export

DICOM-SR, DICOM-SEG and GSPS to PACS; a FHIR R4B DiagnosticReport to the EHR; DICOMweb retrieval, HMAC-signed webhooks and a PDF narrative — so downstream systems stay in sync without re-entry.

Governance, RBAC & immutable audit

JWT auth with role-based access (radiologist / admin / integrator), scoped API keys, model-version and threshold management with never-miss safeguards, and an append-only audit log enforced by both a DB trigger and privilege separation.

20-finding abdomen-pelvis taxonomy — active hemorrhage, ruptured AAA, pneumoperitoneum, mesenteric ischemia, appendicitis, cholecystitis, bowel obstruction and more — each with tier, SNOMED code, sex-specificity, threshold and never-miss flag as code-level clinical-safety invariants. DIMSE C-STORE push ingestion and multi-tenant SaaS are on the near-term roadmap (coming soon).

Outcomes & maturity

Honest about what's measured, modelled, and yet to be proven.

Radiant runs end-to-end with a full radiologist and admin UI and a deployed Pillar-0 inference endpoint. It is not an FDA-cleared / UKCA / CE-marked diagnostic device, and every model-performance figure below is a validation / held-out-test result — not a realized clinical outcome.

0.71
Model macro AUC (held-out test)
v3.7 bundle · 5,428 studies · validation-set, not a clinical outcome
20
Calibrated abdomen-pelvis findings
Code-defined taxonomy · capability
30–60s
End-to-end pipeline latency
Design target · not for real-time use
<500ms
Deterministic rules-engine decision
Performance target · authoritative stage
What we claim — and what we don't
The value is faster, better-ordered triage: time-critical studies surface first, with evidence and clinical context side by side.
Time-to-triage, radiologist review time and AI-radiologist concordance are instrumented through the clinical-quality reporting and immutable audit trail — to be measured in a clinical pilot. We make no diagnostic-accuracy or clinical-efficacy claim.

Engineering rigour: ~25 registered FastAPI routers; a six-stage resumable pipeline with a structured error taxonomy; CI standards targeting 80% coverage on services/rules plus integration and Playwright E2E tests. Calibration: a reported probability matches observed prevalence post-isotonic — honest values, not overconfident raw scores. No hospital, site or client is named; reference material is illustrative only.

Deployment & clinical governance

Your patient data. Your cloud. Your control.

Radiant runs as SaaS on Amplinth's AWS, in any cloud, or as a single-tenant customer-VPC / on-prem appliance — so PHI can stay inside your estate. It slots into existing PACS/EHR over open standards, with the radiologist accountable at every step. Regulatory clearance (FDA 510(k) / UKCA / CE) is a separate workstream and not in place; Radiant is positioned as a clinical decision-support / triage aid.

DICOM-SR / SEG / GSPSFHIR R4BDICOMwebHuman-in-the-loopImmutable auditPHI retention & purgeSOC 2ISO 27001
SaaS
Amplinth SaaS on AWSFully managed — fastest path to value.
VPC
Your own cloud / VPCSingle-tenant; PHI never leaves your control.
On-prem appliancePer-hospital deployment with a local GPU endpoint.
Why Amplinth

Explainable, human-in-the-loop, standards-based.

Decisions you can defend

A deterministic rules engine owns tier and concordance with a recorded rules_fired list; the language model writes prose only and can never override it. Every result is auditable to the prompt.

Evidence, not a black box

Attention overlays and key slices show where the model looked, in your own viewer. Calibrated probabilities mean a reported number matches the real-world rate — honesty radiologists can trust.

Built to fit, not to replace

DICOM-SR/SEG/GSPS, FHIR R4B and DICOMweb let Radiant slot into your existing PACS/EHR estate. The radiologist remains the decision-maker — always.

See Radiant on your worklist

Surface the time-critical study first — with the evidence in view.

Book a walkthrough on de-identified demo cases: the tier-sorted worklist, in-viewer attention overlays, deterministic clinical concordance, and one-click standards-based export to PACS and the EHR.