Temporal detection
Detect sustained changes-not merely isolated spikes-using historical baselines and transparent surveillance models.
- EWMA and CUSUM
- Change-point detection
- Seasonal adjustment
Salus Agentic connects public health laboratories, electronic reporting, disease surveillance, epidemiology, and human-supervised AI in one federated platform-so a signal detected in one laboratory can become coordinated public-health action across an entire region.
No gated presentation. No fictional screenshots. Enter the working platform.
Inherited from our Snowflake data cloud foundation under a signed BAA. See our full compliance posture.
Every day, they see the first fragments of a larger story: one unusual result, then another; a wastewater increase; a cluster spanning counties; a new variant appearing quietly in the background.
Too often, those fragments remain separated by incompatible systems, manual spreadsheets, delayed interfaces, and organizational boundaries. By the time the pattern becomes obvious, the opportunity to act early may already be gone.
Salus changes the role of the laboratory information system. It unifies the work inside the laboratory, the reporting that leaves it, and the epidemiological intelligence that must follow. The result is not simply a modern LIMS. It is a living public-health network.
CDC describes ELR as a way to reduce manual entry errors and improve standardized, complete laboratory reporting. CDC also emphasizes that public-health data modernization must address outdated technology, complicated processes, and systems that do not work well together. Salus is designed directly around that mission.
Every screen below is a live capture of the working application - the same dashboard, results worklist, and ELR feed a technologist or lab director uses every day.
Each layer is valuable on its own. Together, they create a continuous path from laboratory operations to evidence-backed public-health action.
Accessioning, testing, instruments, quality, chain of custody, result verification, inventory, and laboratory operations.
HL7, FHIR, ELR, eCR, instruments, terminology, secure routing, acknowledgments, and inter-laboratory exchange.
Historical baselines, trend analysis, anomaly detection, geospatial clusters, wastewater signals, and forecasts.
Cases, investigations, timelines, exposure relationships, hypotheses, outbreak workflows, and reporting.
Human-supervised agents that assemble evidence, correlate signals, explain risk, and accelerate investigation.
Salus combines laboratory truth with complementary public-health signals. It does not let an AI “declare” an outbreak. It builds an explainable evidence package for epidemiologists to review.
Detect sustained changes-not merely isolated spikes-using historical baselines and transparent surveillance models.
Reveal clusters across counties, facilities, schools, sewersheds, and neighboring jurisdictions while controlling geographic precision.
Compare wastewater movement with clinical and laboratory activity to identify community changes earlier.
Connect lineage, mutation, resistance, and sequence-cluster information with laboratory and geographic context.
Bring environmental, veterinary, foodborne, water, and vector signals into the same investigative picture.
Generate short-horizon forecasts with prediction intervals, model comparisons, and visible limitations.
Laboratories, clinical systems, wastewater programs, and environmental sources contribute governed data.
Salus validates, normalizes, deduplicates, geocodes, and preserves the provenance of every observation.
Statistical, geospatial, graph, and cross-source models identify meaningful departures from expected activity.
Epidemiologists review the evidence, open investigations, coordinate laboratories, and authorize escalation.
Salus agents are scoped workers-not an all-powerful chatbot. Every agent has defined tools, permissions, evidence requirements, and approval boundaries.
Extracts manifests, label sheets, and files into a review queue and pre-fills accessioning.
Human-verified intakeFlags anomalous results, drafts verification work, and brings urgent exceptions forward.
Never releases aloneMonitors QC drift, out-of-spec events, documentation gaps, and recurring operational risks.
Evidence preservedValidates HL7/FHIR structures and LOINC/SNOMED mappings before reporting.
Rules remain authoritativeTracks turnaround, capacity, bottlenecks, inventory, and testing-network load.
Advisory actionsAssembles surveillance evidence, identifies correlations, and recommends investigation steps.
Cannot declare outbreaksDetects schema drift, duplicates, missing fields, delayed feeds, and source-quality changes.
No silent correctionMonitors agent scope, autonomy, performance, policy compliance, and human overrides.
Kill switch remains humanEvery laboratory and jurisdiction retains control of its detailed records. Salus shares only the signals, summaries, and authorized evidence required for collaboration.
Qualified public-health organizations receive the complete platform-not a stripped-down edition. Deployment scope, infrastructure consumption, implementation, and support are structured transparently around the organization.
Enter the live Salus environment and experience laboratory operations, interoperability, surveillance, and agentic intelligence as one connected system.