Safety Modeling Assurance Toolkit (SMAT) is an AI-powered platform designed to model clinical pathways, surface hazards early, and predict adverse outcomes before they ripple through care delivery. It helps healthcare organizations move from reactive reviews to proactive safety assurance.
SMAT factors in people, tasks, tools, environments, and processes to understand how care truly works in practice.
Current development is focused on two high-risk pathways with real-world validation in view.
Novart Technologies is reframing its landing page around a focused startup vision: a digital toolkit that proactively models clinical pathways, identifies potential hazards, and predicts adverse outcomes in healthcare delivery. The goal is to strengthen patient safety, build confidence in digital care, and support safer adoption of emerging technologies.
Healthcare systems are under pressure to modernize quickly, yet every new workflow, digital tool, and process change can shift risk in ways that are hard to see. SMAT is designed to help clinicians, safety leaders, and administrators spot those risks earlier—before they become harmful outcomes.
The platform combines modern AI, machine learning, and explainability to move from fragmented signals to actionable safety insight. Each module mirrors a clear capability within the startup vision and can evolve as new data and feedback arrive.
SMAT connects data sources such as electronic health records, clinical guidance, incident reports, and workflow logs to reconstruct real-world care pathways.
The toolkit uses anomaly detection, supervised learning, and GenAI-assisted scenario discovery to reveal safety risks that may otherwise stay hidden.
SMAT forecasts care outcomes and continuously learns from fresh data and user feedback, while keeping predictions interpretable for users.
The product vision is simple: ingest trustworthy clinical inputs, model the pathway, detect safety threats, simulate possible outcomes, and return clear dashboards or alerts that teams can act on.
Upload or connect to clinical data sources including EHRs, workflow logs, and safety reports.
AI reconstructs the clinical pathway and maps the important interactions that shape care delivery.
Machine learning identifies deviations, bottlenecks, and signals that point to emerging hazards.
Generative models simulate possible failure scenarios and predictive models estimate likely outcomes.
Teams receive visual dashboards and actionable alerts for clinical, operational, and safety decisions.
The startup concept is being developed with practical validation in mind, combining a systems approach to patient safety with a modern AI stack and targeted MVP execution.
SMAT maps and analyzes the components that influence patient safety across people, tasks, tools and technologies, environments, and operational processes. This systems perspective helps move beyond siloed incident review toward whole-pathway understanding.
Designed to provide transparent reasoning for hazard alerts and predictions.
Incorporates new data and user feedback to evolve with clinical practice.
Current build targets two high-risk pathways as an initial proof of value.
Development is being informed by collaboration with secondary and tertiary healthcare providers.
We are building the Safety Modeling Assurance Toolkit (SMAT) to help healthcare organizations anticipate harm earlier, validate digital change more safely, and strengthen confidence in modern care delivery. If you believe in safer systems, smarter tooling, and practical innovation in healthcare, we invite you to contribute to the next stage of development.
Share your details and sponsor message below, and we will respond by email soon.
