Run AI Agents Locally. Patient Data Never Reaches a Cloud AI API.
For organizations where patient data cannot be sent to an external AI service for processing — Clinexa deploys its Agentic AI stack with a locally hosted LLM. The model runs on hardware inside your network. Clinical data is processed on-device. Seven real-world deployment scenarios.
Reference Hardware: NVIDIA DGX Spark — 1 Petaflop AI compute · 128GB unified memory · Runs LLaMA, Mistral, Qwen locally · 100% local AI inference
Book Edge AI Consultation →Correctional Health Services
Federal Penitentiaries · Provincial Detention Centres · Youth Custody Facilities
“Full AI-powered clinical intake. Inside the wire.”
The Problem
Correctional Service Canada and provincial correctional health services operate on strictly air-gapped networks — by policy, no external cloud connectivity is permitted. Inmates receive medical care, mental health services, and addiction treatment, but health data cannot cross the perimeter. Traditional cloud-based clinical AI is simply not permitted.
The Edge AI Solution
Clinexa Edge AI deploys the full clinical intelligence stack on hardware within the correctional facility's secure network perimeter. Mental health screening, substance use assessment, chronic disease management, and clinical documentation all happen locally — zero data ever leaves the facility.
Why Clinexa for This Use Case
Clinexa's Mental Health Complete Bundle (19 forms including PHQ-9, AUDIT, DAST, PCL-5) directly addresses the most common clinical needs in correctional populations: depression, anxiety, PTSD, substance use disorders, and chronic pain. The AI generates SOAP notes and risk flags inside your secure environment.
Compliance: Meets Correctional Service Canada health service standards and provincial correctional health policies. No PHI ever crosses network boundaries.
Edge AI Capabilities for This Deployment
- Mental health intake: PHQ-9, GAD-7, PCL-5, AUDIT, DAST-10 — auto-scored locally
- Substance use and addiction screening — flagged before provider consultation
- Chronic disease management forms — diabetes, cardiovascular, respiratory
- AI-generated SOAP notes without any external API calls
- Secure role-based access: health staff, nurse practitioners, physicians
- Complete audit trail meeting Correctional Service Canada documentation standards
- Works on existing correctional LAN infrastructure
Deployment: NVIDIA DGX Spark, equivalent NVIDIA hardware, or customer-specified edge compute. We configure to your environment.
Canadian Military & DND Medical Facilities
CAF Base Clinics · Field Medical Units · DND Classified Networks · Veterans Affairs
“Military-grade clinical AI. On the base. On the network.”
The Problem
Canadian Armed Forces medical facilities operate on DND classified networks with strict separation from public internet. Personnel screening for PTSD, traumatic brain injury, operational stress injuries, and functional capacity requires validated clinical assessments — but cloud AI processing is prohibited on DND networks.
The Edge AI Solution
Clinexa Edge AI runs on DND's internal network infrastructure using the NVIDIA DGX Spark or equivalent approved hardware. The full suite of clinical AI agents processes operational health assessments locally — PTSD screening, concussion evaluation, functional capacity, and return-to-duty protocols — without any data leaving the DND network.
Why Clinexa for This Use Case
Clinexa's Concussion Bundle (SCAT5, VOMS, MoCA, Return-to-Play), Mental Health Bundle (PCL-5 for PTSD, PHQ-9, GAD-7), and Functional Assessment Bundle (WAI, FIM, Return-to-Work Protocol) are precisely the clinical tools CAF medical staff need for operational health management.
Compliance: Designed for deployment on DND Protected B/C networks. No external API calls. Configurable for ITAR-restricted environments.
Edge AI Capabilities for This Deployment
- PTSD screening: PCL-5, Impact of Event Scale-Revised — processed on DND network
- Concussion and TBI: SCAT5, VOMS, MoCA — scored by local AI agent
- Operational stress injury assessment — confidential, no external exposure
- Functional capacity for return-to-duty determinations
- Field hospital deployment: runs on portable edge hardware in forward operating bases
- Multi-classification network support — configuration for different security zones
- Veterans Affairs transition forms — processed locally before discharge
Deployment: NVIDIA DGX Spark, equivalent NVIDIA hardware, or customer-specified edge compute. We configure to your environment.
Indigenous Community Health & OCAP Compliance
Nunavik · Nunavut · NAN First Nations · Remote Reserve Communities · Fly-In Health Centres
“Your community's health data. Processed on your land. Under your control.”
The Problem
Indigenous communities operate under OCAP principles — Ownership, Control, Access, and Possession. Many communities in Nunavik, Nunavut, and Nishnawbe Aski Nation territory have satellite-only internet with strict policies against health data crossing to southern data centres or being processed by external AI systems. Sovereignty over health data is not a preference — it's a right.
The Edge AI Solution
Clinexa Edge AI processes all clinical intake locally on community-owned hardware. Health data never crosses to southern data centres. The AI — running open models locally — generates SOAP notes and clinical summaries that stay within the community's information systems. When the community approves, structured (non-PHI) aggregated data can sync for regional health reporting.
Why Clinexa for This Use Case
Clinexa's Primary Care / AWV Bundle (22 forms), Mental Health Bundle, and Pediatric / Developmental Bundle cover the primary health priorities in remote Indigenous communities: chronic disease management, mental health and addiction, maternal health, and pediatric wellness checks.
Compliance: Aligned with OCAP principles, FNIHB (First Nations and Inuit Health Branch) data sovereignty requirements, and NIHB program documentation standards.
Edge AI Capabilities for This Deployment
- Satellite-resilient: full AI processing when internet is down, sync when restored
- Community-controlled data — no southern data centre processing
- OCAP-aligned: data stays within community jurisdiction by design
- Mental health and addiction screening for high-risk populations
- Maternal health and prenatal intake forms — processed locally
- Pediatric developmental screening — M-CHAT, ASQ-3, Pediatric Symptom Checklist
- Chronic disease management: diabetes, cardiovascular, respiratory — all local
- French language support for Northern Quebec communities
Deployment: NVIDIA DGX Spark, equivalent NVIDIA hardware, or customer-specified edge compute. We configure to your environment.
Forensic Psychiatric Units
NCR Facilities · Provincial Forensic Hospitals · Secure Mental Health Units · Assessment Courts
“Legally mandated isolation. Clinically complete AI.”
The Problem
Forensic psychiatric units manage NCR (Not Criminally Responsible) patients under court orders that mandate strict information barriers. Data about patients in forensic psychiatry cannot be exposed to any external system — not for legal, not for clinical, not for administrative purposes. Most cloud AI platforms are simply excluded by policy.
The Edge AI Solution
Clinexa Edge AI provides complete mental health assessment and documentation capabilities within the forensic unit's secure network. Risk assessments, cognitive evaluations, treatment progress tracking, and legal documentation are all generated by local AI agents — with no possibility of external data exposure.
Why Clinexa for This Use Case
Forensic psychiatry requires exactly the clinical tools Clinexa specializes in: standardized risk assessment instruments, longitudinal mental health tracking, cognitive screening, and comprehensive SOAP documentation. The Mental Health Bundle (19 forms) covers the full forensic clinical workflow.
Compliance: Meets Ontario Review Board and provincial forensic review body documentation requirements. Information barrier architecture prevents cross-contamination between clinical and legal data streams.
Edge AI Capabilities for This Deployment
- Risk assessment tools: PCL-5, AUDIT, MDQ — auto-scored, no external processing
- Cognitive screening: MoCA, Trail Making Test — scored by local AI
- Longitudinal tracking: PHQ-9 and GAD-7 over treatment course
- Court-ready clinical documentation — AI-generated SOAP notes locally
- Strict role-based access: psychiatrists, nurses, social workers, security staff
- Legal hold documentation — timestamped, immutable audit trail
- Separate data environments for assessment, treatment, and legal documentation
Deployment: NVIDIA DGX Spark, equivalent NVIDIA hardware, or customer-specified edge compute. We configure to your environment.
Remote Industrial & Occupational Health
Northern Ontario Mine Sites · Offshore Oil Platforms · Arctic Exploration Camps · Forestry Operations
“Full occupational health AI. 200 km from the nearest cell tower.”
The Problem
Mine sites, offshore platforms, and remote industrial operations employ thousands of workers who require occupational health services — injury intake, functional capacity evaluations, return-to-work assessments, and WSIB/WCB documentation. These sites have no reliable internet. Clinic staff currently use paper forms and send physical documents to urban centres for processing — creating days of delay in critical clinical decisions.
The Edge AI Solution
Clinexa Edge AI on portable edge hardware (fits in a Pelican case) gives remote site medical staff the full clinical AI capability of a tertiary care centre. Forms are completed on site, AI agents process them instantly, SOAP notes are generated, and WSIB documentation is ready — all without any connectivity.
Why Clinexa for This Use Case
Clinexa's Functional Assessment Bundle (19 forms: WAI, FIM, DASH, Roland-Morris, NDI, Oswestry) and Post-Incident Assessment Bundle (Whiplash DQ, IES-R, Acute Stress) are directly designed for occupational health and return-to-work workflows — exactly what remote industrial sites need.
Compliance: Meets WSIB (Ontario), WCB (BC, Alberta), CNESST (Quebec) documentation standards. Forms and SOAP notes generated locally are legally equivalent to cloud-generated documentation.
Edge AI Capabilities for This Deployment
- Functional capacity evaluation: FIM, WAI, DASH, Oswestry — locally scored
- WSIB/WCB documentation generated by AI agent without internet
- Post-incident assessment: whiplash, acute stress, tinnitus — completed on site
- Return-to-work and return-to-full-duty protocols
- Portable deployment: NVIDIA DGX Spark fits in a standard equipment case
- Batch sync to corporate health system when satellite window is available
- MSK and pain assessment: NPRS, McGill, Pain Catastrophizing Scale
Deployment: NVIDIA DGX Spark, equivalent NVIDIA hardware, or customer-specified edge compute. We configure to your environment.
Home Care & Palliative Nursing
Rural Atlantic Canada · Northern Ontario Home Care · Indigenous Home Care Programs · Hospice Services
“The AI that goes where your nurses go. Even when the internet doesn't.”
The Problem
Home care nurses in rural New Brunswick, PEI, Nova Scotia, and Northern Ontario visit patients in locations with no reliable internet — farmhouses, remote communities, island communities accessible only by ferry. They carry clipboards and paper forms, return to the clinic, manually enter data, and wait for documentation — a process that creates errors, delays, and administrative burden on already strained nursing staff.
The Edge AI Solution
Clinexa Edge AI on a nurse's laptop or tablet processes intake forms completed during home visits. The AI generates SOAP notes, functional assessments, and care plan updates instantly — while the nurse is still at the patient's home. When the nurse returns to connectivity, structured clinical summaries sync to the health record system.
Why Clinexa for This Use Case
Clinexa's functional assessment forms, chronic disease management tools, and palliative care questionnaires cover exactly what home care nurses document: medication management, daily living assessments (IADL), fall risk, pain scales, and end-of-life care planning.
Compliance: Meets provincial home care documentation standards. PIPEDA and PHIPA-aligned with local processing. Supports NS Home Care Program, NB Extra-Mural Program, and Ontario HCCSS documentation requirements.
Edge AI Capabilities for This Deployment
- IADL (Instrumental Activities of Daily Living) — completed and scored offline
- Fall Risk Assessment — processed by local AI, flags high-risk patients instantly
- Medication review and reconciliation forms — no connectivity required
- Palliative care comfort assessments — AI generates care plan summaries locally
- PHQ-9 and mood assessments for isolated elderly patients
- Sync-when-connected to Accuro, Cliniko, or provincial home care systems
- Works on standard nursing laptops — no specialized hardware required
Deployment: NVIDIA DGX Spark, equivalent NVIDIA hardware, or customer-specified edge compute. We configure to your environment.
Academic Medical Research — IRB-Mandated Data Isolation
University Health Networks · Hospital Research Institutes · Sensitive Cohort Studies · Multi-Site Trials
“Your ethics board said the data stays. So does the AI.”
The Problem
Many IRB/REB approvals for sensitive research cohorts — addiction studies, HIV research, mental health, racialized communities, LGBTQ2+ health — explicitly prohibit research data from being transmitted to external systems or crossing international borders. Standard cloud AI platforms violate these ethics conditions by design. Researchers either forgo AI assistance entirely or risk their ethics approval.
The Edge AI Solution
Clinexa Edge AI runs on research site hardware, processing all participant intake data locally. AI agents analyze responses, identify trends, generate participant summaries, and flag eligibility criteria — all without the data leaving the research site's approved computing environment. Ethics conditions are met by architecture, not policy.
Why Clinexa for This Use Case
Clinexa's longitudinal intake capabilities — time-based survey scheduling, cohort tracking across study phases, and automated scoring of validated instruments — make it ideal for research applications. The AI can track PHQ-9 trajectories, AUDIT trends, and functional capacity changes across a study cohort without any data leaving the approved environment.
Compliance: Meets Tri-Council Policy Statement (TCPS 2) data handling requirements. Supports CIHR, SSHRC, and NSERC funded research data governance requirements. FDA 21 CFR Part 11 compliant audit trails.
Edge AI Capabilities for This Deployment
- Longitudinal participant tracking: PHQ-9, GAD-7, AUDIT trends across study phases
- Ethics-compliant by architecture — no external data transmission
- Eligibility screening against protocol criteria — processed locally
- Sensitive cohort data: HIV, addiction, mental health — stays on research network
- Cross-border restriction compliance: no US or EU cloud processing
- De-identification pipeline runs locally before any data export
- Multi-site research: each site runs local Edge AI, aggregated de-identified data syncs
- REB-ready data handling documentation — automatic audit trail
Deployment: NVIDIA DGX Spark, equivalent NVIDIA hardware, or customer-specified edge compute. We configure to your environment.
We handle the full Edge AI deployment
Hardware recommendation, local LLM selection and configuration, Clinexa AI agent integration, network validation, and staff training — all delivered in your environment. We don't just sell software. We configure your local AI clinical system end-to-end.