Eight categories, more than fifty device types, over a thousand devices tested. Class II and Class III, across every FDA specialty. Each carries its own architecture and its own way to reach a patient. Here is how we test them.
Software, motion, and instrument activation meet in the same clinical moment. We test where a network fault can become a patient-safety event.
Surgeon console, patient-side cart, instrument controllers, and a latency-sensitive vision pipeline.
We decompose the console, cart, and instrument-control nodes into a digital twin, then drive the real USB, service, and CAN interfaces across each procedure state. Coverage maps every path that could touch the motor-control boundary.
Preoperative planning, intraoperative tracking, and robotic cutting or guidance hardware.
We model the planning-to-intraoperative transfer and test implant-library and plan integrity end to end. Testing centers on what has to stay trustworthy between the plan and the cut.
Navigation carts, optical tracking, and calibration data feeding a guidance display.
We exercise the calibration and tracking inputs and validate how the system authenticates the data it steers on. The twin models every source the guidance view trusts.
Insufflators, electrosurgical generators, and stateful instrument controllers.
We build the procedure-state machine into the twin and fuzz each state on its own, pre-case, intra-case, and service mode. That is where behavior actually diverges.
Surgical laser instruments and RF or laser aesthetic platforms.
We test the activation and parameter paths over the internal bus, and check service ports for default credentials before they ship line-wide.
Powered drivers, staplers, and handpieces with embedded controllers.
We map the controller and accessory interfaces into the twin and exercise the command and identity paths under real activation sequences.
Cath-lab systems, contrast injectors, and procedure-data synchronization.
We pair the injector and procedure-record twin and test the injection-command and DICOM-sync paths from a credible local foothold. Scope stays tied to the interface under test.
Shock-wave and laser lithotripsy consoles with treatment controllers.
We model the treatment controller and exercise its parameter and interlock paths across the therapy sequence.
DICOM everywhere, enormous parsers, and clinical reads downstream. We test the acquire, move, and read chain as one system.
Archives, cardiology viewers, and enterprise image distribution.
We run the archive, router, and viewer as one twin and stress the DICOM parsers, cache integrity, and study-routing paths. Coverage includes the read path a radiologist actually trusts.
Acquisition console, reconstruction pipeline, and vendor network segment.
We model the console and recon nodes and exercise the DICOM ingest and vendor-network interfaces, including how the segment is expected to be isolated.
Scanner host, in-suite controllers, and procedural peripherals.
We map the suite as a twin and test the trust assumptions between the scanner host, in-bore peripherals, and the vendor segment.
Cart and handheld ultrasound, intravascular imaging, and catheter consoles.
We test the probe pairing, mobile-app trust, and image-export paths, down to the key exchange and the exact transport settings.
Acquisition units with computer-aided detection in the read path.
We validate how the CAD path loads and verifies its model artifact, and model the integrity controls the read depends on.
Fluoroscopy systems paired to hemodynamic recorders.
We test the pairing and data-sync link between the imaging system and the recorder, and keep scope tied to that interface.
Segmentation, 3D-model, and surgical-planning software.
We test the model-import and export paths and the kiosk boundary, and validate how patient models are authenticated before they reach a clinician.
Retinal OCT, fundus, and ophthalmic diagnostic imaging.
We test the study-export and device-management paths and validate the integrity controls around the diagnostic image.
Slide scanners and enterprise pathology viewing platforms.
We run the scanner-to-viewer twin and test the access controls and object-reference handling around each case.
High-energy delivery where software decides where the beam stops. We test to the ceiling of medical-device risk.
Synchrotron beam-delivery control, gantry motion, and safety interlocks.
Among the most dangerous devices in medicine. We model the beam-delivery control chain and safety interlocks and drive the real interfaces, mapping which paths the interlocks contain and which they do not.
LINAC delivery control, MLC, and record-and-verify integration.
We test the plan-transfer and record-and-verify paths and validate that delivery trusts content, not filenames.
Dose-planning and RT records software.
We test the dose-plan transfer, records integrity, and audit trail, and model the validation the workflow depends on.
PET/SPECT consoles and nuclear imaging workflow.
We model the console and workflow twin and test the acquisition, routing, and export interfaces.
The device leaves the clinic. Radios, long-lived firmware, and a threat model that follows the patient home.
Pacemakers, ICDs, home monitors, and clinician programmers.
We model the implant, home monitor, and programmer as one twin and test the telemetry-radio session and pairing controls under real proximity conditions.
Implanted pump, external controller, and clinical hub.
We test the controller-to-hub trust and configuration paths and scope every result to the interface and the reach it actually has.
DBS, spinal cord, and vagus nerve stimulators with patient controllers.
We model the implant, patient controller, and clinician tablet and test the pairing window and stimulation-parameter path against replay and downgrade.
Implant, sound processor, and clinical fitting software.
We test the processor and fitting-software link and the pairing controls, scoping the stimulation-parameter path to real proximity.
Public and clinical AEDs with connected update and readiness.
We test the firmware-update and readiness-reporting paths and validate the signing the device depends on.
Body-worn sensors, phone apps, and cloud follow.
We test the sensor-to-app trust and the cloud channel, focused on the integrity of the reading a clinical decision rides on.
Wearable therapy, home hubs, and remote patient devices.
We test the cloud channel, certificate handling, and OTA path a home device depends on, and validate the fallback behavior before it becomes a field issue.
Connected stimulators and their control apps.
We test the app-to-device link and the dose-schedule integrity path over BLE.
Availability is a safety property. We test the devices that keep a patient alive and the networks they sit on.
Continuous ventilators with serial and network control.
We drive the serial and network control paths and test how alarm and setting messages are authenticated, rating everything against loss-of-therapy impact.
Anesthesia delivery with agent control and monitoring.
We build the delivery state machine into the twin and exercise the agent-control and monitoring interfaces across the case.
Bedside monitors and central-station telemetry.
We model the bedside-to-central twin and test the waveform and alarm-routing trust across the monitoring VLAN.
Connected SpO2 and capnography with app or network links.
We test the pairing and transport for the connected link, down to the key exchange and transport settings.
Fetal, maternal, and NICU apnea monitoring systems.
We test the trace and alarm-routing trust across the ward network and model the availability the unit depends on.
Connected incubators and radiant warmers with setpoint control.
We test the setpoint and control APIs and validate the authentication around thermal control.
Dialysis machines on the clinical network.
We test the treatment-parameter and monitoring interfaces and validate the integrity controls the therapy depends on.
Neuro monitoring carts and bedside ICP monitors.
We model the monitor twin and test the sensor and network trust, scoping results to the interface under test.
Small parsers, big consequences, and a supply chain of disposables. We test the pump, the server, and the consumable together.
Large-volume and syringe pumps, drug libraries, and pump servers.
We run the pump-and-server twin and test the drug-library update path and dose-limit controls, keeping scope to the interface, not the alarm count.
Automated dispensing and pharmacy workflow systems.
We test the kiosk boundary, the local credential store, and the dispensing-log integrity from realistic physical access.
Connected feeding pumps with app control.
We test the rate-command integrity path over the app and BLE link and validate the checks the pump depends on.
Docks, cartridges, tubing, and authenticated consumables.
We test the consumable-authentication scheme on the bus and model the memory protection that keeps cloning uneconomical.
Software as the device. We test cloud tenancy, mobile clients, and the AI pipelines regulators now expect you to cover.
Cloud-connected clinical software and multi-tenant platforms.
We test the tenant boundary, API authorization, and PHI handling under each role, and model the access controls the platform relies on.
Clinician programmers and patient-facing mobile apps.
We test the build for secrets, the deep-link surface, and local storage, and validate the platform integrity checks.
Model training, inference, and data pipelines inside a device.
We test the model-loading and data-pipeline integrity and frame the controls in the twin a reviewer will read.
Autonomous detection systems that return a clinical result.
We test how the system verifies its model and inputs before it returns a result, and model the integrity path end to end.
Fleet telemetry, RMM, and connected device management.
We test the fleet-push and token-scope paths and model the blast radius a management action can have.
Computer-aided detection add-ons in the read or scope path.
We test how the add-on loads and verifies its model and validate the integrity controls the read depends on.
Benchtop instruments and the embedded platforms under them. We test the instrument, the OS, and the data path.
Sequencers and analyzers on embedded Linux.
We model the instrument twin and test the exposed services and update-verification path on the embedded OS, distilling coverage to the roots that matter.
PCR and molecular diagnostic systems.
We test the run-configuration and result-routing integrity on the lab network.
Identification and susceptibility platforms.
We test the result API and reporting path and validate the authentication the workflow depends on.
Immunoassay and clinical chemistry analyzers.
We test the result-export and instrument-management paths and model the integrity controls around the reported value.
Benchtop hematology and coagulation systems.
We model the analyzer twin and test the exposed services and result-routing integrity.
Donor management and blood-bank workflow systems.
We test the role boundaries and PHI handling and model the access controls the workflow enforces.
Shared embedded platforms underneath many device families.
We model the shared platform once and map every device family that inherits it, so a platform-level control becomes a line-wide fix.
New modality, legacy platform, or something nobody has looked at in years. Bring it to us.