Why I built it
I wanted a practical system for turning noisy physiological data into a validated, clinically useful shunt-monitoring workflow.
Medical devices
FlowSense Clinical / ACE
Clinical-grade wearable system for noninvasive assessment of CSF shunt patency using thermal transport, with algorithm validation and FDA-ready workflows.
completed 2022–2025 Medical devicesAlgorithmsValidationThermal sensing
Type product / algorithm
Skills Wearable sensing, Thermal physiology, ML/DSP, Clinical validation
Project summary
Why it exists, what I built, and what I learned.
What I built
Clinical and home-use sensing workflows, model-training pipelines, feature engineering, validation tooling, and FDA-support documentation.
What worked
The strongest part was connecting thermal sensor physics, model validation, and real-world patient data into one workflow.
What failed
Small and biased datasets required careful grouping, leakage control, and feature selection to avoid overfitting.
What I learned
Algorithm work is strongest when it is grounded in the sensing physics and the clinical workflow, not just model metrics.
Stack
Tools, systems, and technical areas involved.
Thermal sensingFeature engineeringONNXPythonValidationFDA documentation
Links and direction
Public links and next steps.
Carry the same validation pattern into adjacent hydrocephalus and wearable-monitoring problems.
Related projects
Other projects in the same neighborhood.
Medical devices
FlowSense Home / Lynx
Longitudinal home-use wearable for continuous hydrocephalus monitoring in daily life and sleep.
Medical devicesWound Monitoring Platform / Tabby
Soft multimodal sensing patch for continuous wound monitoring and prediction of wound-healing progression.
ImplantablesWireless Battery-Free Bioelectronics
Fully implantable wireless and battery-free bioelectronics for neural, skeletal, and physiological monitoring.