Projects
Selected medical-device, implantable, sensing, data, and AI systems.
Concise project evidence from commercial medical-device work, Ph.D. implantable bioelectronics, early biomedical sensing, physiological data workflows, and selected AI/infrastructure builds.
Selected projects
Project evidence with the numbers pulled forward.
FlowSense Clinical / ACE
Clinical wearable and DSP/ML workflow for noninvasive CSF shunt-flow assessment; improved from a 74% baseline to 0.81 AUC / 82% blinded validation accuracy.
- Expert / Gen1 baseline: 74% accuracy.
- FlowSense Gen2 blinded clinical validation: 0.81 AUC / 82% accuracy.
FlowSense Home / Lynx
Home-use hydrocephalus wearable advanced across 3 design generations, 100+ units, 200+ participants, 2,800+ wear hours, and 70% → 96% reliability improvement.
- Field reliability improved from about 70% in the initial Gen 1 home device to about 96% by the final Gen 3 clinical study.
- Clinical-study build lots were typically 20–30 devices.
Wound Monitoring Platform / Tabby
NIH R43 Phase I multimodal wound patch with thermal/humidity/temperature sensing, 40 mAh battery, 20-subject dataset, 172 logs, and 5-fold wound-model validation.
- NIH R43 Phase I sensing program with prototype-to-grant-deliverable execution in approximately 1 year.
- Device/study metrics: 40 mAh battery, 20 subjects, 172 six-minute logs, and 7 days hourly use per charge.
Wireless Battery-Free Bioelectronics
Shared Ph.D. implantable platform work across 14 journal articles, 20+ fixtures, 6 tuned antennas, 3 simulation frameworks, <50 mg implants, and up to 2 m wireless power.
- Publication output: 14 peer-reviewed journal articles, 5 conference proceedings/abstracts, and 1 book chapter.
- Mentored or managed 10+ researchers/students across graduate research work.
Subdermal Photometry Implant
PNAS wireless battery-free subdermal photometry platform for chronic neural recording; approximately 10.5 mm × 7 mm with ~27 Hz, 12-bit data streaming.
- PNAS 2020 publication: wireless, battery-free subdermally implantable photometry for chronic neural dynamics.
- Approximate platform size: 10.5 mm × 7 mm.
Implantable Electrical Neurostimulation
Microsystems & Nanoengineering wireless battery-free neurostimulation platform with ~5.5 V compliance and ~18 mW wireless harvesting in relevant test conditions.
- Microsystems & Nanoengineering 2021 publication: wireless, battery-free, fully implantable electrical neurostimulation in freely moving rodents.
- Approximate stimulation compliance: 5.5 V.
Osseosurface Electronics
Nature Communications thin wireless battery-free osseosurface biointerface with multimodal sensing and up to ~87 Hz, 14-bit communication/readout.
- Nature Communications 2021 publication: thin, wireless, battery-free, multimodal osseosurface electronics.
- Communication/readout metric: up to approximately 87 Hz with 14-bit resolution.
High-Power FES Implant
Nature Communications fully implanted battery-free high-power FES platform with ~20 V compliance, 10 µA–1 mA spinal output, and 1–5 mA muscle output.
- Nature Communications 2023 publication: fully implanted battery-free high-power platform for chronic spinal and muscular functional electrical stimulation.
- Approximate compliance: 20 V.
Non-Invasive Neural Recording Hardware
EUNIL research across low-noise neural recording, 4D acoustoelectric current-density imaging, analog front-end optimization, EMI control, DSP/wavelets, and experimental phantoms.
- EUNIL research role: Jan 2016 – Oct 2019.
- Work covered non-invasive neural recording, current-density simulation, low-noise amplifiers, amplifier optimization, EMI control, DSP, wavelets, and experimental phantoms.
Low-Cost Nanoparticle Analyzer Capstone
Biomedical engineering capstone combining fluorescence microscopy, microfluidics, and Brownian-motion tracking in a ~$4,500 prototype with <$5 chambers and <1.9% sizing error across 50 nm to 1 µm particles.
- Undergraduate biomedical engineering capstone completed in 2016–2017.
- Prototype cost: approximately $4,500, with disposable chamber cost below $5 each.
Wearable Physiological Data Analysis
iCAMP analysis of 800+ hours of chest-worn ECG/accelerometer data across 31 subjects, supporting fall-risk modeling from 0.73 baseline AUC to 0.969 mixed-model AUC.
- iCAMP role: 31 subjects and 800+ hours of chest-worn wearable data.
- Fall-risk/frailty model reference: 0.73 baseline AUC, 0.921 posture/activity AUC, and 0.969 mixed-model AUC.
OpenClaw AI Agent System
A local AI-agent workflow system for project execution, memory, automation, and technical planning.
- Current source uses capability-based wording until measured uptime, storage scale, GPU specs, indexed-file count, retrieval performance, or backup frequency are added.
Home Server Infrastructure
Self-hosted infrastructure for storage, media, AI services, documentation, networking, and personal automation.
- Current source uses capability-based wording until measured uptime, storage scale, GPU specs, indexed-file count, retrieval performance, or backup frequency are added.