01 FLAGSHIP · Closed-Loop Embedded
VVIR Cardiac Pacemaker
Duke · BME 354A closed-loop pacemaker built across six modules and demonstrated on live (pithed) frog-heart tissue — from microvolt sensing to rate-adaptive pacing.
Raw signal straight off the electrode — buried in noise, drift, and 60 Hz hum.
A Sensing + VVI Pacing
A bandpass sensing front-end feeding VVI pacing firmware on an Arduino Nano Every.
- Bandpass sensing (~1–48 Hz) to detect intrinsic cardiac activity.
- VVI pacing firmware with refractory and escape-interval logic.
- Strength-duration characterization of the stimulus.
B Stimulation + VVIR Rate-Adaption
MOSFET-driven stimulation validated against real stimulator output, with an accelerometer closing the loop for rate-adaptive pacing.
- MOSFET-driven stimulation circuit.
- Validated against real stimulator output.
- ADXL335 accelerometer drives activity-based rate adaptation (VVIR).
- Demonstrated end-to-end on live (pithed) frog heart tissue.
02 FLAGSHIP · Embedded + Full-Stack
mHealth Tympanometer
Duke · Dr. Mark PalmeriWorking on a layperson-operable handheld tympanometer for pediatric middle-ear screening in underserved communities — funded by PCORI and Duke Global Health, with clinical and industry partners. My part: adding an on-device display, standing up the tympanogram-labeling web stack, and building the labeling pipeline for its planned on-device ML classifier.
A Firmware / Embedded
Added an OLED display subsystem to a BLE medical device on Zephyr RTOS / nRF52833.
- Designed a 10-state display state machine that mirrors the measurement flow.
- Ran a display-hardware tradeoff study (mono OLED vs Memory LCD vs e-paper vs TFT) against power, BOM cost, and measurement-noise constraints.
- Ported the display driver from I²C to SPI.
- Diagnosed a blank-screen failure down to a pin conflict with the debugger's internal UART and resolved it on validated hardware.
B Full-Stack / Infra
Stood up a Flask web app for audiologists to label ~5,000 tympanograms — the ground-truth dataset for the device’s planned on-device ML classifier.
- Python · Jinja · HTMX · SQLite with server-side matplotlib rendering, running on a Linux VM.
- Built the labeling pipeline that will feed a planned on-device ML classifier — the classifier is in development; my role is the data pipeline, not training the model.
- Deployment: nginx reverse proxy, HTTPS/SSL, systemd service, cron backups.
02 / RESEARCH / WET-LAB
Research & Wet-Lab
Bench science and neuroengineering — cells, assays, and electrodes.
- 01 Duke · Undergraduate Researcher Aug 2025 – Present
Collective for Psychiatric Neuroengineering
Connexin-43 oligomerization and gap-junction channel formation in neuropsychiatric models — from wet-lab assays to the electrodes that record neural signals.
- Transfected HEK293FT across 16 construct conditions; validated interactions via 40+ Co-IP and Western blots
- DNA extraction (Mini/Midiprep), gel electrophoresis, and mammalian cell culture
- Fabricated multi-channel neural electrodes under stereomicroscopy (jig-guided, caliper-calibrated) for in-vivo murine recordings
Cell cultureCo-IPWestern blotElectrode fabrication - 02 Biomedical Engineering Intern · Istanbul May – Jun 2025 Internship
Turkish Ministry of Health
Field service and device prototyping across the national hospital network — calibrating clinical imaging systems and prototyping mammography hardware.
- Calibrated / serviced 30+ X-ray, CT, and angiography systems across 8 hospitals
- CAD + 3D-printed low-compression mammography prototypes across 5 design iterations
- Evaluated imaging-system proposals from Canon, Siemens, and Philips on cost and performance
Imaging hardwareCalibrationCAD3D printing - 03 Duke · BME 260L
CAR-T Tumor–Immune Modeling
“Cells Born to Kill” — an ODE model of CD19-directed CAR-T therapy in pediatric B-ALL, tracking malignant and normal B-cells, effector and memory CAR-T cells, and a lumped cytokine (CRS) compartment to link efficacy against toxicity.
- Logistic tumor growth + CAR-T-mediated killing; memory cells reactivate with residual tumor burden
- Parameter sweeps over cytotoxic potency, memory reactivation, and effector proliferation revealed remission-vs-toxicity thresholds
- Framed CAR-T dynamics as a PK/PD-like system to relate dose and timing to outcome
ODE modelingPythonPK/PDImmunology
03 / IMAGING / SIGNAL & FREQUENCY
Imaging & DSP
Signal and frequency-domain work across optical, MRI, and X-ray.
- 01 Duke · BME 303
OCT Simulator
A full frequency-domain OCT pipeline built from scratch — a Michelson-interferometer model, a simulated resolution phantom, spectral interferograms, and FFT depth reconstruction, with an analysis linking source bandwidth to axial resolution. Drag the bandwidth below to watch two reflectors resolve.
- Axial resolution follows Δz = (2 ln2 / π)·λ₀²/Δλ — wider bandwidth, sharper depth
PythonFD-OCTFourierDSPaxial resolution 7.5 µm FWHM
two reflectors 12 µm apart — resolved
- 02 Duke · BME 271
Brain Tumor Segmentation
“Tuning into Tumors” — a 2D-Fourier segmentation pipeline that treats each MRI slice as a signal, using high-pass and band-pass FFT filters (plus Canny edges) to isolate glioma boundaries, benchmarked against Otsu intensity thresholding.
- Evaluated across 50+ annotated scans (UCSF preoperative-glioma set) on Dice, IoU, and boundary accuracy
- Frequency-domain filters recovered tumor edges that intensity thresholding missed
- Three-person student team; I built the FFT filtering and evaluation pipeline
Python2D FFTMRIDice / IoUCanny - 03 Research & Technical Intern · Istanbul Jul 2025 Internship
Canon Medical Systems
Detector-physics and image-quality work for a clinical imaging vendor — benchmarking hardware, evaluating angiography suites, and briefing engineering teams.
- Benchmarked 12 flat-panel detectors (Siemens / GE / Philips / Samsung) on DQE, spatial and contrast resolution
- Analyzed charge-trapping in a-Si / CsI detectors and produced image-quality reports
- Evaluated clinical workflows across 5 biplane angiography suites (dose, protocols, interface)
- Supported government imaging-tender evaluations with detector benchmarking and competitive analysis
- Delivered a 45-minute X-ray physics seminar (bremsstrahlung, collimation, dose) to 20+ engineers
DQEDetector physicsX-rayAngiographyBenchmarking
04 / EMBEDDED / DEVICE DESIGN
Embedded & Device Design
Circuits, firmware, and enclosures — built and bench-validated.
- 01 Duke · BME 254L
Night King Lightbox
A full custom light-box, designed and fabricated end to end: KiCad schematic and PCB layout, interrupt-driven Arduino firmware with PWM dimming, and a 3D-printed enclosure — then bench-validated on the scope.
- Interrupt-driven state-machine firmware with PWM brightness control
- Custom PCB (KiCad schematic + layout) housed in a 3D-printed enclosure
- Oscilloscope-validated — R² ≈ 0.99 on brightness-vs-voltage
- Root-caused and re-spun a PCB routing fault found via scope probing
KiCadPCB designArduinoPWMOscilloscope - 02 Duke · Client Project (CapCell / Ramanujam Lab)
Automated Fluorescence Microscope
Maps six user-defined parameters to physical light and stage movements for mouse-tissue imaging.
EmbeddedMotion controlImaging - 03 Duke · Client Project
Eye-Drop Administration Device
Assistive device helping elderly users self-administer eye drops safely.
Assistive deviceCADPrototyping
05 / PREPRINTS / RESEARCH OUTPUT
Preprints / Research Output & Honors
- 01 Preprint · not peer-reviewed
PK/PD Model of Metoprolol for Doping Detection
Co-authored a 7-compartment PBPK model with CYP2D6 phenotype variability and Monte Carlo population simulation; the analysis found no exploitable undetectable window for evading detection. (BME 260.)
Read the preprint ↗ - 02 Scholarship
Defne – Muhtar Kent Education Foundation Scholarship
Merit scholarship.
- 03 Honor
BME Design Fellow · FOCUS Program
Selected biomedical design fellowship and first-year FOCUS interdisciplinary program.
06 / CAPABILITIES
Skills
- Embedded
- Embedded CZephyr RTOSnRF Connect SDKdevicetree / KconfigArduinoSPI / I²CKiCad PCB designOscilloscope validation
- Software / Data
- Python (signal processing, FFT, simulation)C / C++MATLABFlaskSQLiteGit / GitHubnginx / Linux server admin
- Imaging / DSP
- MRI segmentationOCT / FD-OCTCTUltrasoundECG / EEG / EMG acquisitionk-space analysis
- Wet-lab / Device
- Cell cultureDNA extractionWestern blotCo-IPNeural electrode fabricationStereomicroscopyCAD3D printingBiocompatible materials
EDUCATION
Education
Duke University
B.S. Biomedical Engineering
GPA 3.40 · Pratt School of Engineering
Aug 2023 – present