Research Portfolio · 2025-2026

Hadi Abdul

Building open-source tools at the intersection of biomedical engineering, digital health, and global health systems.

NCSSM '26Duke BIG IDEAs LabHarvard '30 Incoming

Reflection Essay

Background & Academic Interests

I am a senior at the North Carolina School of Science and Mathematics, and research now feels less like a distant category and more like a way of building for real people. I will enter Harvard University in the fall, and my academic interests sit at the intersection of biomedical engineering, digital health, and global health systems. I am drawn to work that can move from a bench, a model, or a circuit board to a person who can use it. My courses at NCSSM, including Applications in Entrepreneurship, Mathematical Modeling, Biophysics, and Organic Chemistry, have helped me connect technical depth with practical constraints. Maintaining a 4.92 GPA has mattered to me, but the lesson has been learning how fields translate into one another.

Research Environment

The Duke BIG IDEAs Lab, led by Dr. Jessilyn Dunn, gave me my clearest view of that translation. The lab brings together biomedical engineers and data scientists around open-source wearable tools, and the environment is collaborative in a specific way: people are generous with expertise, but they also expect work to be reproducible, useful, and honest about its limits. My work on VitalWave fit that culture because the project required hardware, firmware, mobile software, cloud storage, and dashboard design to behave as one system. I also worked with the WHO Innovation Hub on digital health innovations across the Eastern Mediterranean Region under Dr. Ahmedali, and with the Oxford Systems Biology Group under Dr. Bela N. on MemoryAssist. Being in a U.S. academic lab, a global health policy hub, and a translational systems biology group showed me how research culture changes when the pressure is validation, implementation, or clinical use.

Strengths & Weaknesses

One strength I bring to research is comfort moving across the hardware-software stack. I like the messy middle where a sensor, a phone, a database, and a visualization tool all have to agree. I am also willing to debug for longer than is convenient. At Katrick, a thermal-resistance bug took weeks to resolve, and the answer was not dramatic: a 0.3 mm interface was causing more than 40 percent of the thermal resistance. That experience changed how I think about precision. I have also learned to communicate technical work to audiences outside the project through TEDx, popular article writing, and explaining why raw sensor access matters.

My main weakness is that I over-scope. If a prototype can support one sensor, I start imagining six. If a dashboard shows one signal, I want cohort analytics, alerts, export tools, and publication-ready charts. Research has forced me to ask whether a feature serves the question or only satisfies my urge to build. I am also learning that a literature review is not a formality. Sometimes I treat building as the fastest way to understand a problem, but strong projects begin by knowing what has already been tried.

Favorite & Least Favorite Parts of Research

My favorite moment in research is when a signal comes through clean for the first time. In VitalWave, seeing raw physiological and motion data stream reliably made the system feel real in a way no schematic could. I also value mentors who are generous with time and precise in feedback. The best conversations help me see the next question I should have asked.

I also care about the moment when research reaches users. MemoryAssist became more meaningful during pilot deployments at Savoy Nursing Home because the device was no longer just an idea about navigation, face recognition, and caregiver alerts. It was something people could hold, react to, and challenge.

The parts I like least are less elegant but just as real: waiting on parts, moving through IRB paperwork, and confronting the distance between a working prototype and a robust one. A demo can succeed once. A research tool has to keep succeeding when conditions are imperfect, the battery is low, or the data format changes.

Future Endeavors

At Harvard, I plan to continue in biomedical engineering and digital health, with the possibility of an MD/PhD or a research-driven engineering path. Long term, I want to help build open-source health infrastructure for low-resource settings: systems that are technically rigorous, locally adaptable, and not locked behind proprietary tools. This year taught me that research is not only about producing an answer. It is about building enough trust in the method that other people can extend it, question it, and use it where it matters.

Video reflection — also available as written essay above.

VitalWave: An Open-Source End-to-End Wearable Platform for Digital Health

Duke University · BIG IDEAs Lab · Dr. Jessilyn Dunn

100 Hz

Raw multimodal sampling across physiological and motion channels.

0

Packets lost in streaming trials during initial system evaluation.

3+ days

Battery life under standard use for longitudinal collection.

Lauren Lederer and Hadi Abdul holding a VitalWave wearable device at Duke

Featured In · Duke Vertices

The BIG IDEAs Lab at Duke is shaping the future of medical wearable technology

The article highlights Dr. Jessilyn Dunn's lab, Lauren Lederer's work on Project VitalWave, and the importance of open-source wearable tools that can move from research prototypes into real clinical and field settings.

Read the Duke Vertices feature

System Architecture

Wearable Device

Multimodal physiological and motion sensing at research-grade frequency.

iOS App

Bluetooth streaming, session control, and offline logging for field use.

Cloud Storage

Secure synchronization for processing, validation, and longitudinal studies.

Web Dashboard

Visualization and export tools for digital biomarker analysis.

Sensor Field Guide

VitalWave collects raw physiological, motion, and environmental channels rather than only low-frequency summaries. That matters because digital biomarker research often depends on interpreting how these signals interact.

PPG

Photoplethysmography

An optical pulse signal that tracks blood-volume changes under the skin.

Captures
Pulse waveform, heart rate, and heart-rate variability features from reflected light.
Project role
VitalWave used PPG to estimate heart rate and compare physiologically relevant trends against a Polar H10 reference.

ACC

Accelerometer

A three-axis motion sensor that measures linear acceleration.

Captures
Walking, activity intensity, posture shifts, steps, and motion artifacts in the signal.
Project role
ACC contextualizes physiology: elevated heart rate during high movement means something different than elevated heart rate during low activity.

GYR

Gyroscope

A rotational motion sensor that measures angular velocity.

Captures
Device rotation, wrist turns, gesture dynamics, and movement direction changes.
Project role
GYR complements accelerometry so downstream models can separate true physiology from movement-driven noise.

MAG

Magnetometer

An orientation sensor that measures the local magnetic field.

Captures
Heading and device orientation relative to the surrounding environment.
Project role
MAG helps stabilize motion context when the wearable changes orientation during everyday use.

HUM

Humidity

An environmental channel that records local humidity around the device.

Captures
Ambient moisture and skin-adjacent humidity conditions during collection.
Project role
HUM adds context for field studies because moisture and device contact can affect signal quality and user comfort.

TMP

Temperature

A thermal channel for skin-adjacent and ambient temperature context.

Captures
Temperature changes around the sensor package and local environment.
Project role
TMP supports richer interpretation of wearable data by adding thermal context to cardiovascular and motion signals.

Research Abstract

VitalWave: An Open-Source End-to-End Wearable Platform for Digital Health

Hadi Abdul, North Carolina School of Science and Mathematics; Jessilyn Dunn, Duke University

204 words
01Introduction
02Methods
03Results
04Conclusions

Digital biomarkers derived from wearable devices have strong potential to support continuous, real-time health monitoring, but many commercial systems provide only low-frequency, preprocessed outputs and limited access to raw signals, restricting their utility for research and machine learning. To address this gap, we developed VitalWave, an end-to-end open-source wearable platform for high-frequency digital health data collection and analysis. VitalWave integrates a multimodal wearable device with an iOS mobile application, secure cloud storage, and a web-based dashboard for visualization and downstream analysis. The device captures raw physiological and motion data, including photoplethysmography, accelerometry, gyroscope, magnetometer, humidity, and temperature signals, at up to 100 Hz, while supporting real-time Bluetooth streaming and offline logging. Initial system evaluation using 10-minute walking trials assessed signal quality, data transmission, cloud synchronization, and dashboard functionality. VitalWave successfully streamed and stored high-quality multimodal data with no reported packet loss, produced physiologically relevant heart rate estimates compared with a Polar H10 reference, and supported automated cloud processing and web-based visualization. Battery life exceeded three days under standard usage. These findings demonstrate the feasibility of VitalWave as a low-cost, research-grade platform for digital biomarker development and wearable health studies. Future work will focus on broader validation, additional sensor integration, and expanded machine learning applications.

NCSSM Research Symposium — Slide Deck

Final oral presentation for the Comprehensive Research Portfolio.

Download PDF (2.2 MB)

Structure: Title · Introduction · Methods · Results · Discussion · Conclusions · Acknowledgements · References

Title slide for the VitalWave presentation

Title

Research problem slide for the VitalWave presentation

Problem

System architecture slide for the VitalWave presentation

System

Signals collected by VitalWave slide

Signals

End-to-end performance results slide for VitalWave

Results

Discussion slide for VitalWave

Discussion

Profile · VitalWave Research

A "Wave" of Vital Information: How an NCSSM Researcher's AI Wearable Enables Continuous Biometric Data Monitoring

Joshua Chilukuri

Written by Joshua Chilukuri

Interviewed researcher: Hadi Abdul

A profile of VitalWave, Hadi Abdul's open-source wearable platform for continuous biometric data monitoring and research-grade digital health data.

You just finished your morning run. 160 BPM heart rate, not bad at all. Glancing at your FitBit enables you to quickly track your vital data, and note how effective the workout was. That's real data - information that enables you to better understand your own body and health, serving you in improving your well-being. That said, imagine how important this data is for scientific researchers. Worldwide, active research across health domains faces a significant bottle-neck: there's limited access to open-source, accessible biometric data or information. Traditionally, collecting this data requires sophisticated, often IRB-approved, user studies, and even then, labs often keep such data private. What if there was a device and online platform that could enable biometric data to be easily collected, documented, and uploaded for use by scientific research communities? That's the idea NCSSM student researcher, Hadi Abdul, sought to chase after with Duke University's Big Ideas Lab under Professor Jessilyn Dunn.

At the start of his senior year, Hadi began his research term with NCSSM's mentorship program unsure of what to expect. He knew he was generally interested in bio-medical technology but didn't understand what exactly biomedical engineering research looked like. It wasn't until joining Dr. Jessilyn Dunn that he'd be exposed to a pertinent issue influencing modern BME research, especially given the modern relevance of data-intensive statistical models, such as AI, within the field.

"Data is challenging to access," Hadi told me. "Especially in developing countries. For example, I learned about researchers in Uganda whose efforts were being halted by the low access to biometric marker data sourced within the region." That inspired him to pursue an interesting challenge - what if Hadi could make his own wearable device, a watch similar to a FitBit or Apple Watch, that could source biometric data, such as heart rate, temperature, and orientation, for both user study use and public use, to build biometric datasets. Through NCSSM's mentorship program, he had the opportunity to be one of the only high school students in the Big Ideas Lab pursuing the exciting project as a culminating endeavor of his senior year.

It wasn't easy by any means at the start for Hadi. Having had limited experience on the engineering side, he first had to understand how non-invasive health sensors function and can be integrated into a device.

"A huge win for us is that we were able to source a chip with access to all 6 integrated sensors I wanted, simplifying the project design," Hadi told me. His device was effectively designed to collect signals PPG, ACC, GYR, MAG, HUM, and TMP, at a 100 Hz sampling rate. The signals represent the collection of heart rate, motion activity, rotational movement, orientation sensing, humidity, and skin temperature respectively. Throughout the first semester, Hadi learned the device design process to manufacture his first example watch.

Simultaneously, he also gained deep exposure to the research ecosystem at Duke, a skill which he noted will be super valuable for the rest of his life. Hadi was able to meet an assortment of researchers in his lab focused on a diverse set of problems within his general field of interest. He'd often gain help from other student researchers, including PhD students, while working on his own device. An exciting side project he also participated in was writing an article for Duke Vertices, where he wrote about the Big Ideas Lab and their work in medical device research. That experience taught him that scientific research isn't just about the technical side, but also how to communicate scientific ideas.

In the second semester, Hadi got to see his work result in tangible impact and results. With a partner in the lab, he was able to test his biomedical device and track real biometric data effectively. He tracked heart rate versus physical activity, leveraging the PPG and ACC sensors he integrated into the watch. The accuracy of his tests validated that his watch could serve as a real research-grade device. Now, Hadi is continuing work attempting to program a digital platform for the Big Ideas Lab to upload their biometric data for public access. He hopes the platform will serve as a powerful resource for groups, such as the researchers in Uganda he learned of, who could use the data to conduct their own research on health conditions.

It's hard to distill Hadi's journey into a single lesson. His work demonstrates that research can be an unexpected experience that stretches and challenges high school students in more ways than one. Yet, beyond that, his work also directly reflects the importance of designing systems that make data safe to procure, and publicly available, to advance spaces as sensitive as medical research. Moving forward, Hadi hopes to study biomedical engineering, or a related field, in college, where he is excited to continue pursuing research.

Resume

Education

Harvard University, Cambridge, MA

Incoming First-Year Student

    North Carolina School of Science and Mathematics, Durham, NC

    Aug 2024 – May 2026

    • Courses: Applications in Entrepreneurship, Mathematical Modeling, Biophysics, Organic Chemistry
    • GPA: 4.92

    Relevant Experience

    World Health Organization Innovation Hub — Eastern Mediterranean Region

    Research Intern · Supervisor: Dr. Ahmedali M.

    Sept 2025 – Present

    • Currently mapping 80+ COVID-era digital health innovations across the Eastern Mediterranean, assessing scalability barriers including infrastructure, interoperability, and digital literacy gaps.
    • Synthesizing input from 40+ regional experts to produce ranked research agendas across digital health, pandemic preparedness, and health systems strengthening.

    Duke University — Big Ideas Lab (Dr. Jessilyn D., Dept. BME)

    Research Intern · PI: Dr. Jessilyn D.

    Mar 2025 – Present

    • Developed VitalWave, an open-source wearable biosensing platform for digital biomarker discovery using high-frequency physiological data acquisition.
    • Integrated near-infrared spectroscopy (NIRS) sensors for real-time biochemical composition sensing.
    • Collaborated with Duke BME researchers on wearable health-data analysis pipelines.

    University of Oxford — Systems Biology Group (Dr. Béla N.)

    Co-Founder, MemoryAssist · PI: Dr. Béla N. — U.S. Provisional Patent Application No. 63/755,185

    Jun 2025 – Dec 2025

    • Co-Founded MemoryAssist, an AI-powered wearable to support Alzheimer's and dementia patients with real-time navigation, face recognition, and caregiver alerts.
    • Piloted two units at Savoy Nursing Home, MA; recognized Congressional App Challenge Winner (CD 12, 2024).

    Katrick, Scotland — Passive-Cooling Technology with Clinical Application

    Developer · Supervisor: Mr. Farshad Q.

    Jun 2023 – Jan 2025

    • Assisted Katrick's vibration-powered passive cooling module: fixed a 12% thermal-limit overshoot by identifying a 0.3 mm interface causing >40% of thermal resistance.
    • Contributed to Katrick + iomart's award-winning live deployment at Glasgow data centre, ~70% reduction in energy, ~100 tonnes of carbon emissions avoided — Digital City Awards: "Best Use of Emerging Technology."

    Lucent — PFAS Water Filtration System

    Co-Founder & Engineer

    Sept 2024 – Mar 2025

    • Founded 3D-printable PFAS water filtration system with 8 filters deployed delivering ~600,000+ liters of clean water annually across partner sites including Daffodils Ladies Hostel in Koche, India and Africa Safe Water Foundation in Kenya — advanced to Diamond Challenge Grand Finals (Top 60 / 3,400 innovations).

    Other Achievements & Activities

    • 2026 Coca-Cola Scholar — Recognized among 150 of 100,000+ applicants nationwide for $20,000 scholarship (Feb 2026)
    • USA Mathematical Olympiad (USAMO) Qualifier — Top 250 of 300,000 (Jan 2026)
    • U.S. Physics Olympiad (USAPhO) Silver Medal — Top 90 of 6,000 (April 2025)
    • Diamond Challenge Grand Finalist — Top 60 of 3,400 teams (Mar 2025)
    • TEDx Speaker — "Breaking Through Perceived Limits," 8.6k YouTube views (Oct 2024)

    Recognition & Relevant Documents

    Selected honors that contextualize the research portfolio and broader academic work.

    2026

    Coca-Cola Scholar

    150 / 100,000+ · $20,000

    National scholarship recognizing leadership, service, and academic achievement.

    2026

    USAMO Qualifier

    Top 250 / 300,000

    Qualified for the USA Mathematical Olympiad through national competition.

    2025

    USAPhO Silver Medal

    Top 90 / 6,000

    Recognized among top U.S. Physics Olympiad competitors.

    2025

    Diamond Challenge Grand Finalist

    Top 60 / 3,400

    Advanced Lucent PFAS filtration system to global finals.

    2024

    TEDx Speaker

    8.6k views

    "Breaking Through Perceived Limits" talk on achievement and self-expectation.