Undergraduate Summer Research Opportunities

Primary Investigators:
Yayun Du
 
Brief Description of Project:
Human–robot physical interaction is significantly influenced by a user’s internal physiological state, including stress, pain, arousal, and emotional tension. SYMBIO-X Lab is developing multimodal models that infer these states in real time using physiological signals (EMG, IMU, heart rate and respiration rate proxies, etc.). While current efforts focus on sensing and machine learning, the next critical step is designing adaptive robot behaviors that use these predictions to enhance safety, comfort, and performance.

In this 10-week project, the student will create and evaluate simulation-based control strategies that adjust a robot’s compliance, speed, and motion trajectories based on predicted human state. Working with synthetic or recorded human-state predictions from ongoing lab work, the student will implement adaptive controllers in PyBullet, Isaac Gym, or Gazebo, and test how robot behavior changes when a user is inferred to be stressed, pained, or startled.

This project does not involve human subjects and requires no IRB. All robot-human interactions are conducted in simulation to ensure safety and rapid iteration. Results will feed directly into a forthcoming IEEE RAL / T-RO manuscript.

Desired Qualifications:
Majors: ECE, CS, ME, Robotics, or related fields
Some experience in Python (required), ROS and ROS2
Familiarity with PyTorch, reinforcement learning, or robot simulation (helpful but not required)
Interest in HRI, robotics control, ML, or physiological sensing

Nature of Supervision:
Primary day-to-day supervision by a senior master’s student working on physiological state modeling
Weekly meetings with Dr. Yayun Du (PI, SYMBIO-X Lab) for scientific guidance and integration with ongoing projects
Access to lab group meetings, code frameworks, and simulation environments
No dependency on in-lab robot access; project is entirely simulation-based

A Brief Research Plan (period is for 10 weeks):

Weeks 1–2:

• Orientation to SYMBIO-X Lab

• Review HRI control literature and current lab physiological-state models

• Set up simulation environment (PyBullet, Isaac Gym, or Gazebo)

Weeks 3–5:

• Implement baseline robot behaviors (standard impedance/compliance control)

• Integrate synthetic or recorded physiological-state predictions (stress/pain/arousal)

• Begin basic adaptive-behavior prototypes (speed modulation, compliance adjustment)

Weeks 6–8:

• Develop advanced adaptive controllers (trajectory shaping, force-limiting, RL-based policy shaping)

• Perform simulation experiments comparing baseline vs. adaptive robot behavior

• Analyze performance, comfort, and safety metrics

Weeks 9–10:

• Finalize adaptive-behavior module

• Generate figures and simulation results suitable for publication

• Prepare a final poster and written summary for the VUSE program

Number of Open Slots: 1 - 2

Contact Information:
Dr. Yayun Du
yayun.du@vanderbilt.edu
Electrical and Computer Engineering (VUSE); VISE affiliate
https://duyayun.github.io/

Primary Investigators:
Shekhar Bhansali
 
Brief Description of Project:
This project develops a fundamental microfluidic droplet generator and using computer vision on an NVIDIA Jetson to monitor and control droplet formation in real time. A single microchannel with a T-junction directs liquid into a main channel while air is injected through a lateral branch, producing liquid slugs or droplets interspersed with air pockets. A camera aimed at the channel will relay footage to the Jetson, where OpenCV algorithms will independently extract droplet parameters like as size, velocity, and frequency. The student will generate a dataset linking operational conditions to droplet characteristics by systematically varying the liquid and air flow rates. This dataset will be employed to train a fundamental predictive model that suggests flow parameters to achieve a designated droplet size or frequency, so completing the feedback loop between sensing and actuation.  

The project combines microfluidics, embedded artificial intelligence, and control mechanisms, so laying groundwork for autonomous lab-on-chip systems. 

Desired Qualifications:
Interest in microfluidics and/or embedded AI, Basic Python programming; some familiarity with OpenCV or willingness to learn, Comfortable with hands-on lab work (tubing, pumps, mounting hardware), (Optional) Experience with Arduino or other microcontrollers
Academic requirement 3.0+ GPA 

Nature of Supervision:
The student will be mentored by Dr. Shekhar Bhansali (Vanderbilt ECE) with assistance of research scientist/postdoc expert in soft material and robotics and haptic interfaces. Weekly meetings and laboratory demonstrations will focus on design iteration, calibration, and physical testing.

A Brief Research Plan (period is for 10 weeks):
Week 1–2: Design & Printing Setup, build and test the T-junction droplet generator; get stable droplet trains with dyed water + air
Week 3–4: Baseline Mechanical and Permeability Testing, set up camera + Jetson; acquire and store video of droplets under different flow conditions. 
Week 5-6: Degradation Study Setup, implement image processing to automatically measure droplet size, speed, and frequency; validate on multiple videos.
Week 7-8: Continued Degradation & Data Collection , train a simple model relating (Q_liquid, Q_air) → (size, frequency); implement a basic control script that chooses flow settings for a target droplet size.
Week 9-10: Exhibition and evaluation

Number of Open Slots: 2 - 5

Contact Information:
Shekhar Bhansali
shekhar.bhansali@vanderbilt.edu
Electrical and Computer Engineering

Primary Investigator:
James Trippe


Brief Description of Project:
The Institute for Space and Defense Electronics (ISDE) is a key member of the Center for the Advancing the Radiation Resilience of Electronics (CARRE) project. CARRE’s aim is to develop a next-generation radiation transport code capable of handling advanced electronics scaled to the current limit of Moore’s Law. This undergraduate research project will focus on performing radiation transport simulations using both existing software and the early versions of the new code, as available, to model radiation experimental data on these cutting-edge electronics. The student will have the opportunity to interface with faculty and graduate student mentors involved in radiation effects, participate in radiation experiments needed to support the simulations performed, and involve themselves with broader CARRE activities in the larger collaboration.


Desired Qualifications:
Undergraduate in a STEM field. U.S. citizen. Familiarity with Python or another high-level programming language.


Nature of Supervision:
Professor Trippe will be the primary supervisor of the student. Additional mentors will include other ISDE faulty, graduate students, and members of the CARRE collaboration.


A Brief Research Plan (period is for 10 weeks):
The student will immediately begin participating in CARRE activities, including regular meetings and testing/simulating with the assistance of mentors. The focus for the first three weeks will be developing familiarity with the relevant radiation transport codes and planning a relevant simulation campaign based on existing data. Then the student will spend the next six weeks executing the plan, with the last week will be spent preparing a report and presentation. A critical part of the report will be recommendations for improvements to the useability and accuracy of the CARRE developed code.


Number of Slots: 1

Contact Information
James Trippe
Electrical and Computer Engineering
james.m.trippe@vanderbilt.edu
https://www.vanderbilt.edu/isde/
https://carre-psaapiv.org/

Primary Investigator:
Sharon Weiss

Brief Description of Project:
While silicon is typically considered as the predominant material for microelectronics technology, such as computers and cellphones, it can also serve as a medium for the transfer of information via light signals. The use of silicon for light-based technologies, such as the encoding, transfer, and detection of information on light waves, enables increases in operating bandwidth and reductions in power consumption. Today, on-chip silicon photonic devices are poised to transform capabilities across multiple sectors, including data center transceivers, optical interconnects, LiDAR, and immunoassays. This project will explore new opportunities to enhance the performance of silicon photonic devices by introducing subwavelength engineered features while at the same time enabling new capabilities that are not achievable with traditional photonic components. The project will involve a combination of simulations and measurements using techniques that are well-established in the Weiss group.

Desired Qualifications:
Highly motivated student with at least basic physics knowledge. 

Nature of Supervision:
The student will work closely with graduate students in the Weiss lab and have regular meetings with Dr. Weiss to provide additional guidance and feedback.

A Brief Research Plan (period is for 10 weeks):
Lab safety training and learning necessary background for project
Training on experimental and computational methods necessary for the project
Simulations, measurements, and analysis
Written summary of results and poster preparation

Number of Slots: 1

Contact Information
Sharon Weiss
Electrical & Computer Engineering
Vanderbilt Institute of Nanoscale Science and Engineering
sharon.weiss@vanderbilt.edu
https://my.vanderbilt.edu/vuphotonics/

Primary Investigator:
Bennett Landman

Brief Description of Project:
This project will focus on large-scale, cross-modality segmentation of medical images using nnU-Net, a self-configuring deep learning framework for biomedical image segmentation. The student will work on applying and extending nnU-Net to segment the optic nerve and brain across multiple imaging modalities (e.g., MRI, CT). This will involve dataset organization, preprocessing, model training, and quantitative evaluation across diverse cohorts. The ultimate goal is to produce generalizable, open, and well-documented segmentation tools that can be integrated into MASI Lab’s broader neuroimaging research workflows.

Desired Qualifications:
Required: Strong experience with Linux and Python
Preferred: Familiarity with PyTorch or deep learning frameworks
Helpful: Background in medical imaging, artificial intelligence, or computational modeling

Nature of Supervision:
Primary mentor: Bennett A. Landman (weekly 1:1 meetings)
Graduate mentor: A MASI Lab Ph.D. student will provide day-to-day guidance and code review
Students will participate in MASI’s weekly group meetings for progress updates, technical discussions, and collaborative learning

A Brief Research Plan (period is for 10 weeks):
Weeks 1–2: Setup environment, explore nnU-Net, and select relevant datasets
Weeks 3–4: Data preprocessing and baseline model training
Weeks 5–7: Model tuning, cross-modality experiments, and performance benchmarking
Week 8: Error analysis and segmentation quality assessment
Weeks 9–10: Documentation, packaging, and presentation of results

Number of Slots: 1 - 3

Contact Information
Bennett Landman
Electrical & Computer Engineering
bennett.landman@vanderbilt.edu
https://my.vanderbilt.edu/masi/

Primary Investigators:
Jason Valentine
 
Brief Description of Project:
Meta-optics are artificially structured materials wherein the structure of the unit cells dictate the optical response allowing for unique optical properties, not existing in nature, to be achieved. Recently, we have shown how meta-optics can be used in concert with digital ML architectures to achieve optical image processing for spatial and spectral classification. The use of meta-optics allows one to off-load computationally expensive operations into ultrafast optical operations for large increases in processing speed while also decreasing energy consumption.  The purpose of this summer research experience is to build on this work, investigating co-design of meta-optics and machine learning (ML) algorithms for tasks such as object segmentation and spectral classification. Hybrid meta-optics and digital designs will be benchmarked against traditional techniques to determine the efficacy of the approach.

Desired Qualifications:
This project is best suited for individuals with interests in optics and machine learning. Experience with numerical modeling is also beneficial.  No experience is needed in meta-optic design.

Nature of Supervision:
The student will be co-advised by Prof. Valentine and a graduate student on this project.

A Brief Research Plan (period is for 10 weeks):
Weeks 1–3: The student will familiarize themselves with hybrid meta-optic and ML techniques.
Weeks 3-8: The student will assist a graduate student in the design and fabrication of hybrid meta-optic systems.
Weeks 8-10: The student will assist in experimental validation of the hybrid meta-optic system.

Number of Open Slots: 1

Contact Information:
Jason Valentine
Professor
Mechanical Engineering
332 Olin Hall
615-875-5508
jason.g.valentine@vanderbilt.edu
https://valentineoptics.github.io/LabWebsite/index.html

Primary Investigators:
Justus Ndukaife
 
Brief Description of Project:
This project will expose students to the trapping and manipulation of nanoscale objects using light and electric field within microfluidic channels. The students will gain exposure on experiments on nano-optical trapping with fabricated structures and data analysis.

Desired Qualifications:
Undergraduate student with basic knowledge of Programming, General Physics, and Chemistry.

Nature of Supervision:
The student will take part in weekly meetings with the PI and also work closely with graduate students in the PI’s lab.

A Brief Research Plan (period is for 10 weeks):
Week 1-3: Literatures review
Week 2-4: Training
Week 4-8: Experiments
Week 7-10 Data analysis and writing paper

Number of Open Slots: 1

Contact Information:
Justus Ndukaife
justus.ndukaife@vanderbilt.edu
Electrical and Computer Engineering
https://my.vanderbilt.edu/ndukaifelab/ 

Primary Investigators:
Shekhar Bhansali
 
Brief Description of Project:
This project will investigate how the interplay between material composition and scaffold architecture controls mechanical support and self-absorption in bioresorbable 3D-printed scaffolds. Using Fused Deposition Modeling (FDM) printed, the student will fabricate porous scaffolds either from PLA or PCL and a 50% each PCL+PLA hybrid with the same external dimensions but systematically varied porosity (ranging from 20%- 80%) and unit-cell geometry, and infill volume (e.g., gyroid vs simple rectilinear grid, 20-80% infill). These architected scaffolds will then be evaluated to understand how: Mechanical integrity (compressive modulus and strength), Permeability (fluid transport through the scaffold), and Early-stage degradation (mass loss and residual mechanical strength over time) are coupled during the initial 4–6 weeks of in PBS at 37 °C, mimicking early post-implant conditions. 

The outcome will be a set of design rules mapping porosity + geometry affecting mechanical support + degradation profile, directly relevant to self-absorbing soft scaffolds materials. 

Desired Qualifications:
Interest or background in materials science, mechanical engineering, biomedical engineering, or additive manufacturing. Basic experience with CAD modeling (SolidWorks, Fusion 360, or Blender) is helpful. Comfort with lab work and mechanical testing (tensile/compression machines, sample preparation) 

Academic requirement 3.0+ GPA 

Nature of Supervision:
The student will be mentored by Dr. Shekhar Bhansali (Vanderbilt ECE) with assistance of research scientist/postdoc expert in soft material and robotics and haptic interfaces. Weekly meetings and laboratory demonstrations will focus on design iteration, calibration, and physical testing.

A Brief Research Plan (period is for 10 weeks):
Week 1–2: Design & Printing Setup, create CAD models of unit cells (gyroid + simple grid) and derive scaffold designs with fixed external dimensions. Slice and print initial scaffold sets at different porosities (e.g., 40%, 60%, 80%). 
Week 3–4: Baseline Mechanical and Permeability Testing, conduct compressive tests (modulus, yield strength) on dry scaffolds for all architectures and porosities.
Week 5-6: Degradation Study Setup, controlled degradation in PBS (2-3 weeks)
Week 7-8: Continued Degradation & Data Collection, measure mass loss, perform compressive testing, Capture microscopy images to observe surface erosion and pore changes, material leaching assessment
Week 9-10: Exhibition and evaluation 

Number of Open Slots: 2 

Contact Information:
Shekhar Bhansali
shekhar.bhansali@vanderbilt.edu
Electrical and Computer Engineering

Primary Investigators:
Michael Alles
VU Institute for Space and Defense Electronics (ISDE)
 
Brief Description of Project:
ISDE leads a government-funded workforce development program in the area of Radiation-Hardened Microelectronics. The goal of this summer project will be to introduce the student to the technical area, or to build on and advance their existing experience. Specific technical details of the research project will be tailored to the student's experience and direction of interest. The student will be paired with one or more mentors for technical guidance and support. The student will participate in weekly SCALE programming during the 10 week period.

Desired Qualifications:
Undergrad in STEM field with a preference for EE.
US Citizen (due to funding rules)

Nature of Supervision:
Professor Alles will serve as the primary supervisor. The student will be paired with one or more additional mentors, which may include other students/grad students, faculty, and staff, depending on the specific project.
 
A Brief Research Plan (period is for 10 weeks):
The student will interact with the SCALE student cohort and participate in weekly programming (multi-university, hybrid-virtual). A project suitable for the student's specific experience and interests will be defined and one or more mentors assigned for technical support. The workforce development mission allows for wide technical latitude for the specific projects, allowing us to define a project that has synergy with our existing funded research programs.

Number of Open Slots: 1

Contact Information:
Michael Alles
Electrical and Computer Engineering
ISDE
mike.alles@vanderbilt.edu

Primary Investigators:
Brian Sierawski
 
Brief Description of Project:
Student will support models and software tools for microelectronics radiation hardness assurance. Electronics used in space systems must be robust in order to reliably operate in the natural space environment. In particular, ionizing radiation degrades and disrupts microelectronics. Commercial electronics are especially at risk but are desirable for their size, power, and cost reductions. Assessments of on-orbit failures are made from ground-based tests and simulations. This project will support the development of methods for radiation effects assessments. The thrust of this summer project is to involve the student in graduate research and introduce the student to radiation hardness assurance activities.

Desired Qualifications:
ECE students with sophomore or junior standing. Basic electrical circuits knowledge and Python programming experience.

Nature of Supervision:
Regular weekly meetings with Dr. Sierawski including participation in technical research meetings with faculty and graduate students.
 
A Brief Research Plan (period is for 10 weeks):
Learn fundamentals of space radiation and radiation effects on electronics
Become familiar with modeling and simulation tools used in radiation-effects research
Assist with data analysis, simulation, or development of research tools
Document results and prepare poster

Number of Open Slots: 1

Contact Information:
Brian Sierawski
Research Assistant Professor
Electrical and Computer Engineering
brian.sierawski@vanderbilt.edu

Primary Investigators:
Bharat Bhuva

Brief Description of Project:
Students will be helping with reliability assessment of advanced silicon technologies from commercial foundries. Reliability topics may include aging, catch-up, and radiation effects.  Students will help design test setup as necessary and make measurements.

Desired Qualification:
Junior standing with solid background in circuits. E experience with FPGA is a plus. 

Nature of Supervision:
Students will interact with faculty and graduate students on a weekly basis.   

A Brief Research Plan (period is for 10 weeks):
Learning about reliability issues, learning about measurements, learning test setup and test IC design, understanding FPGA firmware and making changes as needed, making measurements, analyzing data

Number of Slots: 2

Contact Information:
Bharat Bhuva
Professor
(615) 343-3184
Bharat.bhuva@vanderbilt.edu
Please see papers published in IEEE Transactions on Nuclear Science and Proceedings of International Reliability Physics Symposium on the IEEE website. 

Primary Investigators:
Shekhar Bhansali
 
Brief Description of Project:
This project aims to design, fabricate and test a soft robotic hand (fingers) integrated with embedded multimodal sensors to assess force, temperature, and texture (based on force feedback) for intelligent object detection and manipulation. The student will incorporate piezoresistive, capacitive, and thermosensitive components into flexible 3D-printed polymers (PDMS) that mimic the tactile sensitivity of human skin in the soft fingers. 

The study will focus on identification of material properties (e.g., soft/hard, metal, plastic, or cloth) and then optimize the grasping techniques informed by sensor feedback. The tactile data will serve as the basis for context-aware manipulation systems and will subsequently be integrated with AI models operating on Jetson-based devices. Students will acquire practical experience in the manufacture of soft actuators, calibration of sensors, and real-time data logging with embedded microcontrollers.

Desired Qualifications:
Experience or interest in electrical engineering or mechanical, robotics, soft materials. 
Proficiency in Arduino or Raspberry Pi, 3D printing, and fundamental circuit design 
Academic requirement 3.0+ GPA  

Nature of Supervision:
The student will be mentored by Dr. Shekhar Bhansali (Vanderbilt ECE) with assistance of research scientist/postdoc expert in soft material and robotics and haptic interfaces. Weekly meetings and laboratory demonstrations will focus on design iteration, calibration, and physical testing.

A Brief Research Plan (period is for 10 weeks):

Week 1–2: CAD and Soft Actuator Design , generate finger geometries and model deformation via COMSOL/Blender.
Week 3–4: Fabrication, fabricate prototypes using 3D printing, incorporating touch and temperature sensors within silicone or TPU
Week 5-6: Integration and Testing, integrate with Arduino, calibrate sensor outputs, and get material/texture data
Week 7-8: Data Analysis, assess sensitivity, repeatability, and accuracy of material classification
Week 9-10: Exhibition and evaluation 

Number of Open Slots: 2 - 5

Contact Information:
Shekhar Bhansali
shekhar.bhansali@vanderbilt.edu
Electrical and Computer Engineering

Primary Investigators:
Mona Ebrish
 
Brief Description of Project:
Gallium Nitride (GaN) devices integrated with diamond substrates offer unmatched potential for high-power, high-frequency electronics due to diamond’s exceptional thermal conductivity. However, managing and characterizing heat flow across the GaN–diamond interface remains a key challenge for reliable device operation.

In this project, the student will work with graduate students to investigate thermal management in GaN-on-diamond devices. The effort will include the setup, alignment, and calibration of a state-of-the-art thermal imaging camera for measuring device heating profiles during operation. Calibration will involve working with vendor-provided software and building scripts in Python or LabVIEW to automate data acquisition and improve spatial/temporal resolution. The objective is to generate reliable thermal maps of operating transistors, enabling correlation between device structure, interface quality, and heat dissipation pathways.

Depending on interest and skills, the VUSE student may focus on:

• Experimental setup and calibrationof the thermal imaging system.

• Software integration and automationof measurement workflows.

• Data analysis, including extraction of thermal resistance and hot-spot characterization.

This project provides hands-on experience with both semiconductor device physics and modern instrumentation, while also contributing to the broader effort of improving thermal design in wide-bandgap electronics. All steps are novel and may lead to contributions in peer-reviewed journals or conference proceedings.

Desired Qualifications:
This project is ideal for students interested in thermal management, microelectronics, and device characterization. Students should have a basic understanding of semiconductor physics (e.g., diodes, transistors) and some familiarity with programming in Python or LabVIEW. Previous exposure to cleanroom environments, microelectronics fabrication, or device testing is helpful but not required.

Nature of Supervision:
Students will be working with graduate students and the PI.
 
A Brief Research Plan (period is for 10 weeks):
In the first couple of weeks we will focus on training and getting the student familiarized with the equipment in the lab.  
Between the 3rd and 7th week we will conduct several experiments. 
At the last couple of weeks we will work on summarizing the results and preparing a presentation to the group with our finding and I will gouge if the findings are worthy of peer reviewed publication.

Number of Open Slots: 1

Contact Information:
Mona Ebrish
Electrical and Computer Engineering
mona.ebrish@vanderbilt.edu

Primary Investigator:
Bennett Landman

Brief Description of Project:
Students will use VolView (https://volview.kitware.com/) to visualize and interpret AI-generated predictions on lung CT scans, combining Vanderbilt’s research in low-dose CT and nodule detection with interactive 3-D exploration. The goal is to create an educational prototype that demonstrates how AI insights can be integrated into intuitive surgical and clinical visualization platforms.

This lab is part of VISE (Vanderbilt Institute for Surgery and Engineering). Students accepted into this project will have the opportunity to participate in VISE-specific activities in addition to other Summer Research Program activities and events.

Desired Qualifications:
Curiosity about medical AI, coding experience in Python or JavaScript, and enthusiasm for visualization or user design.


Nature of Supervision:
Joint mentoring by AI Scholars and Bennett Landman, with structured milestones and feedback sessions.


A Brief Research Plan (period is for 10 weeks):
Weeks 1-4 – explore VolView and lung AI models
Weeks 5-8 – integrate AI outputs and refine visualization
Weeks 9-10 – final presentation and documentation

Number of Slots: 1 

Contact Information
Bennett Landman
Electrical & Computer Engineering
VALIANT
bennett.landman@vanderbilt.edu
https://my.vanderbilt.edu/masi/

Primary Investigators:
Catie Chang
 
Brief Description of Project:
Functional magnetic resonance imaging (fMRI) is a widely used technology for studying human brain activity in health and disease. Yet, fMRI provides an indirect measure of neural activity, and fMRI signals also reflect blood-flow changes associated with physiological processes such as breathing and cardiac activity. This project uses signal processing methods to characterize specific neural and physiological contributions to fMRI data, and to relate these variables to cognitive and clinical measures.

This lab is part of VISE (Vanderbilt Institute for Surgery and Engineering). Students accepted into this project will have the opportunity to participate in VISE-specific activities in addition to other Summer Research Program activities and events.

Desired Qualifications:
Programming experience (preferably MATLAB or Python), and a strong interest in brain imaging data.

Nature of Supervision: 
The student will work closely with the PI, graduate students, and postdocs in the lab.

A Brief Research Plan (period is for 10 weeks):
The first couple of weeks will involve becoming familiar with fMRI and learning relevant data analysis techniques. The remainder of the program will focus on analyzing neuroimaging and physiological signals, and preparing a final report and poster.

Number of Open Slots: 1

Contact Information:
Catie Chang
catie.chang@vanderbilt.edu
Electrical and Computer Engineering

Primary Investigators:
Yayun Du
 
Brief Description of Project:
Falls are a major cause of injury in older adults, yet most fall-detection systems are reactive and ignore the rich neural information associated with balance threat. Recent studies demonstrate that EEG features—particularly those measurable from ear-EEG—contain identifiable signatures of postural disturbance, startle, and fear-of-falling states. This 10-week project focuses on developing a signal processing and machine-learning pipeline to detect early balance-threat events using existing EEG + IMU datasets, and on designing a feasibility prototype of an earable EEG–IMU device and a galvanic vestibular stimulator.
The student will analyze publicly available balance-perturbation datasets, extract spectral/complexity features, and build EEG/IMU sensor-fusion models. They will also design a concept earable form factor (3D shell + electronics layout) informed by state-of-the-art ear-EEG and vestibular literature.
The long-term vision of this research is a closed-loop system where ML-detected balance threats could trigger safe galvanic vestibular stimulation (GVS) or vibrotactile cues to improve stability. During the summer project, the student will only explore this closed-loop GVS conceptually (e.g., simulation of timing and trigger logic). No human studies or stimulation experiments are performed, and no IRB approval is required.

This lab is part of VISE (Vanderbilt Institute for Surgery and Engineering). Students accepted into this project will participate in VISE-specific activities in addition to other Summer Research Program activities and events.  

Desired Qualifications:
Majors: ECE, BME, CS, Neuroscience, or related fields
Experience with Python or MATLAB
Interest in neural engineering, signal processing, wearables, or rehab robotics
No EEG or hardware experience required

Nature of Supervision:
The student will be supervised by Dr. Yayun Du (ECE, SYMBIO-X Lab) with day-to-day mentorship from a senior graduate student. Weekly PI meetings, regular check-ins, and integration into lab meetings ensure close guidance.

A Brief Research Plan (period is for 10 weeks):

Weeks 1–2: Literature review (ear-EEG, balance-threat EEG, GVS), dataset onboarding, goals
Weeks 3–5: Preprocessing of EEG/IMU datasets, feature extraction, exploratory analysis
Weeks 6–8: ML model development, EEG+IMU fusion, explainability analysis
Week 9: Design of wearable form factor + electronics concept
Week 10: Conceptual closed-loop GVS trigger model; final poster + report

Number of Open Slots: 1

Contact Information:
Dr. Yayun Du
yayun.du@vanderbilt.edu
Electrical and Computer Engineering (VUSE); VISE affiliate
https://duyayun.github.io/