Projects | Anh Duong Vo

Applying new data analysis or ML methods to analyze multi-modal time series data

Mon, 01 Jan 0001 00:00:00 +0000

Work:

Using parallelization and GPU/CPU to improve data analysis pipelines
Denoising neural data and extracting important features
Dealing with small datasets
Classifying behavior in images

Application in industry:

Healthcare: Analyzing condition of motion and sensory processing of each patient
Robotics: Develop robots with motion inspired by real humans
Assistive Devices: tackling only motion dimension of a human and built assistive devices based on that
Automated data labelling to train models for smart wearables
Various Data analysis cases where limited time series data is a problem

LLMs, Generative models and smart wearables

Mon, 01 Jan 0001 00:00:00 +0000

Predicting eye movement with EEG/EOG data

Work:

multimodal system to predict eye movement and focus
supervised Master students with following code:

Applications in industry:

Automotive
Wearables

Generating images and text based on EEG data

Work:

Using diffusion models to generate images and text based on neural data
supervised Master student with following code:

Applications in industry:

Consumer Behavior
Healthcare: Information about how users process environment
Wearables and app development

Limitations of Large Language models

Work:

Collaborating with researchers all over the world to create a study on limitations of Large Language models for different languages across the world
Publication: Noga Mudrik, …, Anh Duong Vo, Others (2025). Lost in Translation? LLMs, Education, and Linguistic Fairness. IEEE ISEC 2025.

Applications in industry:

Applying LLMs for different languages

Python coding course for students with disability in Vietnam

Mon, 01 Jan 0001 00:00:00 +0000

In 2024, I launched a Python and Data Science course in collaboration with industry partners, providing adults with disabilities in Vietnam access to higher education in technology. Designing and delivering these lectures reinforced my ability to make complex tools accessible.

Understanding more about the brain and developing models

Mon, 01 Jan 0001 00:00:00 +0000

Work:

Movement prediction
Transfer learning and generalization
Learning on a synaptic scale

Applications in industry:

Wearables: user intent such as clicking on a button
Automotive
Assistive devices
Healthcare: Using wearables to measure emotions in prefrontal cortex (front head) for therapeutic purposes.
Advertising: Measure the emotional effect of advertisement.