Tien Truong

I am interested in

Technical skills

- HTML
- CSS
- JavaScript
- Python
- R
- Boostrap
- Numpy
- Pandas
- Matplotlib
- Seaborn
- Scikit learn
- Tensorflow
- Keras

Let's Chat!

Email: tientruong@usf.edu

Designed and Coded by Tien Truong

Hello! I am a PhD student in Industrial Engineering at the University of South Florida, specializing in AI-driven computer vision systems, multimodal data processing, and machine learning applications. With a background in Economics and Cognitive Science from UC Berkeley, I focus on developing trustworthy AI solutions for computational healthcare, combining advanced deep learning architectures with practical medical applications to drive meaningful impact in healthcare technology.

On the news

On the news

'Vietnam strength' turns student into an international teacher

A desire to help the Vienamese disapora learn their mother tongue has turned a student into a teacher with students from many countries.

On the news

How to conquer the top two universities in the world

Less than a month to prepare the application, Truong Thoi Tien, 21 years old, completed 4 main essays and gained admissions into two top universities in the world.

Recent Publications

First Author

Grad-CAM-Enhanced Dual-Image Prompting of Multimodal LLMs for Trustworthy Brain Tumor Classification

Reliability-centered AI pipeline for brain tumor classification improving accuracy from 0.64 to 0.67 and enabling 68% case automation. Submitted to IEEE-RAMS.

First Author

SwinNeXt-HAADA: Hierarchical Adaptive Attention-Driven Dual-Stream Architecture for Medical Image Classification

Advanced deep learning model achieving 99.56% accuracy across 22,000+ medical images. Submitted to IEEE Journal of Biomedical and Health Informatics.

First Author

Sleep State-Dependent EEG Spectral Alterations During Obstructive Sleep Apnea in Chronic Traumatic Brain Injury

Novel EEG biomarker discovery from 220 patient datasets for traumatic brain injury screening. IEEE-ICHST-2025.

First Author

Integrating Advanced Sensing Technologies and Artificial Intelligence for Predicting Cardiovascular Risks

Book chapter analyzing 193 sources and coordinating 13 international co-authors across major medical institutions for cardiovascular disease prediction.

Co-Author

Learning Disorder Detection Using Eye Tracking: Are Large Language Models Better Than Machine Learning?

GenEAI Best Paper Award winner at Symposium on Eye Tracking Research and Applications. Novel approach combining eye tracking with LLMs for learning disorder detection.

Co-Author

ECGConVT: A Hybrid CNN and Vision Transformer Model for Enhanced 12-Lead ECG Images Classification

Published in IEEE Access. Hybrid CNN-Vision Transformer architecture for enhanced ECG image classification, combining convolutional and attention mechanisms.

Co-Author

Multi-Scale System Reliability Analysis of Multi-Layer Network Infrastructures

Published in 2025 Annual Reliability and Maintainability Symposium (RAMS). System reliability analysis for multi-layer network infrastructures.

Co-Author

Multi-Level Phenotypic Models of Cardiovascular Disease and Obstructive Sleep Apnea Comorbidities

Published in PLOS One. Longitudinal Wisconsin Sleep Cohort Study analyzing cardiovascular disease and obstructive sleep apnea comorbidities.

Data Science Projects

Turning curiosity into discovery changing what we know about the world.

"We think that when we make a discovery, we've answered a question; but almost always whar we've done is pose a new question" - Nergis Mavalvala

Developing a Convolutional Neural Network (CNN) Model for Accurate House Number Classification in the SVHN Dataset: Importance and Applications

The main goal of this project is to develop a CNN model capable of accurately classifying house numbers in the Street View House Numbers (SVHN) dataset, enabling address recognition, automation, urban planning insights, and improving navigation systems.

Forecasting Google Stock Prices using Recurrent Neural Networks (RNN) from 2012 to 2017

In this project, I aim to utilize Recurrent Neural Networks (RNN) to develop predictive models that can capture patterns and dependencies in the complex and volatile stock price data of Google (GOOGL) from 2012 to 2017, enabling more accurate stock market predictions.

Car Price Prediction using Linear Regression and Lasso

Utilizing Linear Regression and Lasso regularization, this model enables precise car price predictions, empowering buyers and sellers with informed decisions. Various attributes and features contribute to a reliable car valuation system, facilitating well-informed choices in the market.

Predictive Modeling of California Housing Prices: A Comparative Analysis with XGBoost

The aim of this project is to develop a machine learning model that predicts housing prices in California based on the California housing dataset. The dataset contains various features related to houses in different locations across California, such as median income, house age, average number of rooms and bedrooms, population, and geographical coordinates.

Battling Misinformation: Predictive Modeling for Fake News Detection

This project aims to develop a machine learning model for detecting and predicting fake news. By leveraging advanced natural language processing techniques and various features extracted from news articles and related data, I strive to build a robust model capable of distinguishing between genuine and deceptive information.

Modified National Institute of Standards and Technology Database Project

The goal of the project is to develop a machine learning model capable of accurately recognizing and classifying these handwritten digits. By training a neural network on the MNIST dataset, I aim to build a model that can generalize well and correctly classify new, unseen handwritten digits.