Research Projects
Tuan Minh Le's research program operates at the interface of advanced evolutionary computing, artificial intelligence, and applied cyber-physical instrumentation. Select a project area below to deep dive into published breakthroughs.
Advanced Metaheuristics for Feature Selection
Pioneering hybrid evolutionary computation frameworks, including Adaptive Particle Swarm Optimization (PSO) and Grey Wolf Optimization (GWO). These techniques reduce search space and eliminate redundant variables in large-scale medical and industrial datasets, significantly boosting classifier accuracy while lowering training costs.
Clinical Signal Processing & Diagnostic Classification
Developing cutting-edge machine learning and deep learning architectures to analyze non-invasive neurophysiological datasets. Developed automated diagnostic platforms that utilize Fast Fourier Transforms (FFT) and deep convolutional neural networks to perform real-time epileptic seizure classification, Parkinson's vocal degradation screening, and early diabetes forecasting.
IoT Telemetry & Predictive Fault Isolation
Prototyping high-reliability embedded control frameworks and deep-learning-driven edge diagnostics for industrial mechanical units. Implemented automated fault isolation pipelines that employ multi-domain feature extraction and state-of-the-art Kolmogorov-Arnold Networks (KAN) to accurately classify rotating bearing faults, fan motor imbalances, and real-time UAV flight stabilization.
