Recent research:
The researcher’s work spans natural language processing, health informatics, and signal analysis, blending AI-driven methods with practical sensing and accessibility applications. Across publications, a core thread is the use of advanced machine learning and deep learning (including transformers, CNNs, VGGish, and GANs) to interpret complex signals—text, voice, and biomedical data—and translate them into actionable insights. Several studies emphasize human-centered design and accessibility (sign language translation, deaf education tools) while others advance healthcare analytics (noninvasive glucose estimation, asthma/COPD detection from lung sounds, COVID-19 cough classification). Methodologies routinely combine feature extraction (MFCCs, Mel spectrograms, phonotactics, n-grams), state-of-the-art classifiers (CNNs, transformer encoders, transfer learning, ensemble/voting schemes), and lightweight, deployable implementations (Arduino, Unity 3D, Node-RED, LoRaWAN). Impact is demonstrated through practical systems (sign language translator, urban water monitoring with predictive analytics, portable diagnostic prototypes) and comparative analyses showing competitive accuracy, correlation, and improved generalization with data augmentation and multimodal features. The portfolio reflects a multidisciplinary trajectory aimed at enabling precise, accessible, and scalable AI solutions in education, health, and environmental monitoring.

Key Topics:

• Natural Language Processing, Conversational AI & Text Analytics
• Artificial Intelligence, Machine Learning & Data Science
• Healthcare, Clinical Epidemiology & Public Health
• Sensors, Instrumentation & Measurement Systems
• Sign Language Translation, Deaf Education & Accessibility
• Image/Video and Audio Signal Processing
• Smart Grids, IoT, Edge/Fog & Industry 4.0
• Environmental Monitoring & Urban Water Management