Research

My research focuses on using AI and wearable technologies to support neurodiverse individuals. I work on projects that integrate data-driven solutions and engineering innovations to improve everyday experiences for people with neurodiverse conditions, particularly in social and workplace settings.

Research Interests

My research lies at the intersection of biomedical engineering, artificial intelligence, and human-computer interaction, with a focus on developing intelligent systems to support neurodiverse populations. I am particularly interested in developing multimodal, wearable-based AI systems that interpret physiological and contextual signals in real time to support autistic adults in everyday social environments.
I focus on building human-centred, explainable, and ethically grounded AI that not only achieves high technical performance but also respects autonomy, privacy, and lived experience within the autistic and broader neurodivergent communities.

My interests are centered on the following interconnected themes:

  • AI for Neurodiversity: Designing adaptive artificial intelligence systems that bridge social cognitive gaps. My work specifically focuses on decoding neurotypical social intent to empower autistic adults, promoting greater autonomy and understanding in daily interactions.
  • Multi-Modal Data Fusion: Developing novel computational frameworks, including transformer-based deep learning models, to intelligently fuse heterogeneous data streams, such as physiological signals (ECG,GSR, EDA, PPG), contextual behavioral cues, and environmental data, for a holistic interpretation of human state and intent.
  • Wearable Sensing & Biomarker Discovery: Pioneering the use of non-invasive, wearable sensors for the discovery and validation of physiological biomarkers related to social cognition and stress. This involves advanced signal processing to extract robust features from noisy, real-world data.
  • Ethical Co-Design in Assistive Technology: A core component of my research is the co-creation of assistive technologies with the autistic community. I am deeply interested in developing ethical feedback mechanisms that balance real-time assistance with user autonomy, ensuring technology serves the user's preferences and goals.
  • Real-Time Signal Processing & Embedded AI: Engineering efficient algorithms for the on-edge processing of physiological signals, enabling low-latency, privacy-preserving inference crucial for real-world deployment of assistive systems.

Projects

1: UPEC: Portable AI-Based Cardiac Diagnostic Device

A portable AI-powered device that integrates ECG, PCG, and PPG sensing to deliver fast, reliable, and multi-modal cardiac diagnostics, enabling early detection and accessible point-of-care monitoring.

2: CardioRespi Analyzer: Integrated Heart and Respiratory Sound Tool

Cardio Respi Analyzer is a MATLAB-based application that analyzes heart and respiratory sounds from auscultation recordings for cardio-respiratory screening and visualization. It lets users load stethoscope audio, apply filters, inspect frequency content with FFT, and estimate breaths per minute and heartbeat statistics from detected peaks.

Conferences

1: 7th IEEE International Conference on Energy Conservation and Efficiency (ICECE)

2: 6th International Conference; Scientific Outlook of Sustainable Development Goals

3: 3rd International Conference on Modern Technologies in Mechanical and Materials Engineering

Abstract

1: A portable AI enabled Multimodal sensing for Cardiopulmonary Disease Detection