Research agenda
I build reliable AI systems for clinical, speech, multimodal, agricultural, and scientific applications.
My research asks how learned representations can remain useful, accountable, and resilient when they meet people, sensitive data, and changing environments.
Healthcare & Clinical AI
Clinical AI work focused on privacy-preserving language systems, uncertainty, unlearning, and dependable decision support.
This domain treats privacy and reliability as model-design constraints rather than post-processing concerns.
Current questions
- How can clinical language systems protect sensitive tokens without discarding useful context?
- How can uncertainty and unlearning improve the reliability of multimodal clinical models?
Methods
- Client-side token selection and cryptographic redaction
- Vision-language modelling, uncertainty estimation, and selective unlearning
Representative papers
- Selective Token-Level Cryptographic Redaction for Privacy-Preserving Clinical Deployment of Large Language Models
- Uc-PrUn: Uncertainty-Calibrated Machine Unlearning using Vision-Language Models for Clinical Decision Support
- A discrete mathematical model and cryptography for secure medical image analysis: Encrypted chest X-ray classification
Collaboration interests
- Privacy-preserving clinical NLP
- Calibrated and auditable multimodal decision support
Privacy, Trust & Safety
Trustworthy AI research spanning encrypted inference, secure information handling, and robust detection of harmful inputs.
The emphasis is on explicit threat, utility, and evidence boundaries for applied systems.
Current questions
- What privacy utility trade-offs appear when only sensitive content is protected?
- How can evaluation expose brittle behaviour under constrained or adversarial inputs?
Methods
- Cryptographic protection and privacy-utility evaluation
- Robustness testing across curated and external datasets
Representative papers
- Selective Token-Level Cryptographic Redaction for Privacy-Preserving Clinical Deployment of Large Language Models
- A discrete mathematical model and cryptography for secure medical image analysis: Encrypted chest X-ray classification
- Discrete mathematical models for enhancing cybersecurity: A mathematical and statistical analysis of machine learning approaches in phishing attack detection
Collaboration interests
- Privacy-aware deployment of language and vision systems
- Reliable evaluation for security-sensitive ML
Speech, Audio & Synthetic Media
Speech-forensics work on cross-paradigm detection, generator attribution, open-set recognition, and emotional manipulation.
This line of work connects geometric representation learning with practical forensic questions about provenance and generalization.
Current questions
- Which representation geometries remain useful across synthesis paradigms and languages?
- How should a detector respond when a generator was not present during training?
Methods
- Speech foundation models with hyperbolic, spherical, and Euclidean fusion
- Graph-based attribution and confidence-aware open-set inference
Representative papers
- Bridging Attribution and Open-Set Detection using Graph-Augmented Instance Learning in Synthetic Speech
- Curved Worlds, Clear Boundaries: Generalizing Speech Deepfake Detection using Hyperbolic and Spherical Geometry Spaces
- Towards Attribution of Generators and Emotional Manipulation in Cross-Lingual Synthetic Speech using Geometric Learning
- Synergizing Zero-Shot Cross-Lingual Alzheimer Detection with Language-Invariant Multimodal Bi-Geometric Adversarial Learning
Collaboration interests
- Robust speech deepfake detection
- Cross-lingual and open-set synthetic-media forensics
Multimodal & Human-Centered AI
Multimodal learning for human traits and performance, with datasets and models designed around naturalistic evidence.
The work begins with the quality and structure of human-centred data, then carries those constraints into model design and evaluation.
Current questions
- How can video, audio, and language representations retain complementary structure?
- How should model capacity and geometry be balanced for practical training?
Methods
- Multimodal dataset construction and pairwise comparative annotation
- Manifold-specific experts with adaptive routing and tangent-space fusion
Representative papers
- RecruitView: A Multimodal Dataset for Predicting Personality and Interview Performance for Human Resources Applications
- Synergizing Zero-Shot Cross-Lingual Alzheimer Detection with Language-Invariant Multimodal Bi-Geometric Adversarial Learning
- Data-efficient neuroimaging classification via Wasserstein Autoencoder-based augmentation and hybrid deep learning
Collaboration interests
- Human-centred multimodal benchmarks
- Efficient models for behavioural and clinical signals
Scientific & Applied AI
Scientific machine learning for thermal-property prediction using augmentation, ensembles, and metaheuristic search.
Scientific applications provide a test of whether a model remains useful beyond benchmark settings and under limited observations.
Current questions
- How can scarce experimental data support high-fidelity property prediction?
- Which combinations of augmentation, ensembles, and optimization are useful across thermophysical tasks?
Methods
- Literature-curated experimental datasets and controlled augmentation
- Ensemble learning, neural models, and metaheuristic hyperparameter search
Representative papers
- Learning heat: High-fidelity experimental and Koo-Kleinstreuer-Li thermal conductivity predictions in nanofluids via advanced data augmentation and metaheuristic search
- A Computational Intelligence Framework Integrating Data Augmentation and Meta-Heuristic Optimization Algorithms for Enhanced Hybrid Nanofluid Density Prediction Through Machine and Deep Learning Paradigms
- Comprehensive framework of machine learning and deep learning architectures with metaheuristic optimization for high-fidelity prediction of nanofluid specific heat capacity
Collaboration interests
- Data-efficient scientific ML
- Reproducible surrogate modelling for physical systems
Earth & Agricultural Intelligence
Computer vision and decision support for soil, rock, crop, and medicinal-herb intelligence.
This domain links visual evidence to practical decisions in agriculture and earth science while keeping dataset scope explicit.
Current questions
- How can visual models handle small, heterogeneous datasets in environmental settings?
- How can predictions become actionable recommendations without hiding uncertainty?
Methods
- Transfer learning, CNN-ViT ensembles, and image augmentation
- Fuzzy decision systems and cross-validation for applied recommendations
Representative papers
- An advanced artificial intelligence framework integrating ensembled convolutional neural networks and Vision Transformers for precise soil classification with adaptive fuzzy logic-based crop recommendations
- Advancing Geological Image Segmentation: Deep Learning Approaches for Rock Type Identification and Classification
- Herbify: An ensemble deep learning framework integrating convolutional neural networks and vision transformers for precise herb identification
- Deep insight: Mathematical modeling and statistical analysis for mango leaf disease classification using advanced deep learning models
Collaboration interests
- Accessible agricultural decision-support tools
- Robust visual models for environmental and botanical data