Projects

Computer Vision for Wildlife Monitoring

Automated wildlife census and habitat analysis across camera-trap, thermal, and aerial imagery

Perception pipelines spanning ground-level RGB, thermal infrared, and drone-based orthoimagery to support large-scale ecological surveys and wildlife population studies.

Smartphone-Based Chlorophyll Monitoring System Using AI Models for Precision Stress Management in Soybeans

Award: Soybean Center, Southern Illinois University Carbondale, $15,000

Smartphone-based AI for precision stress management in soybean fields

Developing a smartphone-based chlorophyll monitoring system powered by AI models to detect early nutrient and stress indicators in soybean fields, enabling low-cost, in-field precision stress management.

AI-Driven Discovery of Sustainable Ionic Liquids: A Multimodal Deep Learning Approach for Enhanced Lignin and Plastic Deconstruction

Award: IIN Sustaining Illinois

Multimodal deep learning for accelerated discovery of sustainable solvent systems for biopolymer and plastic valorization

Vision-centric deep learning frameworks that fuse 3D imaging cues with cross-modal signals to forecast solvent–substrate interactions, enabling faster screening of sustainable solvent candidates.

AI-Powered Early Detection of Foodborne Pathogens in Microscopy

Growth-stage-aware instance segmentation of Salmonella and E. coli microcolonies for rapid food-safety screening

Neural-architecture-searched instance segmentation that detects and measures small, low-contrast Salmonella and E. coli microcolonies in bright-field microscopy, conditioned on growth stage and robust to food-matrix interference, to enable earlier, physically calibrated pathogen screening.

Clinical Language AI for Patient Sentiment Analysis

Distilling frontier large language models into compact, deployable clinical sentiment classifiers

Benchmarking open-source transformers on medical dialogue and distilling a frontier-LLM sentiment labelling policy into a compact clinical classifier, bringing LLM-level understanding of patient sentiment to efficient, privacy-friendly on-premise deployment.