Majid Bagheri
Savannah State University, GA·Computer Science
Pollution in water, soil and air is hard to track, and some of it ends up in the food people and livestock eat. This research applies artificial intelligence (AI) and machine learning (ML), meaning computer programs that learn patterns from data, to two linked problems.
- Fields
Computer Science
Earth & Environmental Science
Plant Science & Agriculture
- Topics
artificial intelligence
machine learning
TinyML
- Grants
- 2 known grants, through Aug 2027
What they work on
- Train deep neural networks, convolutional neural networks and long short-term memory networks on water, soil and air datasets to support pollution control, using both training from scratch and transfer learning.
- Compress TensorFlow deep learning models with TensorFlow Lite and TensorFlow Lite Micro into TinyML models that run on microcontrollers for on-site monitoring of environmental contaminants.
- Use molecular dynamics simulations to examine how emerging contaminants interact with plant root materials, linking contaminants' physical and chemical properties to how well they cross membranes.
- Apply neural networks, fuzzy logic and clustering to predict how contaminant uptake varies among plant species, to find candidates for phytoremediation and crops at risk of contaminated edible tissues.
- Develop a TinyML computer-vision system that detects plants affected by emerging contaminants from images.
More about the research
One is monitoring environmental contaminants. The other is predicting how plants take up these contaminants from soil. A recurring tool is tiny machine learning (TinyML): AI models shrunk enough to run on cheap, low-power microcontroller chips in the field.
Why it matters
Contaminants absorbed by crops can reach people and livestock through food. These tools could help farmers and regulators find which crops are most at risk and flag contaminated plants with simple image-based devices. They could also point to species suited for cleaning polluted soil. Low-cost TinyML sensors may make on-site monitoring of water, soil and air pollution more affordable, which matters for underserved communities.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Aaron Maxwell
West Virginia University, WV·Earth & Environmental Science·Early-career
Develop convolutional neural network methods for semantic and instance segmentation to extract geomorphic features from LiDAR point clouds and terrain derivatives such as hillshades.
James S Duncan
Yale University, CT·Biomedical Engineering
Develop image-analysis methods that learn jointly from repeated MRI and CT scans of liver cancer patients. The goal is to track the tumor's surroundings during combined chemoembolization and immunotherapy and to predict outcomes.
Tahir Ekin
Texas State University, TX·Computer Science
Develop an Applied AI Summer School and new course modules in data analytics, Python, deep learning, natural language processing and computer vision for students in non-AI disciplines, using inclusive and contextualized teaching methods.
The school
Savannah State University: 1 labs on PI Pages.
Reach out
A start for your email. Fill in the brackets.