Ashfak Yeafi

East Lansing, Michigan, USA

👨‍🔬 About Me

I’m Ashfak Yeafi, a graduate researcher in Biosystems and Agricultural Engineering at Michigan State University (MSU), where I work under the guidance of Prof. Yuzhen Lu. My research focuses on integrating Artificial Intelligence (AI) and Hyperspectral Light Scattering Imaging to develop non-invasive methods for early-stage rot detection in fruits, particularly blueberries. This work aims to revolutionize quality control in the agricultural supply chain using advanced machine learning, computer vision, and spectral analysis techniques.

My long-term goal is to pursue a Ph.D. in Computer Science or AI, with a focus on vision-language models, generative AI, and transformer architectures applied to medical and agricultural imaging.

🎓 Academic Background

M.S. in Biosystems Engineering (2025 – Present)
Michigan State University, USA
Research: AI-driven rot detection using hyperspectral imaging
B.Sc. in Electrical and Electronic Engineering (2019 – 2024)
Khulna University of Engineering & Technology (KUET), Bangladesh
Thesis: 3D GAN & dynamic convolution for brain tumor segmentation

🔬 Research Interests

Computer Vision in Healthcare and Agriculture
Deep Learning (CNNs, RNNs, Transformers, Hybrid Models)
Hyperspectral and Multispectral Imaging
Biomedical Image and Signal Processing
Generative Models (GANs, VAEs, Diffusion Models)
Embedded AI with ROS, Gazebo, and real-time systems
Digital Agriculture & Non-destructive Testing

🤖 Technical Skills

Programming: Python, C, C++, C#
AI Frameworks: PyTorch, TensorFlow, Keras
Vision & Imaging: OpenCV, Scikit-image, Pillow
Web: Flask, HTML, Jinja2, WordPress
Robotics: ROS, Gazebo, Rviz, CMake
Modeling/Simulation: AutoCAD, Fusion 360, Simulink
EDA & Tools: MATLAB, Proteus, Excel, Tableau
Fabrication: 3D Printing, Laser Cutting, Cura, PrusaSlicer
OS: Linux (Ubuntu), Windows

🏆 Achievements

🥇 Kaggle Datasets Grandmaster (Ranked #24 worldwide)
🥈 Kaggle Notebook Master (Top 250)
🛰️ Winner, Indian Rover Design Challenge 2024
🛸 Software Sub-Team Co-Lead, KUET Mars Rover – Developed autonomous navigation system

🌐 Online Profiles

🔗 GitHub – Research projects, tools, and open-source contributions
🔗 Kaggle – Competitions, datasets, and notebooks
🔗 LinkedIn – Professional profile
🔗 Google Scholar – Research profile

📫 Get in Touch

I’m always open to research collaborations, internships, and discussions about AI in medicine, agriculture, or embedded systems.

📧 Email: yeafiashfak@gmail.com
🏛️ Department of Biosystems and Agricultural Engineering, Michigan State University
🗓️ Available for academic collaborations, proposal writing, and paper reviews

“The true sign of intelligence is not knowledge but imagination.”
— Albert Einstein

selected publications

Implementation of Encoder-Decoder based Long Short-Term Memory Network for Short-Term Electrical Load Forecasting

Amit Deb Roy , and Ashfak Yeafi

In 2022 4th International Conference on Sustainable Technologies for Industry 4.0 (STI) , 2022

Bib HTML PDF

@inproceedings{10103285,
  author = {Roy, Amit Deb and Yeafi, Ashfak},
  booktitle = {2022 4th International Conference on Sustainable Technologies for Industry 4.0 (STI)},
  title = {Implementation of Encoder-Decoder based Long Short-Term Memory Network for Short-Term Electrical Load Forecasting},
  year = {2022},
  volume = {},
  number = {},
  pages = {1-6},
  keywords = {Radio frequency;Measurement;Recurrent neural networks;Load forecasting;Predictive models;Power systems;Forecasting;Short-Term Load Forecasting;Long-Short-Term Memory Network;Bangladesh Power System;Evaluation Metrics},
  doi = {10.1109/STI56238.2022.10103285},
}

Advancing automatic plant classification system in Saudi Arabia: introducing a novel dataset and ensemble deep learning approach

Emna Amri , Yonis Gulzar , Ashfak Yeafi, and 3 more authors

Modeling Earth Systems and Environment, 2024

Bib HTML PDF

@article{amri2024advancing,
  title = {Advancing automatic plant classification system in Saudi Arabia: introducing a novel dataset and ensemble deep learning approach},
  author = {Amri, Emna and Gulzar, Yonis and Yeafi, Ashfak and Jendoubi, Siwar and Dhawi, Faten and Mir, Mohammad Shuaib},
  journal = {Modeling Earth Systems and Environment},
  pages = {1--17},
  year = {2024},
  publisher = {Springer},
}

GSNet: a multi-class 3D attention-based hybrid glioma segmentation network

Md Tasnim Jawad , Ashfak Yeafi, and Kalyan Kumar Halder

Optics Express, 2023

Bib HTML PDF

@article{jawad2023gsnet,
  title = {GSNet: a multi-class 3D attention-based hybrid glioma segmentation network},
  author = {Jawad, Md Tasnim and Yeafi, Ashfak and Halder, Kalyan Kumar},
  journal = {Optics Express},
  volume = {31},
  number = {24},
  pages = {40881--40906},
  year = {2023},
  publisher = {Optica Publishing Group},
  url = {https://opg.optica.org/oe/fulltext.cfm?uri=oe-31-24-40881&id=542523},
}

A Semi-supervised Approach For Brain Tumor Classification Using Wasserstein Generative Adversarial Network with Gradient Penalty

Ashfak Yeafi, Monira Islam , Sohag Kumar Mondal , and 2 more authors

In 2023 6th International Conference on Electrical Information and Communication Technology (EICT) , 2023

Bib HTML PDF

@inproceedings{yeafi2023semi,
  title = {A Semi-supervised Approach For Brain Tumor Classification Using Wasserstein Generative Adversarial Network with Gradient Penalty},
  author = {Yeafi, Ashfak and Islam, Monira and Mondal, Sohag Kumar and Nashad, KM Ishraq Hussain and Yusuf, Md Salah Uddin},
  booktitle = {2023 6th International Conference on Electrical Information and Communication Technology (EICT)},
  pages = {1--6},
  year = {2023},
  organization = {IEEE},
}

ADTNet: Attention-Guided U-Net with Dynamic CNN and Transformers for Skin Cancer Detection

Ashfak Yeafi, and Liton Sarker

In 2024 13th International Conference on Electrical and Computer Engineering (ICECE) , 2024

Bib HTML PDF

@inproceedings{11024876,
  author = {Yeafi, Ashfak and Sarker, Liton},
  booktitle = {2024 13th International Conference on Electrical and Computer Engineering (ICECE)},
  title = {ADTNet: Attention-Guided U-Net with Dynamic CNN and Transformers for Skin Cancer Detection},
  year = {2024},
  volume = {},
  number = {},
  pages = {679-684},
  keywords = {Image segmentation;Technological innovation;Accuracy;Convolution;Computational modeling;Computer architecture;Transformers;Feature extraction;Skin;Skin cancer;Skin Cancer Segmentation;Dynamic Convolution;Attention Mechanism;Transformer},
  doi = {10.1109/ICECE64886.2024.11024876},
}

EF-SwinNet: A Hybrid EfficientNet-Swin Transformer Model for Skin Cancer Classification

Liton Sarker , and Ashfak Yeafi

In 2024 International Conference on Recent Progresses in Science, Engineering and Technology (ICRPSET) , 2024

Bib HTML PDF

@inproceedings{sarker2024ef,
  title = {EF-SwinNet: A Hybrid EfficientNet-Swin Transformer Model for Skin Cancer Classification},
  author = {Sarker, Liton and Yeafi, Ashfak},
  booktitle = {2024 International Conference on Recent Progresses in Science, Engineering and Technology (ICRPSET)},
  pages = {1--4},
  year = {2024},
  organization = {IEEE},
}