Summary

Computer Engineering PhD candidate and experienced software engineer with a strong background in machine learning, data analysis, image processing, and cognitive engineering. Dedicated to developing innovative solutions and making significant contributions to the field.

Education

• Ph.D. Candidate, Computer Engineering, University of Memphis (GPA: 3.90, Expected: May 2024)
• M.S. in Electrical and Computer Engineering, University of Memphis (GPA: 3.88, Dec 2021)
• Graduate Certificate in Data Science, University of Memphis (May 2022)
• B.S. in Electrical and Computer Engineering, Korea Advanced Institute of Science and Technology (May 2019)

Professional Experience

Graduate Assistant, University of Memphis (2019-Present)

1. Research Experience

• Predicting Cognitive Load from EEG with Optimized CNNs (2019-2021)
  • Developed parameter-optimized CNN models for cognitive load prediction from raw EEG recordings.
  • Compared the predictive power of individual and composite frequency band representations, achieving up to 90% accuracy.
  Funding: Carnegie R1 Doctoral Fellowship
• Enhancing Cognitive Load Prediction Using Deep Learning Leveraging Multimodal Representations of EEG (2021-present)
  • Developed a deep learning framework for modeling spatial-spectral dynamics of cognitive load (CL) and achieved 94% CL prediction accuracy.
  • Implemented eigenspace bootstrap sampling to address EEG noise and enhance ERP data generation.
  • Developed GAN methods for synthetic EEG data generation and achieved >4% improvement on the CL prediction task.
  • Currently designing deep CNN models of CL prediction using a spatial-spectral-temporal representation of EEG.
  • Currently designing graph-based deep-learning models for CL prediction using static and dynamic EEG connectivity measures.
  • Developing deep learning techniques for mental load detection in real-time scenarios (e.g., driver drowsiness detection).
  Funding: NIH/NIDCD Grant R01DC016267, Graduate Assistantship through the EECE department
• Currently building deep and interpretable neural network models for Epileptic Seizure Detection and Classification
  Funding: Graduate Assistantship through the EECE department
• Speech Categorization with Convolutional Neural Networks (2022)
  • Investigated response time variation in speech perception using EEG data and parameter-optimized CNNs.
  • Identified key brain regions and probable factors influencing response time.
  Funding: NIH/NIDCD Grant R01DC016267
• Application of Machine Learning for Modeling Risk Decision-Making in Rats (2022)
  • Trained various classical ML models (e.g., Random Forest) for rodents’ risky choice prediction (risky vs. safe) using micro-electrode data from single neurons, achieving >85% accuracy.
  • Identified temporal stages that influence the rodent’s decision-making process.

2. Teaching Experience

Skills and Expertise

Programming Languages: Python, R, Java, C#, LabView, MATLAB. Software and Tools: LaTeX, Git, OpenCV, Jupyter Notebook, Keras, Microsoft SQL Server. Libraries: NumPy, pandas, statsmodels, seaborn, Matplotlib, Plotly, Keras, TensorFlow, PyTorch. Techniques: Machine Learning, Data Analysis, Optimization, Image Processing, Deep Learning, Computer Vision, EEG Analysis, 3D Visualization, Desktop and Web Applications Development.

Awards and Honors

• 1st Prize at 9th EECE Annual Poster Competition, University of Memphis (April 2023)
• Carnegie R1 Doctoral Fellowship, University of Memphis (2019-2021)
• KAIST International Excellence Scholarship, KAIST (2015-2019)

Professional Membership

IEEE, NSBE, University of Memphis GSA, African Student Association

Publications

• Havugimana F., Moinudin K. A., & Yeasin M. (2023). Deep Learning Framework for Modeling Cognitive Load from Small and Noisy EEG data. IEEE Transactions on Cognitive and Developmental Systems.
• Moinuddin K. A., Havugimana F., Al-Fahad R. B., Bidelman G. M., & Yeasin M. (2022). Unraveling Spatial-Spectral Dynamics of Speech Categorization Speed Using Convolutional Neural Networks. Brain Sciences, 13(1).
• Havugimana F., Muhammad M. B., Moinudin K. A., & Yeasin M. (2021). Predicting Cognitive Load using Parameter-optimized CNN from Spatial-Spectral Representation of EEG Recordings. In 2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA), pp. 710-715.