Tasnim Ferdous completed her Master's in Electrical and Computer Engineering from the Oregon State University, USA. Prior to this, she received her BSc. in EEE from AIUB, Bangladesh in 2012. After her masters, she worked as an Electrical Engineer in Electro Scientific Industries, Portland, OR, USA.
Once she returned back to Bangladesh, she joined Northern University Bangladesh as a faculty. After serving there a year, she joined Brac University.

Her current research interests are in Computational Neuroscience, Neuroinformatics and Digital Health, Biomedical Signal processing and Deep learning. Prior to this, she had research interest and hands-on experience in Analog/Mixed-signal Integrated circuit analysis, design and testing. 

Beside working, she loves to sketch and frame beautiful sights on paper. She also spends time reading on various topics, hiking within the woods and trying out different recipes.

Google Scholar
ORCID

Conferences

Ferdous, T. (2012). Design and FPGA-based implementation of a high performance 32-bit DSP processor. 2012 15th International Conference on Computer and Information Technology (ICCIT), 484–489. https://doi.org/10.1109/ICCITechn.2012.6509808

1. Computational Neuroscience/ Brain Computer Interface(BCI)
   Neuroinformatics and digital health
   Biomedical Signal processing.
   Sensor electronics, Machine Learning and Deep learning

1. Brain Computer Interface based gamified neurofeedback 
   This study integrates Brain-Computer Interface (BCI) technology to enhance attention, memory, and manage fatigue-induced drowsiness. It combines Beta-to-(Alpha+Theta) Ratio and Frontal Midline Theta (FM-theta) neurofeedback with eye tracking to stimulate brain activity associated with cognitive performance. A gamified platform provides real-time EEG feedback, while advanced neural networks classify mental states for effective cognitive enhancement.

Machine and Deep learning on PPG based automated Sleep stage classification  
This project focuses on creating a deep learning model that uses PPG signals to classify sleep stages accurately. The model will be trained on data from both healthy people and those with mild sleep problems, aiming to detect abnormal Critical Arousal Periods (CAP) linked to sleep disorders. It will use techniques like PCA for feature reduction and attention-based CNNs, along with Explainable AI (XAI), to highlight the most important features. The model will also be adapted for smartwatches, allowing real-time, home-based sleep tracking for both healthy individuals and those with sleep issues.

AI-Based Early Detection of Parkinson’s Disease from Facial Emotion and Motor Symptoms 
An AI-based approach using computer vision for the early detection of Parkinson’s disease by analyzing facial emotions and motor symptoms. Deep learning models will process subtle changes in facial expressions and movement patterns from video data to identify early biomarkers. The study aims to develop an efficient, non-invasive diagnostic tool for timely intervention and improved patient outcomes.