Faraz Moradi

• Real-Time QRS Complex Detection: Implemented a QRS detection algorithm for ECG signal analysis, ensuring accurate heart rate reporting in real-time applications.
• Algorithm Optimization for Embedded Systems: Optimized ECG processing algorithms for real-time execution on embedded systems using C (Nordic Semiconductors, Zephyr).
• Respiratory Signal Estimation: Developed methods to extract respiratory signals and rates from ECG and PPG data, enhancing prediction accuracy.
• Feature extraction and classification of different sleep stages using ECG, PPG and Accelerometer data.

• Thesis: A protocol for isolating neural activity of neurons and analyzing their behavior in a pattern separation task. (paper)

• Spike sorting, manual curation, and separation of mossy and granular cells based on physiological properties and clustering algorithms (k-means).

• Characterizing different stages of sleep (wake, REM, nonREM) in rats through analysis (frequency domain analysis) of local field potentials and EMG signals.

• Large scale spike train analysis and matching behavioral correlates to neural activities using classic methods (ISIs, autocorrelations, serial correlations, PSTH, and causal filters).

• Actively movement tracking of animals in the maze using object detection and tracking methods (OpenCV: CSRT, KCF).

• Comparative analysis of local field potentials (power spectrum, time-frequency analysis, and frequency coupling) between transgenic F344AD and healthy rats

• Multi-unit extraction (spike sorting) and manual curation using toolboxes such as Tridesclous and Phy.

• Comparative analysis of spike trains between AD and normal rats using spike train distance metrics (VR-d,ISI-d,ES-d) to identify pertinent features.

• Building a machine learning classification paradigm (meta learning model) to detect AD using high and low frequency features at the early stage of the disease (6-months-old).