From 1,052 to 8,563: How AI Coding Agents Transformed My Research Workflows
How integrating Claude Code into academic research workflows led to an 8x increase in output, and a 10-week course to help other researchers do the same.
Seyed Yahya Shirazi, Ph.D.
Recent content on Seyed Yahya Shirazi, Ph.D.
Latest posts
Annotation Garden Initiative: Collaborative Annotation for Neuroscience
The Annotation Garden Initiative (AGI) establishes an open infrastructure for collaborative, multi-layered annotation of stimuli used in neuroscience research, building on BIDS and HED standards.
Image Annotation Tool: VLM-Based Annotation for Neuroscience
A VLM-based image annotation system for the Annotation Garden ecosystem, providing collaborative annotation capabilities for static image datasets, starting with the Natural Scene Dataset (NSD).
Age-related Reorganization of Corticomuscular Connectivity During Locomotor Perturbations
Young and older adults show fundamentally different patterns of brain-muscle communication during perturbed recumbent stepping, revealing an age-related shift from flexible, error-driven control to a more constrained, stability-focused approach. Preprint available on bioRxiv, September 2025.
The Lab Streaming Layer for Synchronized Multimodal Recording
LSL synchronizes data across devices on LANs for neurophysiological research, offering easy, reliable multi-device integration and supporting diverse programming languages.
Signal Processing for Wearable Biosensors: Fundamentals and Techniques for Ring-Based Devices
Comprehensive guide to digital signal processing techniques for ring-based wearable biosensors, covering theoretical foundations, filtering methods, and real-time implementation strategies.
Health-Specific Evaluation for AI Systems
A guide to health-specific evaluation metrics and frameworks for AI systems, focusing on clinical validity, FDA regulatory considerations, bias detection, safety evaluation, and practical implementation strategies.
Statistical Analysis for Evaluation
A guide to statistical methods for evaluating models, including power analysis, mixed-effects models, bootstrap confidence intervals, multiple comparison corrections, and effect size calculations.
LLM Evaluation Methods
A guide to evaluating large language models (LLMs) using automatic metrics, LLM-as-judge paradigms, human-in-the-loop strategies, and domain-specific considerations.
Human Evaluation & Psychometrics for AI Systems
A comprehensive guide to designing and implementing human evaluation frameworks and psychometric principles for AI systems, with practical algorithms and code snippets.
NEMAR Dataset Citations Analysis Dashboard
Interactive dashboard analyzing citation patterns and research impact of NEMAR (NeuroElectroMagnetic Archive and Tools Resource) datasets, featuring network analysis, temporal trends, and research theme identification.
Creating Interactive Dashboards in Hugo: A Complete Guide
A comprehensive guide to building interactive dashboards in Hugo static sites using Chart.js, with best practices for data management and visualization.
HBN Dataset Insights Dashboard
This dashboard provides comprehensive insights into the Healthy Brain Network (HBN) EEG dataset, featuring demographics, task availability, and mental health correlations across all releases.
Hyser: High-Density Surface EMG Dataset for Neural Interface Research
Hyser is an open-access high-density surface EMG dataset with 256-channel recordings from 20 subjects performing hand gestures and finger force tasks, designed for neural interface and prosthetic control research.
Adding Mermaid Diagrams to Hugo: From Partials to Production
A comprehensive guide to integrating Mermaid diagrams in Hugo static sites using partials, with practical examples and deployment strategies.
Simulating Scalp EEG from Ultrahigh-Density ECoG Data Illustrates Cortex to Scalp Projection Patterns
This paper challenges the misleading assumption that EEG channels reflect only local cortical activity by using ultrahigh-density ECoG data to demonstrate how cortical activity projects broadly across the scalp surface through volume conduction.
Python's venv: A Comparison with Conda for Research Computing
A guide to Python's built-in venv module with comparisons to Conda for research computing workflows
Setting Up and Managing Git-Annex with Backblaze B2 Storage
A guide to setting up and managing git-annex with Backblaze B2 storage for large files.
A Standardized Framework for Sensor Placement in Human Motion Capture and Wearable Applications
A unified framework for standardizing sensor placement across different sensing modalities and applications, providing precise anatomical landmarks, coordinate systems, and placement protocols with defined precision levels.
HBN-EEG: Healthy Brain Network EEG Datasets
HBN-EEG is a curated collection of high-resolution EEG data from over 3,000 participants aged 5-21 years, formatted in BIDS and annotated with Hierarchical Event Descriptors (HED). These datasets support large-scale analyses and machine-learning research related to mental health in children and adolescents.
HBN-EEG: The FAIR implementation of the Healthy Brain Network (HBN) electroencephalography dataset
This paper presents the HBN-EEG dataset, a comprehensive and analysis-ready collection of high-density EEG recordings from the Healthy Brain Network project, formatted in BIDS with annotated behavioral and task-condition events, aimed at supporting EEG analysis methods and the development of EEG-based biomarkers for psychiatric disorders.
Older adults use fewer muscles to overcome perturbations during a seated locomotor task
Older adults use fewer muscles to drive a stepper during a seated locomotor task, demonstrating greater co-contraction and similar motor errors compared to young adults. Pulsihed at the Journal of Neurophysiology, May 2024.
System and methods for biosignal detection and active noise cancellation
Innovative EEG system with dual electrodes and advanced signal processing for high-quality, noise-reduced biosignal capture and enhanced user comfort. US Patent Application No. 2023/0240581A1.
Non-Parametric Functional Muscle Network as a Robust Biomarker of Fatigue
This paper shows, for the first time, that using a non-parametric functional muscle network can reliably characterize fatigue-related changes in synergistic muscle coordination and distribution of neural drive at the peripheral level. Published in IEEE Journal of Biomedical and Health Informatics, 2023.
Re-Referencing Methods for High-Density EEG
Analysis of re-referencing methods for high-density EEG recordings and their impact on source estimation accuracy.
Nonlinear functional muscle network based on information theory tracks sensorimotor integration post stroke
Muscle connectivity network based on information theory quantifies the role of sensory information on motor performance. Published in Scientific Reports, 2022.
Perilaryngeal-Cranial Functional Muscle Network Differentiates Vocal Tasks: A Multi-Channel sEMG Approach
This paper shows, for the first time, that the functional muscle network can robustly differentiate vocal tasks, while classic muscle activation assessment fails to differentiate. Published in IEEE Transactions on Biomedical Engineering, 2022.
Sensory feedback and assistive motor control for desk cycling
Investigating the role of sensory feedback in assistive motor control during desk cycling activities for improved user experience and rehabilitation outcomes.
Differential Theta-Band Signatures of the Anterior Cingulate and Motor Cortices During Seated Locomotor Perturbations
We show that seated locomotor tasks can elicit perturbation-related cortical dynamics similar to walking. Published in the IEEE Transactions on Neural Systems and Rehabilitation Engineering.
More Reliable EEG Electrode Digitizing Methods Can Reduce Source Estimation Uncertainty, but Current Methods Already Accurately Identify Brodmann Areas
This paper examines the reliability of EEG electrode digitizing methods using five different techniques, and the potential impact of digitization reliability on source estimation uncertainty. Published in Frontiers in Neuroscience, 2019.
Influence of Mismarking Fiducial Locations on EEG Source Estimation
This paper shows how fiducial mismarking can affect EEG source estimation. 2019 9th International IEEE/EMBS Conference on Neural Engineering (NER)
Location
Mailing and office addresses at UCSD.