Fluent/Daily Use:

  • Python

  • Linux server (Ubuntu)

  • Shell (Bash) scripting

  • Git / Git-Annex / DataLad

  • Docker / Singularity

  • CI/CD

  • Sphinx docs 

  • RST / Markdown


Proficient in:

  • R


  • Amazon Web Service

  • Google Cloud Platform

Examples of projects:

  • Image segmentation deep learning tool (U-Net CNN) for delineating the fetal brain from maternal tissue in functional MRI time-series. https://github.com/saigerutherford/fetal-code

  • Novel deep learning architecture for addressing 3-D spatial invariance present in normalized structural MRI brain images. Image (T1w) to scalar (age) predictive model. https://github.com/saigerutherford/anatomically_defined_CNNs

  • Distilling complex technical topics and scientific jargon into concise and widely accessible to the diverse general public. https://github.com/predictive-clinical-neuroscience/PCNtoolkit-demo

  • Curating mega datasets (multi-study and multi-site) and turning them into federated (distributed/decentralized) privacy-preserving machine learning models that capture population-level centiles of variation. Growth charts for the human brain's lifespan. https://github.com/predictive-clinical-neuroscience/braincharts

  • Building models to predict symptom scores and treatment outcomes of psychiatric patient groups - attention deficit disorder (ADD/ADHD), schizophrenia (SZ), bipolar (BP), major depressive disorder (MDD), social anxiety disorder (SAD), post-traumatic stress disorder (PTSD). Regression and classification.

  • Discovering low-rank structure using dimensionality reduction techniques (PCA, ICA, Factor Analysis, Community Detection) in child, adolescent, and adult fMRI data (rest and task) using large publicly shared datasets, like the Human Connectome Project, ABCD Study, UKBiobank, and the Philadelphia Neurodevelopmental Cohort (PNC). I am comfortable working with large datasets (30TB+). https://github.com/SripadaLab/Predictive_Modeling_Reliability_HCP