N-Back Memory Training
- Java-based Android app on Google Play for n-back working memory training
- 4.6 star rating, 10k+ monthly active users, 25k+ lines of Java
- Scoring with discrimination index (A’) from signal detection theory
- Data management using Firebase Realtime Database and SQLite
NIfTI Studio
- Open-source toolbox for NIfTI and Analyze (img/hdr) image visualization, editing, and 3D rendering
- View on Github, Citation
MRIqual
- Open-source toolbox toolbox for quality assurance analyses on structural and functional MRI images
- View on Github, Citation
Deep Learning Framework for Financial Time Series Prediction in Python Keras
- A set of Python functions for training small, fully connected networks or sequence models (e.g., LSTM) to predict stock prices
- Functions available for randomly partitioning data into Train/Dev/Test, creating sliding windows of various sizes, making data stationary
- Evaluate model performance via RMSE, R2, ROC curve and AUC
- Evaluate trading outcomes via Annualized return, % Profitable Trades, Profit Factor, Max Drawdown
Portfolio Sortino Ratio
- Optimize stock portfolio weights to maximize a combination of Sortino ratio, Sharpe ratio, total return, downside risk, SD, or max drawdown
- View on Github, MathWorks File Exchange
N-Back for Matlab
- The classic dual n-back task implemented in Matlab; includes position, sound, and color
- View on Github, MathWorks File Exchange, G-Node Open Data
Plot Corrmat
- A Matlab utility for plotting correlation matrices, with similar appearance to Seaborn in Python.
- View on Github, MathWorks File Exchange
Simple Cohen’s Kappa
- A simple and easy-to-use implementation of the inter-rater reliability measure Cohen’s kappa (κ)
- Cohen’s kappa is normalized for the percent agreement expected by chance due to class imbalance
- View on Github, MathWorks File Exchange
fast_corr
- Rapidly calculates column-wise Pearson correlations between large matrices using a vectorized method
- An order of magnitude faster than Matlab built-in corr.m to compute the full correlation matrix
- View on MathWorks File Exchange
Intrinsic Connectivity Contrast (ICC)
- A simple utility for efficiently computing the ICC, a measure of degree centrality for weighted networks
- This function was inspired by Martuzzi et al. (2011) and used in some of my neuroimaging research (Layden et al., 2017)
- View on MathWorks File Exchange
References
Layden, E. A., Cacioppo, J. T., Cacioppo, S., Cappa, S. F., Dodich, A., Falini, A., & Canessa, N. (2017). Perceived social isolation is associated with altered functional connectivity in neural networks associated with tonic alertness and executive control. Neuroimage, 145, 58-73.
Martuzzi, R., Ramani, R., Qiu, M., Shen, X., Papademetris, X., & Constable, R. T. (2011). A whole-brain voxel based measure of intrinsic connectivity contrast reveals local changes in tissue connectivity with anesthetic without a priori assumptions on thresholds or regions of interest. Neuroimage, 58(4), 1044-1050.