read-tv

read-tv visualizes longitudinal time-stamped tabular data and detects changepoints to support analysis of temporal patterns such as surgical workflow disruptions.

Key Features:

• Changepoint Analysis (CPA) and Filtering: Performs changepoint analysis (CPA) and applies user-specified filters to identify temporal intervals with altered event rates.
• Customizable Longitudinal Plots and Faceting: Generates customizable longitudinal plots with faceting for stratified visualization of time-series data.
• Reproducible R Code Export: Produces the R source code for each visualization to enable reproduction and verification of plots.
• Simple Data Input Requirements: Accepts tabular input with a time column containing no missing values provided as a file or an in-memory dataframe.
• Generalizability and Extensibility: Handles any tabular dataset meeting input requirements and can be extended by adding functionality in the source code.
• Detection and Characterization of Disruption Periods: Enables identification and characterization of periods with high disruption rates, including disruption types such as training and equipment issues linked to adverse surgical outcomes.

Scientific Applications:

• Surgical Workflow Disruption Analysis: Identifies and characterizes periods of high disruption rates in operating room workflows, including types like training and equipment issues, to relate these patterns to adverse surgical outcomes.
• Exploratory Longitudinal and Time-series Research: Supports visualization and pattern detection in time-series or longitudinal datasets across diverse research contexts.

Methodology:

Uses customizable longitudinal plotting, faceting, changepoint analysis (CPA), user-specified filters, and exports R source code for each visualization.