Claudician
I built Clinical Digital Twin, an open-source R system (Shiny/plotly dashboard, plumber API, SQLite, and a Telegram bot powered by Claude) that gives clinicians in elder-care institutions a personalized, interpretable mo
Core trick
Answers what-if questions by re-running a pooled logistic fall-risk model with changed inputs, and the chat bot quotes only those computed outputs.
Description
I built Clinical Digital Twin, an open-source R system (Shiny/plotly dashboard, plumber API, SQLite, and a Telegram bot powered by Claude) that gives clinicians in elder-care institutions a personalized, interpretable model of each resident's fall risk. An engineered pooled-logistic-regression digital twin ingests simulated wearable data (heart rate, blood pressure, steps, sedentary time) and clinical history to predict fall probability at 24 hours and 7 days, and answers counterfactual "what-if" questions in real time (e.g., what happens to risk if mobility increases or a medication is deprescribed).
I found that a single, transparent model can serve two very different interfaces (a visual dashboard and a natural-language Telegram bot) without sacrificing interpretability, and that grounding Claude's replies in real model output rather than free generation keeps the bot clinically trustworthy. We also surfaced a modeling caveat worth reporting: under our synthetic data-generating process, static risk factors are only identifiable through their effect on sensor trends, not directly; a reminder that interpretability claims need statistical scrutiny, not just a simple model.
This matters because falls are a leading cause of injury death in older adults, and institutional fall-prevention often fails not from data scarcity but from data never reaching a clinician in time.
README excerpt
# Claudician: Clinical Digital Twin Monitoring System An end-to-end R system that lets clinicians visualize patient vitals/activity data and run **"what-if" simulations on statistical digital twins** to estimate fall risk over the next 24 hours and 7 days. > ## ⚠️ Disclaimer > **All patient data in this project is SYNTHETIC.** It is generated by > `data-raw/generate_synthetic_data.R` using random distributions and contains > **no real PHI**. This is a **hackathon prototype and is NOT for clinical use**, > diagnosis, or treatment decisions. The auth layer is an MVP simplification, not > production-grade security. See [Known limitations](#known-limitations). ## Why falls, and why this matters Falls are the single most important safety event clinicians in older-adult institutions are asked to watch for. In this population a fall is rarely "just a fall": it is a potential inflection point in a resident's health trajectory. A single fall can cause a hip fracture or a traumatic brain injury, trigger a hospital admission, and set off a downward spiral of immobilization, loss of independence, fear of falling, functional decline, and death. Because frail older adults have little physiological reserve, the adverse events that follow a fall are frequently **devastating and irreversible** — which is exactly why early, continuous fall-risk monitoring is worth the effort. The burden at the population scale is enormous: - **One in four** U.S. adults aged 65+ — over **14 million** peop