KétaPK - Pharmacokinetic Simulation for Anesthesia & Intensive Care

KétaPK is a modern, cross-platform pharmacokinetic (PK) and pharmacodynamic (PD) modeling application for Esketamine and Racemic Ketamine. It allows anesthesiologists, intensive care physicians, and clinical researchers to simulate, visualize, and predict plasma drug concentrations (C_p) over a 300-minute timeline using multi-exponential models.

1. Overview & Purpose

Original Creator: Dr. Georges Mion (Anesthesiologist-Intensivist)
Purpose: To provide clinical specialists with an instantaneous, offline, and reliable decision-support and simulation tool for IV bolus doses and continuous infusions.
Core PK Models Integrated:
Domino et al. (1982) — Three-compartment model (2.0\text{ mg/kg} reference)
Clements et al. (1981) — Two-compartment model (0.25\text{ mg/kg} reference)
Kamp et al. (2020) — Meta-analytic multi-compartment model (0.5\text{ mg/kg} reference)

2. Project Origin & AI Generation Transparency

The Legacy Application

The initial version of KétaPK was hand-crafted as a Microsoft Excel workbook (.xlsm) using complex VBA (Visual Basic for Applications) macros, custom chart event listeners, and dynamic range colorizations.

Modernization via AI Collaboration

To make the tool responsive, cross-platform, and independent of Microsoft Office, the application was re-engineered into a modern web and desktop software stack.

AI Collaboration Model: Google Gemini
AI Model Version: Gemini 3.6 Thinking
Generation Mode: Context-aware multi-turn prompt engineering & source-code decompilation
Generation Parameters: Standard System Instructions, Default Temperature (~0.7), Strict Module Separation (Decoupled Math Engine / Responsive UI / i18n Dictionary)

3. Legacy Excel Implementation

The original spreadsheet relied on embedded OLE streams, sheet-level event macros (Worksheet_SelectionChange), and manual range formulas to calculate 61 time points (0 to 300 minutes in 5-minute increments).

Learn more about the original implementation: View Original Excel/VBA Application Documentation & Walkthrough

4. Running with Docker

You can run KétaPK in an isolated container using Docker or Docker Compose.

Option A: Using Docker CLI

# Build the Docker image using the non-root secure Dockerfile
docker build -t ketapk-web .

# Run the container mapping port 3000 on your host machine
docker run -d -p 3000:3000 --name ketapk-app ketapk-web

# Access the application in your browser at http://localhost:3000

Option B: Using Docker Compose (Recommended)

# Build and start the containerized application in detached mode
docker compose up -d --build

# Stop and remove the running container environment
docker compose down

5. Compiling Desktop Apps with Tauri (Windows, macOS, Linux)

KétaPK uses Tauri v2 to generate ultra-lightweight (\sim 12\text{ MB}), native desktop executables (.exe, .dmg, .AppImage) that operate 100% offline without requiring Node.js or Docker on the target medical machine.

Prerequisites

Node.js (v20 or higher)
Rust & Cargo (installed via rustup.rs)
Windows: C++ Build Tools for Visual Studio
macOS: Xcode Command Line Tools (xcode-select --install)

Step-by-Step Build Process

# 1. Install project dependencies
npm install

# 2. Compile the static Tailwind CSS bundle
npm run build:css

# 3. Launch the application in desktop development mode (with Hot Reload)
npx tauri dev

# 4. Compile the production installer for your current Operating System
npx tauri build

Output Installation Binaries

After running npx tauri build, your standalone installer will be available at:

Windows: src-tauri/target/release/bundle/nsis/KetaPK_1.0.0_x64-setup.exe
macOS: src-tauri/target/release/bundle/dmg/KetaPK_1.0.0_x64.dmg
Linux: src-tauri/target/release/bundle/appimage/KetaPK_1.0.0_amd64.AppImage

4.2 KiB Raw Blame History