Epidemics (Sentinelles) smart home integration, free and open source

Respiratory infections, gastroenteritis and chickenpox in your French region (Sentinelles network).
Acute respiratory infections, acute diarrhoea (gastroenteritis) and chickenpox in your French region, as the Sentinelles network (Inserm, Sorbonne Université) measures them every week with general practitioners: one device per region, with the weekly rate of each disease and its trend. Scene triggers, a scene action and a dashboard widget come with it.
Epidemiological information, not medical advice. These are estimates for a whole region, with wide confidence intervals. For your health, ask your doctor.
Not affiliated with the Sentinelles network.
What you get
- One device per area you choose: the 13 regions of metropolitan France, and metropolitan France as a whole.
- For each disease, the rate of the latest week (cases seen by general practitioners, per 100,000 inhabitants) and its trend against the week before (rising, stable, falling).
- The week of the data ("43 (19 – 25 Oct 2026)").
- Two scene triggers (a rate rises above or falls below a threshold), a scene action (the week in one sentence) and a dashboard widget.
Requirements
- Gladys Assistant 5.3.0 or later.
- Internet access from Gladys. No account and no key.
- Metropolitan France only: the network does not cover the overseas regions.
Configuration
Settings
| Setting | Default | What it does |
|---|---|---|
| Areas | none | The regions to follow (and metropolitan France), one device each. |
| Language of the feature names | French | Names of new devices and features, language of their texts and summary. |
Save, then open the Discover tab of the integration and create the devices you want.
Devices
| Feature | Type | Value |
|---|---|---|
| Respiratory infections /100,000 | number | Acute respiratory infections (flu, COVID-19, RSV…) |
| Respiratory infections, trend | text | rising, stable or falling |
| Acute diarrhoea /100,000 | number | Acute diarrhoea (gastroenteritis) |
| Acute diarrhoea, trend | text | rising, stable or falling |
| Chickenpox /100,000 | number | Chickenpox |
| Chickenpox, trend | text | rising, stable or falling |
| Week of the data | text | ISO week number and its days |
The trend is "rising" or "falling" when the rate moved by 15 % or more of the week before and by 5 per 100,000 or more; "stable" otherwise. The rates are numbers, so a scene condition such as "respiratory infections ≥ 200" works.
Dashboard widget
Seasonal epidemics: the area chosen (else the first one followed), with one line per disease giving its rate and trend, coloured orange when rising and green when falling, and a caption with the week and the source. One setting chooses the diseases shown, another hides the caption.
Scene triggers
- Epidemic rate rises above a threshold: when a new week is published, the rate of a disease in an area crosses the chosen threshold (25 to 1,000 per 100,000) upwards. Filters: area, disease, threshold.
- Epidemic rate falls below a threshold: the same, downwards: the wave is over in your area.
Variables: area, disease, rate, previous rate, week. The first week read after the installation only sets the reference: it fires nothing. The previous rate is the one the device showed: a revision of past weeks by Sentinelles never fires a crossing twice.
Scene action
Epidemics: get the week in one sentence: returns, for an area (empty: the first one followed), a sentence ready for a notification, such as "Occitanie, semaine 43 (cas pour 100 000 habitants) : Infections respiratoires 196 (en hausse), Diarrhée aiguë 58 (en hausse), Varicelle 4 (stable).", in the language of the configuration, with the week number and each rate as separate outputs.
Scene examples
- Every Thursday at 8 am, send the sentence of your region to the family group.
- When acute diarrhoea rises above 200 in your region, remind everyone to wash their hands.
- When respiratory infections fall back below 100, say the winter wave is over.
Data sources, licence and quotas
- Source: réseau Sentinelles, INSERM, Sorbonne Université, www.sentiweb.fr, public Sentiweb API, read without an account. The data are estimates from the network's general practitioners (and, for the regions, wide confidence intervals).
- Updated once a week, usually on Wednesday afternoon. The API allows 15 requests per 10 minutes and asks its users to download only what changed: the integration asks for the version of the data once an hour, and downloads the datasets (3 per level, regions or France) only when it changed. After a "too many requests" answer it waits for the time the API announces, restarts included.
- Influenza-like illness, the historical flu indicator, is no longer under surveillance: flu is counted in the acute respiratory infections.
Troubleshooting
- No value after creating the device: wait a few seconds for the first download; check that Gladys has internet access.
- The connection status says the Sentinelles data cannot be read: the API is unreachable; the last values are kept and the integration tries again every hour.
- The connection status says the API asks to wait: its request limit was reached (perhaps by another device of your network). The integration waits and tries again on its own.
Limits
- Developed and tested against the real API, not inside a running Gladys yet: feedback welcome.
- The regional rates are estimates with wide confidence intervals: a small change is not significant, hence the margin of the trend.
- No official "epidemic level": Sentinelles publishes rates, and Santé publique France publishes no open regional epidemic level that an integration could read. The thresholds of the triggers are yours to choose.
Configuration settings
These are the settings Epidemics (Sentinelles) asks for in its configuration screen in Gladys.
| Setting | Type | Required | Description |
|---|---|---|---|
| Your areas | section | No | Choose the regions to follow, and metropolitan France if you wish: one device per area, with the weekly rate of acute respiratory infections, acute diarrhoea and chickenpox, and their trend. Save, then create the devices in the Discover tab. |
| Sources and limits | section | No | Data of the Sentinelles network (Inserm, Sorbonne Université), read without an account, updated once a week (Wednesday). Metropolitan France only. Epidemiological information, not medical advice. Not affiliated with the Sentinelles network. |
| Areas | multi_select | No | One device per area chosen. |
| Language of the feature names | select | No | Names given to the new devices and their features, and language of their texts (trends, weeks, scene variables and summary). |
How to install Epidemics (Sentinelles) in Gladys
- In Gladys, open Integrations: Epidemics (Sentinelles) appears in the catalog, next to the native integrations, with a community badge.
- Click Install. Gladys pulls the Docker image (
ghcr.io/guim31/gladys-epidemies:1.0.0), starts it in a sandbox isolated from the core, and generates the integration's interface (devices, discovery and configuration). - Open the Configuration screen of the integration, fill in the settings, and save.
- You can also install it directly from its repository URL: https://github.com/guim31/gladys-epidemies.
Epidemics (Sentinelles) requires Gladys >=5.3.0. The catalog inside Gladys refreshes every hour, so a new version becomes available at most one hour after its release.
Not running Gladys yet? It is free and open source: follow the installation guide to get started.
About external integrations
Epidemics (Sentinelles) is an external integration: a community integration packaged as a Docker container and published on GitHub, that Gladys installs in one click and runs in a sandbox isolated from its core. It is published and maintained by guim31, not by the Gladys core team.
- Browse all integrations, native and community
- Build and publish your own external integration
- Source code on GitHub — source of this documentation