PhenoPad is a note taking application that allows physicians to take free-form notes and capture standard phenotypic information via audio, photo and video.
- Free-form note-taking by handwriting or typing.
- Insert drawings, images to note and capture photos and videos.
- Real-time handwriting to text conversion.
- Real-time speech-to-text and speaker diarization.
- Medical term recognition on notes and ASR transcripts.
- Phenotypes suggestion based on transcripts and differential diagnosis on diseases.
- Edit and annotate EHR.
The app requires Microsoft Windows 10. You also need to download the Visual Studio IDE.
Clone the project to your device. In Visual Studio, open the project by File->Open->Project/Solution... or press Ctrl+Shift+O and select the project's solution (PhenoPad.sln). Use the green start button to build and run the app.
More detailed guide on how to use the app in Wiki.
You need to set up your own speech server to use the speech features. A demo server and the instructions to set it up is available at https://github.com/haochiz/PhenoPad-SpeechEngine.
Once you set up the server and have it running, click the gear button on the top bar and select 'Settings'. Select the Surface Microphone option. Type "your.server.ip:port" (e.g. 127.0.0.1:8888) in the "ASR server" field then click Change Server. You should see a notification indicating ASR server address has been changed.
Click the microphone icon in the bottom left corner to start/stop a speech session, click the converastion icon on top left to open the real-time conversation panel to see the results.
By default, the phenotype suggestion and disease prediction service which relies on PhenoTips is hosted on our own server. If you'd like to set up your own service, you can install the latest stable Phenotips branch at https://github.com/phenotips/phenotips/tree/phenotips-1.4.9. Then go to Application Settings (click the gear button on the top bar) and change the address for Suggestion, Differential and PhenoDetail to the address of your own server.
The following code are used to evaluate ASR and medical term recognition performance: