Freed AI uses advanced natural language processing (NLP), sophisticated summarization, and sensible data organization to write medical notes based on conversations captured between clinicians and their patients. New users are blown away by the accuracy and comprehensiveness of the AI-generated notes.
However, after the newness wears off, users begin to realize that while the notes are impressive, they need to be edited to varying degrees to fit their personal style and format, as well as any style requirements their medical practice may have.
Our overarching mission at Freed is to save clinician's time wherever possible, so we asked ourselves, "how can we use AI to cut down on the time our users spend editing their notes?"
1. Give new users the opportunity to upload past notes that fit their style requirements. This is similar practice to when clinicians begin working with human scribes.
2. When a note is uploaded, identifiable information is anonymized so the user can see that only the style, structure, and anonymized content is being stored for use.
3. Users can upload up to three past notes, but Freed can do a lot with just one.
4. Later in the journey, users can capture examples notes generated by Freed as well as curate their saved examples by clicking "Teach Freed" in the bottom toolbar.
Magic Edit lets users instruct Freed to make changes to the entire note or a specific selection. These instructions can also be saved and applied to future notes.
15% reduction in churn in two months
Drop in churn correlated with drop in complaints about lack of personalization options
22% reduction time spent editing in two months
Users clicked Copy or Send to EHR faster after their note was generated
Sole Product Designer, Researcher, and Content Strategist
Assumed PM duties until a dedicated PM was hired midway through the project
