Agents, persistent memory, and a search of validated literature: Heidi wants to process the approximately 10 million follow-up tasks generated each week by her consultations.
Heidi, known for its ambient medical scribe, began the deployment of Heidi II on September 29, in English and French. The platform is no longer satisfied with writing the report of a consultation: it adds agents capable of taking charge of the administrative work that arises, such as managing addresses, preparing files before consultation, coordinating follow-up, population reviews and patient reminders. The doctor retains control over decisions and final validations.
Heidi now supports about 2.8 million consultations per week worldwide, which bring up nearly 10 million follow-up tasks. Until now, these tasks were the responsibility of the practitioner; Heidi II relies on three bricks — agents, memory and access to peer-reviewed research — to carry them out.
Bypass the lack of integration
Notably, the agents work even without a direct connection with the computerized patient record: a "computer use" agent then takes control of the keyboard and mouse to navigate through the software. These functions will not be offered in the United Kingdom or the European Union.
Read the full article on the original source
Go to the source →