What You Ought to Know. Heidi, a global AI care partner platform formerly known as Heidi Health, has launched Heidi II. This upgrade transforms the platform from a passive system for clinical documentation to an active one that handles administrative tasks related to patient visits.
The platform addresses large-scale clinical needs, supporting 2.8 million patient interactions weekly across 110 languages and 190 countries. Its core engine has supported over 175 million visits, 67 million clinical hours, and generated more than 400 million follow-up tasks since February 2024.
Heidi II fills the gap in task execution by introducing native capabilities such as Agents, Memory, and Peer-Reviewed Research, which allow tasks to be completed instead of being handed back to the physician. Heidi II changes the interaction model from static prompting to an integrated clinical workspace.
Context-Aware “Home” Dashboard: Provides a customized view for the clinician’s daily schedule, automatically displaying patient records, upcoming appointments, and tasks that need attention upon login. Healthcare professionals can link previous appointments, test results, or standard guidelines with Heidi’s tasks.
Plain-Language Autonomous Agents: Doctors tell Heidi to carry out complicated office tasks using simple, everyday language. The primary agent devises a strategy, communicates with subordinate training resources, and delivers finalized versions for medical professional endorsement without further inquiries in between.
Initial applications involve creating daily clinic schedules in advance, conducting audits based on evidence for referral letters, and identifying overdue screenings, tests, and vaccinations through patient panel sweeps. Users can save sequential actions as ‘Routines’ for future use by clinicians or practice managers. ” They operate on pre-set schedules for whole patient groups instead of processing patients individually… Adaptive ‘Memory’: The AI learns and adapts to each clinician’s personal preferences, formatting rules, and templating styles over time, reducing the need for repetitive corrections… Inter-System Computing: It moves smoothly across four to six different software systems commonly used in outpatient practices, such as calendars, EHRs, patient portals, and email inboxes, to extract data and compose messages.
