THE UTILISATION OF AN ARTIFICIAL INTELLIGENCE BASED CONVERSATION AGENT IN CATARACT SURGERY FOLLOW-UP: A SIX-MONTH ANALYSIS OF PATIENT OUTCOMES
Published 2025 - 29th ESCRS Winter Meeting
Reference: PO072 | Type: Presented Poster & Poster | DOI: 10.82333/sm0k-em62
Authors: Vinay Badhwar* 1 , Raneesha Pillay 2 , Aaron Sosah 2 , Aisling Higham 3 , Thomas Poole 2
1Frimley Park Hospital,Camberley,United Kingdom;University Hospitals Sussex NHS Foundation Trust,Sussex,United Kingdom, 2Frimley Park Hospital,Camberley,United Kingdom, 3Moorfields Eye Hospital,London,United Kingdom
Purpose
The increasing incidence of cataracts and demand on healthcare systems highlight the need for AI in post-operative care. This study evaluates the safety and effectiveness of Dora, an AI-based conversational agent, in follow-up care for cataract surgery patients.
Setting
All patients who had routine cataract surgery from the 13th of December 2022 to the 28th February 2023 at Frimley Park Hospital, UK and who could converse in English.
Methods
Dora calls were conducted to discuss symptoms and answer questions, simulating clinician interaction. Patients who “passed” were scheduled for a second surgery or discharged, while flagged concerns prompted a follow-up call. Patient notes were reviewed up to six months post-surgery to identify any unexpected management changes after the Dora call.
Results
458 patients were listed for a Dora call (mean age 75 SD 8.4). Out of these 380 were completed, resulting in an 83% completion rate. Of those completed, 245 patients passed while 135 were flagged for potential concerns. Visual disturbances, particularly vision concerns (46 patients) and new floaters (40 patients), were the most common post-op issues flagged via Dora, while significant redness, pain, and flashing lights were also flagged but less frequently.
Conclusion
This analysis demonstrates Dora’s potential to reduce workforce burden while supporting effective post-op care, especially for high-volume procedures like cataracts.