{"id":120,"date":"2026-03-18T20:10:31","date_gmt":"2026-03-18T20:10:31","guid":{"rendered":"https:\/\/medlearn.imperial.ac.uk\/innovation\/?page_id=120"},"modified":"2026-04-09T13:55:52","modified_gmt":"2026-04-09T13:55:52","slug":"paces-scenario-generator","status":"publish","type":"page","link":"https:\/\/medlearn.imperial.ac.uk\/innovation\/projects\/paces-scenario-generator\/","title":{"rendered":"PACES Scenario Generator"},"content":{"rendered":"\n\n
\n \n \n \n \n \n <\/linearGradient>\n <\/defs>\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \u2726<\/text>\n \n PACES<\/text>\n Scenario<\/text>\n Generator<\/text>\n AI-assisted clinical scenario drafting for MRCP examiners<\/text>\n \n \n Clinical<\/text>\n \n Pilot<\/text>\n \n AI<\/text>\n <\/svg>\n <\/div>\n\n \n

A research pilot exploring whether AI-generated PACES scenario drafts — covering patient demographics, presenting complaint, personality and ICE — can meaningfully reduce the time burden on clinical examiners and scenario writers.<\/p>\n\n \n \n \n \n \n \n \n \n
Status<\/td>\n Pilot — active research<\/td>\n <\/tr>\n
Exam context<\/td>\n MRCP Part 2 PACES — Practical Assessment of Clinical Examination Skills<\/td>\n <\/tr>\n
Team<\/td>\n Adrian Cowell, Ehigie, Naomi E, Corey Briffa<\/td>\n <\/tr>\n
Tech stack<\/td>\n Vite, React, TypeScript, Netlify, Claude AI<\/td>\n <\/tr>\n
Live app<\/td>\n paces-generator-7b40ef.netlify.app<\/a><\/td>\n <\/tr>\n <\/tbody>\n <\/table>\n\n \n

The Challenge<\/h2>\n

PACES examinations require a substantial library of realistic simulated patient scenarios. Each scenario must specify a presenting complaint, relevant history, personality and affect, and — critically — the patient’s Ideas, Concerns, and Expectations<\/strong> (ICE): the structured framework that underpins the communication skills station.<\/p>\n

Writing these from scratch is time-consuming, and recruiting clinical examiners for a full scenario-writing session is logistically difficult. The question this pilot set out to explore: can AI-generated drafts serve as a credible starting point<\/strong>, reducing the editing burden to something that fits within a shorter, more practical working session?<\/p>\n\n \n

What the Tool Does<\/h2>\n

The generator takes a structured set of inputs and produces a complete scenario draft. Clinicians provide the clinical framing; the AI handles the narrative expansion.<\/p>\n\n

\n
\n

Clinical inputs<\/h3>\n