Research, Knowledge Management & Clinical Content Creation

The Diagnostic Prompt IMI Revisited + Advanced Prompting — Chain-of-Thought, Few-Shot & Negative Techniques

Opens with a 15–20 minute formal revisit of The Diagnostic Prompt IMI, now applied to research questions: before you open any search tool, let AI interrogate the question. Live demonstration with a real clinical research question — the AI's questions reshape and sharpen it before a single search is run. Remainder of session covers advanced prompting techniques: chain-of-thought (step-by-step reasoning before conclusion), few-shot (embedding worked examples to teach AI your specialty's language), and negative prompting (explicit exclusions to eliminate predictable failure modes). Includes prompt debugging as a clinical skill and a live comparison of Claude, ChatGPT/GPT-4o, and Gemini on the same complex multi-part clinical query.

Reflection on Advanced Prompting — Reviewing AI Tool Use Practice

Participant applies The Diagnostic Prompt IMI to one real complex clinical or research question from their own practice — a question where standard prompting has previously given insufficient results. They then use advanced prompting techniques (chain-of-thought, few-shot, or negative) to investigate it across two platforms and compare outputs. Structured reflection on their own AI tool utilisation practice.

AI-Powered Deep Research & Knowledge Management — Gemini, Perplexity & NotebookLM

Structured online teaching session covering: the distinction between standard AI queries and deep research mode; Gemini Deep Research and Perplexity compared — what each does well and when to use both; the paired two-tool method applied live to the same structured clinical question; how to read AI-generated research outputs (hedging language, confidence signals, citation gaps); a three-step citation verification protocol (confirm existence, verify author accuracy, verify accuracy of the attributed finding); NotebookLM introduced as a personal clinical knowledge base — uploading guidelines, local protocols, and curated papers, querying in plain language, and generating audio overviews for commute or journal club preparation.

Reflection on Evidence Synthesis Practice & NotebookLM Knowledge Base Setup

Participant applies the paired two-tool research method (Gemini Deep Research + Perplexity) to one real clinical or research question, verifies a sample of citations from each output using the three-step method, and documents findings. Participant also sets up their first NotebookLM notebook with a minimum of three clinical guidelines, local protocols, or curated papers relevant to their practice, and queries it on a real clinical question. Structured reflection on evidence synthesis practice and knowledge management.

AI for Clinical Data & Presentations — Claude and Microsoft Copilot for Excel & PowerPoint

Structured online teaching session with live demonstration in both environments. Data analysis: uploading clinical audit data to Claude on a personal device — generating insight summaries, identifying patterns and outliers, drafting governance-ready findings; the equivalent workflow in Microsoft Copilot within Excel for hospital Microsoft 365 environments; honest comparison of both tools on the same audit dataset. Presentations: converting clinical evidence and audit findings into slide structures using Claude (personal device) and Copilot in PowerPoint (hospital environment); briefing AI for grand rounds, M&M, or governance committee presentations; a practical decision framework for choosing the right tool in the right environment.

Applied Data or Presentation Task — Self-Directed Learning

Participant completes one real professional task using the tools from Hour 5. Either: (a) AI-assisted audit data analysis — uploading real or de-identified audit data and producing a written summary suitable for department use, governance submission, or quality improvement reporting; or (b) AI-assisted clinical presentation — producing a structured presentation using Claude or Copilot. Participant documents which tool they used, what the AI produced, what they changed, and why.

AI Content Creation — Teaching Materials, Patient Education & Audio Overviews

Structured online teaching session covering: AI for clinical teaching content — generating structured teaching sessions from papers, producing MCQs across Bloom's taxonomy levels, generating case vignettes, learning objectives, and facilitator notes; NotebookLM audio overviews — live demonstration transforming a folder of clinical papers into a conversational AI-voiced podcast in under five minutes; patient education content creation — drafting handouts, post-procedure instructions, and chronic disease self-management guides at specified reading levels; quality verification framework for AI-generated patient-facing content (accuracy, reading level, cultural appropriateness, tone); landscape overview of further AI audio and video tools for independent exploration.

Reflection on Clinical Content Deployment — Measuring New Workflow Performance

Participant creates and uses one piece of AI-generated clinical content in a real professional context: a teaching resource used in a session, a patient education document given to a patient or family member, or a NotebookLM audio overview used to prepare for a clinical case or journal club. Participant reflects on its quality, clinical utility, and the workflow used to produce it — prospective evaluation of a newly deployed AI-assisted content creation workflow.