The Problem
Clinical evidence is scattered across hundreds of documents.
A clinician consulting on a complex case needs to cross-reference NICE guidelines, the BNF, relevant published research, and the patient's own history — simultaneously. This typically involves opening multiple browser tabs and searching each source manually.
AI LexNotes collapses this into a single interface. Upload all your reference documents once. Ask clinical questions in natural language. The AI searches across all sources simultaneously and surfaces the relevant passages with citations.
Example Questions
Workspace capabilities
Upload Any Clinical Document
PDFs, text files, scanned images — guidelines, papers, protocols, formulary references
Natural Language Q&A
Ask questions in plain clinical English — no query syntax or keyword search required
Source-Cited Answers
Every answer references the specific document and section it was derived from
Patient History Import
Import any ELLEXMED patient visit as a notebook source for case analysis
Private & Encrypted
Organization-scoped. Not used for AI training. AES-256 at rest, TLS 1.3 in transit.
Persistent Notebooks
Notebooks persist across sessions — build a personal clinical knowledge library over time
FAQ
AI LexNotes — answered.
What file formats does AI LexNotes support?
AI LexNotes supports PDF, plain text (.txt), and image files (JPG, PNG — for scanned documents). The most common use case is uploading clinical guidelines (NICE, BTS, WHO, AHA), research papers from PubMed, drug reference documents, or custom clinical protocols.
How does the AI answer questions about uploaded documents?
LexNotes uses Retrieval-Augmented Generation (RAG). When you ask a question, the system searches the indexed content of your uploaded documents for the most relevant passages, then feeds those passages into the AI model (Gemini 2.5 Pro) as context for generating an answer. The answer includes references to the source document and section, so you can verify the original text.
Can I import patient history from ELLEXMED into a notebook?
Yes. Any ELLEXMED patient visit — including the transcript and structured EHR data — can be exported as a notebook source. This is useful for complex case review, preparing referral letters, or asking analytical questions like 'Has this patient's blood pressure control improved over the last 6 visits?'
Is my notebook content used to train AI models?
No. Notebook content is private and organization-scoped. ELLEXMED does not use patient data, clinical documents, or notebook content for AI model training. Documents are stored encrypted at rest (AES-256) and only accessible within your organization's data boundary.
Can multiple clinicians share a notebook?
Notebooks are owned by the clinician who creates them. The current implementation does not support real-time collaborative editing, but notebooks can be viewed by other members of the same organization with appropriate permissions. Each clinician typically maintains their own personal research notebooks.
What are the most common clinical uses for AI LexNotes?
Common use cases include: (1) Uploading NICE/BTS/WHO guidelines and asking condition-specific questions without reading the full document; (2) Uploading a BNF or formulary reference and querying interaction or contraindication information; (3) Uploading published research papers and asking for summaries, methodology critique, or applicability to a specific patient population; (4) Importing a patient's full visit history to track clinical trajectory.