Document Filing

The AI uses optical character recognition (OCR) to extract text from documents, then applies machine learning models trained on clinical documents to identify document type, extract patient identifiers, and highlight key clinical information. It suggests a patient match and categorisation that a clinician then reviews and approves.

Jump handles common clinical document types: lab results, specialist letters, discharge summaries, radiology reports, GP correspondence, insurance forms, and patient-submitted records. The system learns from your practice's document patterns, improving accuracy over time for the specific document types you receive.

NHS documents received via NHS Mail or other channels are processed through the same workflow. The system recognises NHS document formats and extracts NHS numbers and other identifiers for patient matching. Filed NHS documents are stored securely with appropriate access controls.

Yes, always. The AI suggests patient matches and document categorisation, but filing requires clinician approval. You can correct the patient match, change the category, add notes, or route the document for another team member's attention. High-confidence matches can optionally be auto-filed, but this is configurable.

Documents without confident patient matches are queued for manual review. Staff can search for the correct patient, create a new patient record if appropriate, or mark the document as not for filing (e.g., administrative correspondence). Nothing is discarded—all received documents are tracked until appropriately handled.

All documents are encrypted at rest and in transit. Access is controlled by the same permission system as patient records—users see only documents for patients they are authorised to access. The audit trail records all document access. For sensitive documents, additional access restrictions can be applied.