Violet Review Clinical document summarization

Violet Review summarizes long charts, referrals, treatment histories, and supporting documents to speed up operational review — condensing lengthy clinical records into concise, actionable summaries for staff who need to make fast, informed decisions.

Key capabilities:

• Ingests and parses lengthy clinical documents (charts, referrals, treatment histories, discharge summaries, supporting attachments)
• Generates structured summaries highlighting key clinical events, diagnoses, medications, and treatment timelines
• Surfaces relevant context for the specific review task (e.g., referral intake, prior-auth review, care coordination)
• Reduces manual chart review time for clinical and administrative staff
• Supports multi-document synthesis — pulling a coherent summary across fragmented records from multiple sources

Ideal for: Clinics, referral coordination teams, utilization review staff, and care managers who need to quickly understand a patient's history without reading through hundreds of pages of records.

Pricing: $50,000

Violet Review Clinical document summarization engine

Technical capabilities:

• Document ingestion: Accepts multi-format clinical documents (PDF, scanned images with OCR, HL7/CCD, FHIR DocumentReference, plain text) from EHR exports, fax gateways, and referral portals
• Document parsing: Section-aware extraction (chief complaint, history, medications, labs, assessment/plan) using layout-aware parsing for scanned/unstructured documents
• Multi-document synthesis: Aggregates and deduplicates overlapping information across multiple source documents into a unified timeline (e.g., resolving conflicting medication lists across three referral letters)
• Summarization engine: LLM-based summarization with task-specific prompting (referral intake summary vs. prior-auth summary vs. care-coordination summary), grounded in extracted structured fields to reduce hallucination risk
• Citation/traceability: Each summary claim links back to its source document and page/section for clinician verification
• Human-in-the-loop: Reviewer interface for editing/approving summaries before they enter the patient record, with corrections feeding back into prompt/model tuning
• Integration: Reads from EHR document repositories via API, writes summaries back as structured notes or referral packets

Deployment considerations: Designed as a decision-support and workflow-acceleration tool — summaries are intended for clinical staff review, not as a substitute for reading source records in high-stakes cases.

Here's the architecture:Data schema:

{
  "summary_id": "uuid",
  "patient_id": "uuid",
  "requested_at": "ISO-8601 timestamp",
  "task_type": "referral_intake | prior_auth | care_coordination | general_review",
  "source_documents": [
    { "doc_id": "uuid", "type": "chart | referral | discharge_summary | lab_report", "received_at": "ISO-8601 timestamp", "page_count": "integer" }
  ],
  "extracted_fields": {
    "diagnoses": [{ "value": "string", "source_doc_id": "uuid" }],
    "medications": [{ "value": "string", "dose": "string", "source_doc_id": "uuid" }],
    "treatment_timeline": [{ "event": "string", "date": "ISO-8601 date", "source_doc_id": "uuid" }]
  },
  "summary": {
    "text": "string",
    "model_version": "string",
    "generated_at": "ISO-8601 timestamp"
  },
  "citations": [
    { "claim": "string", "source_doc_id": "uuid", "location": "page/section reference" }
  ],
  "review": {
    "approved": "boolean",
    "reviewer_id": "uuid | null",
    "edits_made": "boolean",
    "reviewed_at": "ISO-8601 timestamp | null"
  },
  "audit_trail": [
    { "event": "string", "timestamp": "ISO-8601 timestamp", "actor": "system | user_id" }
  ]
}

Price breakdown ($50,000):

Same caveat as before: this is a first-year all-in price; renewals typically run license + support only (~$28–32K). Want this and Azalea Triage combined into a single formal proposal document (Word or PDF) for distribution?