Skip to content

workflow-guides.mdx

Latest commit

 

History

History
141 lines (120 loc) · 4.64 KB

workflow-guides.mdx

File metadata and controls

141 lines (120 loc) · 4.64 KB
title Workflow Guides
description Step-by-step guides for common end-to-end workflows in the ClaimMind API.

Overview

The ClaimMind API orchestrates complex medical claim processing through a series of automated workflows. This section provides practical, step-by-step guides for common end-to-end scenarios, demonstrating how different API endpoints work together to achieve a complete business process.

These guides are designed to help you understand the flow of data and operations, enabling you to integrate ClaimMind workflows seamlessly into your applications.

Available Workflows

Here are the detailed workflow guides:

End-to-End Workflow: Claim Submission to Validation

This guide illustrates the typical workflow for submitting a new claim, processing its associated documents via OCR, processing ICD codes, and then triggering a validation process.

Prerequisites

  • An authenticated access_token.
  • A base64 encoded document (e.g., a scanned medical bill or insurance form) to be processed.
First, create a new claim in the system. This will provide you with a `claim_id` that you will use for subsequent workflow steps. ```bash curl -X POST "{api_base_url}/v1/claims/create" \ -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \ -H "Content-Type: application/json" \ -d '{ "claim_id": "a1b2c3d4-e5f6-7890-1234-567890abcdef", "claim_code": "NEW-CLAIM-001" }' ``` With the `claim_id`, you can now trigger the OCR workflow to process the claim's associated document for PII. Provide the document as a base64 encoded string. ```bash curl -X POST "{api_base_url}/v1/workflows/ocr/pii/create" \ -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \ -H "Content-Type: application/json" \ -d '{ "claim_id": "YOUR_CLAIM_ID", "base64": "BASE64_ENCODED_DOCUMENT_DATA" }' ``` After processing for PII, you can trigger the general OCR workflow to extract all information from the document. This process will generate a markdown representation of the document's content. ```bash curl -X POST "{api_base_url}/v1/workflows/ocr/create" \ -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \ -H "Content-Type: application/json" \ -d '{ "claim_id": "YOUR_CLAIM_ID", "base64_list": ["BASE64_ENCODED_DOCUMENT_DATA"] }' ``` After OCR processing (including the normal OCR that generates markdown), you can trigger the ICD processing workflow. The `ocr_result` in the payload should be the markdown content obtained from the "OCR Normal" step. ```bash curl -X POST "{api_base_url}/v1/workflows/icd/create" \ -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \ -H "Content-Type: application/json" \ -d '{ "claim_id": "YOUR_CLAIM_ID", "ocr_result": "MARKDOWN_CONTENT_FROM_OCR_NORMAL_STEP" }' ``` After processing ICD codes, you can trigger the INACBG processing workflow. This step requires the `claim_id`, the `icd_code` obtained from the previous ICD processing step, and the `ocr_result` (markdown content) from the "OCR Normal" step. ```bash curl -X POST "{api_base_url}/v1/workflows/cbg/create" \ -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \ -H "Content-Type: application/json" \ -d '{ "claim_id": "YOUR_CLAIM_ID", "icd_code": { "icd10_list": [ { "code": "F01", "description": "Dementia in Alzheimer''s disease" } ], "icd9_list": [] }, "ocr_result": "MARKDOWN_CONTENT_FROM_OCR_NORMAL_STEP" }' ``` Finally, trigger the claim validation workflow. This will check the claim against predefined rules, utilizing the ICD and INACBG results. ```bash curl -X POST "{api_base_url}/v1/workflows/validation/create" \ -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \ -H "Content-Type: application/json" \ -d '{ "input_data": { "claim_id": "YOUR_CLAIM_ID", "tenant_id": "YOUR_TENANT_ID", "claim_markdown": "Extracted markdown content from OCR (or manual input)", "patient_profile": { "icd10_list": [ { "code": "F01", "description": "Dementia in Alzheimer''s disease" } ], "icd9_list": [], "ina_cbg": "G-4-26-II" } } }' ```