Intelligent document processing for the insurance industry

on YouTube ↗30m

" summary

Anthem replaces a 20‑minute manual review per claim by piping scanned claims and attachments through Amazon Textract for OCR and structure extraction, running a small ML indexer, and sending low‑confidence fields to Amazon Augmented AI for human review. A pilot processes about 80% of attachments end‑to‑end automatically, aiming for >90%, and AWS cites about 10x savings on OCR workloads; outputs are JSON, CSV (tables), and key/value files for downstream systems.

AI-written summary · about

" chapters (13)

1. 0:00Introduction
2. 6:42Problem Statement
3. 13:24Agenda
4. 15:23Demo
5. 17:54CLI
6. 19:31Code Sample
7. 21:30Sample Document
8. 21:52Sample Document Extraction
9. 25:21JSON Parser
10. 26:58Augmented AI
11. 28:14Resources
12. 29:14Thank you
13. 29:30Session Survey

" related talks

• 50mWhy use AI in insurance underwriting and claims? (PRT271)Kenneth Saldanha, Anand Premsundar · AWS re:Invent 2022
• 42mExtract information accurately from documents using computer visionNavneeth Nair, Martin Schade · Amazon re:MARS 2022
• 35mAutomate your document pipeline with intelligent document processingPrem Ranga · AWS Machine Learning Summit 2021
• 32mAutomate data extraction & analysis with intelligent document processingAshish Khare, Saumin Patel · AWS AMER Summit May 2021
• 26mEmpower your business use casesMatteo Gabrielli · Conf42 Machine Learning 2021
• 13mTransforming Insurance Through AIBhaskara Beeraka · Conf42 Cloud Native 2025
• 32mAutomate data extraction & analysis with intelligent document processingAshish Khare, Saumin Patel · AWS Summit Online – Americas 2021
• 01h 00mExtract information using the Queries feature in Amazon TextractJulu Panat · AWS Summit San Francisco 2022