In another big move aimed at its healthcare clients, Amazon Web Services revealed this week that its Textract machine learning technology – which can help healthcare organizations more easily extract data from scanned documents – is now HIPAA eligible, joining a half-dozen other cloud-based AI tools.

WHY IT MATTERS
It's important to note that HIPAA-eligible is not the same as HIPAA-compliant – it just means that the technology is able to be customized and put to use in ways that are. Organizations can't simply install Textract and expect to be compliant.

That said, with proper configurations, the tool can be deployed at healthcare and life science organizations whose tasks that require HIPAA compliance, said Textract Product Lead Kriti Bharti in an Oct. 10 blog post.

"Critical healthcare information often lies within documents such as medical records and forms. Healthcare and life science organizations need to access data that is locked inside those documents in order to fulfil medical claims, streamline administrative processes, and process electronic health records," said Bharti.

"Amazon Textract analyzes virtually any type of document – such as patient information from an insurance claim or values from a table in a scanned medical chart – without requiring customization or human intervention," she added.

Its HIPAA-eligibility, said Bharti, means that "healthcare and life science customers can take full advantage of it." She noted that healthcare and life sciences customers like The American Heart Association, Celgene and Cerner – which recently named Amazon its preferred cloud hosting provider and just announced a new cognitive platform built using AWS infrastructure – are among the organizations exploring how the machine learning tool can automate their workloads

Textract joins six other AI tools that are HIPAA-eligible: Amazon Translate (language translation), Amazon Comprehend (extracts insights from unstructured text), Amazon Transcribe (speech to text conversion), Amazon Polly (text into speech), Amazon SageMaker (managed machine learning service) and Amazon Rekognition (image and video analysis).

THE LARGER TREND
In the blog post, Amazon points to several healthcare companies that are already putting Textract to work.

At Cambia Health Solutions, parent company of six regional health plans, is using it to  optimize care coordination and streamline administrative processes, said Faraz Shafiq, its chief artifical intelligence officer. And at Change Healthcare, Textract is helping it "liberate the information from millions of documents and create even more value for patients, payers and providers,” said Nick Giannasi, the company's chief artificial intelligence officer.

ON THE RECORD
Bharti notes that many AWS customers typically rely on "manual data entry simple optical character recognition software. This is a time-consuming and often inaccurate rocess that requires extensive pre-processing, such as creating custom templates for each unique document type, and extensive post-processing from human reviewers.

"What we’ve learned from our customers, is they instead want the ability to quickly, easily, and accurately retrieve text and data from forms and tables in a wide variety of documents," she said.