form on a top of a colored shape background.


How to automate order form processing for medical devices using OCR and Impira AutoML

Order processing and fulfillment workflows are everywhere, but many industries are struggling with rapid digital transformation and still process orders by paper or fax. Now's the time to transform your order management process by using optical character recognition (OCR) and automated machine learning (AutoML) from Impira.

Download sample files

We all want more sales, but more paperwork?

Despite an increasing consumer demand for more convenient experiences, healthcare and insurance are two example industries that remain heavily reliant on paper workflows. According to a recent survey, a whopping 90% of healthcare providers still use paper. Eliminating that much paper is quite the undertaking, let alone the hundreds of thousands of hours of manual data entry that's highly error prone and slow to process.

Paper is everywhere. From insurance claims to pharmacy prescriptions and medical equipment order forms, there is no shortage of paper-based workflows that can benefit from automation. The burden of manual data entry is real.

Better for your business, better for your customers

Better than traditional OCR technology, Impira AutoML reduces the amount of time and effort required for order management for any order processing or fulfillment teams. Impira gives you instant access to high quality data, allowing your teams to more efficiently process an increased volume of complex, high priority orders, with much greater accuracy. The result is that order forms can be addressed with greater care. This drives better customer experience, greater retention, and increased revenue.

Impira AutoML can be applied to a wide variety of use cases across the enterprise.

Traditional OCR solutions don't work

Many companies have turned to traditional OCR solutions in an effort to diminish the time-consuming workloads involved in manually entering in details from purchase orders. When legacy OCR technology is not supercharged by artificial intelligence (AI), it doesn’t learn from mistakes, is extremely limited when deployed in an enterprise, and doesn’t provide the structured output that businesses require to successfully convert a piece of paper into meaningful information that you can pass down to order fulfillment and accounts receivable.

Automated order entry you can control

Impira AutoML technology combines OCR with artificial intelligence and machine learning to fully automate this effort, enabling smarter access to information in your documents. Impira continuously learns from your interactions to improve results, provide instant access to better quality data, and ultimately get your workflows moving at lightning speed. Whether it be a document, a scan, or a photograph, Impira supports them all, affording you complete flexibility in how you receive your order forms. And our simple APIs can be used to integrate with your order and fulfillment processes, allowing you to support a drastically higher volume of orders and increase your customer satisfaction. 

Want to see it in action? Follow along with your own custom order forms, or you can download some sample Durable Medical Equipment Order Forms from the link on the left.

Step 1: Create a collection

Once you've created a free Impira account and logged in, start by creating a new collection, which is like a folder that holds files with the same layouts and data fields. For example, if you have multiple order intake forms from which you want to extract names, phone numbers, and balances, group them in the same collection.

Create a collection by clicking on the + next to Collections in the left-hand sidebar and give it a name like "Order forms."

Step 2: Upload your order forms

Click to open your newly created collection. There are many ways to get your files into Impira. You can:


animated gif of how to upload files.


Once your file(s) are uploaded, they're immediately processed and placed in a spreadsheet.

Step 3: Training your ML models

Impira takes a unique approach to training which doesn’t require large numbers of training documents and allows you to train your model directly with just a handful of files. Start by double-clicking one of your files to begin extracting data.

Highlight the member's name on the document and name the field "Member's name."

animation on how to highlight and select data to extract.


Impira AutoML now processes all the files in your “Order forms” collection and updates a model based on this field. In just a few seconds you have created and trained an ML model, and Impira ran it across all of your forms to automatically fill in your spreadsheet — a process that usually takes hours when done manually.

You can create as many fields as you need — customer name, address, payments, shipping details, etc.

Review and confirm predictions

Now that Impira has extracted all the data fields you want, we have the opportunity to fine tune Impira by reviewing its work (referred to as "predictions").

Learn how to refine and improve Impira's ability to extract data from any new files in this collection by going through the review workflow.


Step 4: Getting your data out in the form you need 

Once you’ve uploaded your files and documents into a collection and worked with Impira to extract all the data you need, you’ll want to access the data in a format that is useful to you. 

Option 1: Export data via CSV

Click Download in the top-right corner of your browser window and select As CSV

You can now import your data to your spreadsheet application or any other application you typically use to work with CSV files. 

Option 2: Exporting via API (and other fun options)

You can also access your data through an API, and you have the option of utilizing Impira Query Language (IQL) to query specific data, giving you as much control and flexibility you need over your data. This document will also show you how to utilize a Poll API.

And if you're feeling a bit more adventurous, you can always connect a webhook with Impira.

Get started in minutes.

Already using Impira? Sign in.