Every organization struggles with how to store, tag, and search for their documents. In a hospital corporation, the need is particularly critical. Hospital staff need to be able to quickly find the latest policies and procedures. Auditors need to be able to track who made what changes and when. Lawyers want to know which protocols were in place on a particular date. In this session you’ll learn a practical approach to putting a document management system in place that can help address these needs and reduce your exposure to legal, regulatory, and even human health risks.

Based on lessons learned during a real-world project, the session shows that getting your documents under control doesn’t have to be a multi-year, multi-million dollar effort. The session will outline how a hospital corporation in New England used a “start small and grow” approach to piloting and rolling out a document management solution across the corporation.

Topics covered include:

• Business benefits of getting your documents under control
• Document management concepts such as metadata, versioning, and workflow
• A high-level methodology for document management implementations
• Common pitfalls
• Leveraging the benefits of open source software

Most of an organization’s information is stored in unstructured medical records with cursive handwriting. By using a new technology that understands how the documents are used and what role they play, you can classify and extract more information from both handwritten and machine-printed documents. Courtney Rand will focus his discussion around the toughest challenges in optimizing forms processing and medical records management today—unstructured document processing and handwritten cursive recognition—and the new generation of technology available to solve these problems. This presentation will explain the state of current automatic classification technologies, how they interact with recognition technologies, and explain how to implement such a system with maximum functionality and results.

Rand will discuss Intelligent Word Recognition (IWR), a lies at the heart of his company’s recognition software products and is optimized for processing data on real-world documents that contain mostly free-form, hard-to-recognize handwriting. The best use of IWR is to eliminate a high percentage of the manual entry of handwritten data on documents that otherwise could be keyed only by humans. IWR differs from Intelligent Character Recognition (ICR) primarily in that conventional ICR technology recognizes data fundamentally at the character level, while IWR recognizes data at the word or “field” level. A2iA’s IWR engine is capable of extracting all types of field-based information from a form - either constrained (machine print, hand-printed capitals) or unconstrained (freeform hand print, cursive) from virtually any type of document.

The IWR process of converting a written word into computer-usable data occurs in a series of top-down stages. IWR provides a competitive advantage for operations that rely on the conversion of paperwork into computer-usable form, such as data entry departments of big government agencies, large financial institutions, and service bureaus. In many cases, IWR can automate data entry at a cost so low that it enables A2iA customers to successfully compete against offshore service bureaus located in countries such as India and China

The experiences of two distinctly different communities to bring their healthcare information online will be explored in these case studies. In one case, the protagonist is a major institution responsible for setting healthcare standards and distributing funding for innovative public health programs. This institution set out to establish a web portal that would provide an integrated view of useful health information for the public. This particular undertaking immediately ran into a tangled web of issues ranging from jurisdictional sensitivities through to the technological challenges of aggregating, into some kind of useful and authoritative view, tens of millions of pages of health information.

In the second case, an independent agency in the pharmaceutical sector determined that it needed to modernize its venerable compendium of pharmaceutical products and to deliver, using the accumulated content and editorial wisdom, a new generation of intelligent drug look-up tools. In this effort, one of the immediate issues encountered was the fact that information suited to print published will not exist in a form suitable for driving intelligent online behavior.

In both of the cases, a number of key lessons emerged that should resonate with everyone interested in applying modern content technologies to the management, improvement and deliver of effective health information as a digitally accessed service. These case studies illustrate, graphically, the types of challenges that need to be addressed and the magnitude of the benefits that can be realized when these challenges are overcome.