Voice Recognition

Agfa HealthCare at SCAR shared secrets for a successful implementation of speech recognition. According to Agfa, successful plans include the following elements:

Set the Stage for Speech Recognition

Elect a Leader

Selecting a leader may be the most important factor to determine if the installation and implementation of speech recognition will be successful. A strong leader leads by example and provides support, encouragement and discipline to drive the use of the product. Typically, the leader is the chief of the department or a 'champion' of the solution who advocates for the benefits of the technology.

Outline Objectives

Facilities that have successfully implemented speech recognition had a need to change the way things were done to either enhance the delivery of patient care by reducing report turnaround time, decrease reliance on transcription services and the related costs or improve workflow efficiency with PACS or RIS integration. There must be a strong internal push for change to create the commitment needed to succeed.

Confront Challenges

Expect and plan for varied learning curves and understand that there will be a spectrum of acceptance. All communication should be positive and include a recap of the rationale for the purchase and outline of the potential benefits. Talk to potential users to identify their position on, and understanding of speech recognition. Involve potential users in the project planning, goal-setting and decision-making processes. If potential users are not tech-savvy, provide training to bring them up to speed. Be open about how the technology will modify workflow.

Promote Best Practices

Best practices can often be summed up by practicing. Users should have sufficient time in "live-use" to develop consistent habits conducive to productive speech recognition usage. Ensure that all users have a solid understanding of the role macros and templates have to streamline workflow. The administrator should minimize time-consuming physical and mental tasks necessary for creating a report and arrange the necessary tasks to minimize visual keyboard or screen icon search.

Show Support

The introduction of speech recognition requires a plan for change management, and appropriate support staff should be provided. One or more system administrators should be assigned to oversee the project from installation throughout the live productive use of the product. Provide ongoing support and coaching during productive use. Ongoing support can be provided in a variety of methods, yet all must be planned during the purchase process. Commitment to the technology is required for success.

Assess the Acoustics

While the level of technology has dramatically improved to filter out extraneous noise, the success of speech recognition can be enhanced with a few minor changes to the acoustical environment of the dictation stations. Choose a dictation station that is away from major traffic and noisy areas.

Evaluate the Vendor

Select a vendor that has had success in a variety of healthcare settings. Provide site visits to existing facilities to observe users in action. Make sure that the software uses set standards and can integrate with existing technology. Ensure that the potential users are either involved with, or support the decision to go with that specific vendor. While the stability and experience of the vendor will have a direct relationship to the long-term commitment made to customer success, workflow features are key to the ultimate success of speech recognition with the users.

Emphasize Ease of Use

To streamline the process and increase productivity, select a system that provides an efficient method for using worklists, templates and macros and for coding reports. A quality speech recognition program will allow users to modify their own vocabulary either ahead of time or on the fly mid-dictation. If users can teach the system their unique preferences, it will increase recognition rates with product usage. Corrections should be voice-enabled so that the user spends as little time on the keyboard as possible. Corrections should also modify the accuracy of the acoustical recognition rates in future reports.

Focus on Flexibility

To add flexibility and accommodate different learning curves, consider a system with digital dictation options. With digital dictation, users can gradually integrate speech recognition into their daily workflow while the department or facility can maintain one system but offer both functions. To improve adoption, select a system that allows for varying dictation and reporting preferences, i.e., real time versus batch mode.



Dictaphone demonstrated at SCAR enhancements to PowerScribe for Radiology. PowerScribe Workstation Version 4.6 includes a web architecture accessed through a browser-based user interface for dictation, correction and system administration. The Acoustically Train Word feature improves accuracy by allowing users to train problematic words and phrases. The new Dictation Shortcuts (PowerNormals) Viewer displays personal and system shortcuts for easy document insertion. The new version also eliminates the need for a paper requisition by allowing users to select an exam from a sorted work list.

Dictaphone highlighted its new CDx (Clinical Data Exchange) Internet-based file sharing technology, which locates clinical information housed in various databases and combines the data into one organized patient work list and delivers it to physicians on demand through a web-browser interface.



StructuRad launched at SCAR Dictation Protocol and Transcription Protocol. The new technologies incorporate traditional report creation methods such as dictation and transcription and leverage advanced structured reporting and macro management capabilities. Dictation Protocol and Transcription Protocol allow users to create a complete and final reports in a single click and also facilitate automatic bill coding.

Around the web

Richard Heller III, MD, RSNA board member and senior VP of policy at Radiology Partners, offers an overview of policies in Congress that are directly impacting imaging.
 

The two companies aim to improve patient access to high-quality MRI scans by combining their artificial intelligence capabilities.

Positron, a New York-based nuclear imaging company, will now provide Upbeat Cardiology Solutions with advanced PET/CT systems and services.