Continual innovation of front-end speech technology

I sat down with Tim Ruff, senior product manager at Solventum, to discuss the background, potential and benefits of ambient and front-end speech technology in capturing clinical documentation. 
 

Front-end speech technology has experienced quite the evolution over the last decade. Walk us through how the technology has changed how clinicians document. 

Front-end speech was originally positioned as an improvement to transcription – lowering costs of transcription and improving turnaround time for note delivery. This process served as a training ground to support technology improvements that encouraged clinicians to migrate to front-end speech technology, where dictation software is used to document directly into the electronic health record (EHR). In 2009, health systems in the U.S. were incentivized to adopt EHR technologyopens in a new tab to improve quality, drive efficiency and foster information sharing. Many different EHR vendors entered the market, which initiated the desire for speech-enabled workflows that could provide a better clinical documentation experience. The evolution has been continuous and exponential, expanding from speech-to-text and command-and-control, to artificial intelligence (AI) automation and embedded clinical intelligence. 
 

What are some of the most impressive leaps the technology has made?  

Initially, speech recognition was quite a jump forward, providing physicians with real time updates to EHR records. The inclusion of AI increased accuracy and assisted with managing accents. Shortly after that, the introduction of commands, macros and templates further improved the clinician experience by improving productivity and the ability to spend more time focusing on the patient. The technology has continued to advance to ease administration, deployment and remote accessibility. 
 

Ambient technology has really been making waves lately. How can it complement front-end speech recognition?   

Ambient AI was introduced to clinicians within the last few years. Audio captured from the patient-clinician conversation can be automatically transformed into a complete clinical note in the EHR as a by-product of care instead of a separate burdensome task. While ambient documentation is the latest hot topic, it doesn’t solve every documentation need.  Front-end speech recognition is complimentary to ambient documentation by easing corrections to the ambient note as well as speech enabling other documentation workflows such as inbox management. Additionally, the integration of natural language understanding (NLU) as part of front-end speech capabilities provides the ability to nudge clinicians toward best practices so that documentation is efficient and as accurate and appropriate as possible. The nudges help prevent documentation gaps so that all the details necessary for coding and billing are captured the first time. 
 

How do you envision clinicians using front-end speech recognition and ambient technology? 

There is a perception held by many in the medical field that ambient technology is a replacement for front-end speech solutions. In reality, front-end speech and ambient documentation are tools in the clinician’s toolbelt, each serving a purpose and readily available when needed. While ambient technology creates a note automatically, in real time, pulled from conversation from the patient visit, clinicians should still review and edit the automated note. Additionally, front-end speech can still provide greater efficiency than manual navigation and typing. 

Sometimes documentation is most appropriate outside of the presence of the patient. For example, detailing obesity, bad habits or other sensitive topics might be situations where multiple documentation tools are needed and not just ambient alone. Front-end speech recognition offers clinicians the flexibility and the efficiency to navigate quickly and edit the note for the inclusion of the necessary details.

Physicians spend 50% of their workday in the EHRopens in a new tab performing administrative tasks. The combination of front-end speech and ambient technology reduces the manual nature of entering patient notes in the EHR, unlocking some of that time for clinicians, which allows them to spend more time focused on patient care, and less of their personal time in the EHR. 

As speech technology continues to evolve to address the challenges of clinicians, automation and AI will continue to play key roles to further improve their efficiency while requiring less customization to utilize it moving forward. 

Tim Ruff is a senior product manager at Solventum.

Courtney Howell-McAnelly is a marketing communication specialist at Solventum.