Dr. Davin Lundquist, MD is the Chief Medical Officer at Augmedix. Augmedix provides remote medical documentation and live clinical support so clinicians can focus less on administrative tasks and more on what matters most: patient care.
The Augmedix platform converts natural clinician-patient conversation into accurate, comprehensive, and timely-delivered medical documentation. The platform is powered by a combination of proprietary natural-language-processing modules and human-expert assistants operating in HIPAA-secure locations.
You joined Augmedix as the Chief Medical Officer in late 2019, what had initially attracted you to the company?
Before joining Augmedix as the CMO, I was an Augmedix customer for more than 7 years. What attracted me to the organization was their passion to help doctors connect with patients and re-humanize healthcare. Burn out is a real thing and in my experience, nothing has made as big an impact as Augmedix in helping me focus on why I went into medicine – to care for patients. Augmedix is the helping doctors optimize their technology, trying to make it better, and consistently trying to improve EHR.
What are some of your primary duties?
I bring the voice of the clinician, the physician’s perspective, and the perspective of a physician executive at a large hospital system. I help keep it real in terms of what physicians need and I understand what it takes to deliver care and balance the demands of technology.
Additionally, I am a subject matter expert. I guide the development of solutions and consult with marketing and sales.
Augmedix is leading the development of hybrid AI systems, could you clarify what this is specifically?
Augmedix is a tech-enabled service that converts real-time natural clinician-patient conversations into precise medical documentation.
The idea that AI can replace or be better than humans is a misconception and is especially challenging in healthcare. AI is a part of the care team with medical documentation support, which is a great application of AI. And humans partnering with AI is even more effective.
Could you describe how the natural language processing Augmedix Notebuilder automation technology works?
From a clinician perspective, Notebuilder (the NLP) is in the background and doesn’t require any additional work. There’s no need to “wake it up” with a prompt such as “Hey Alexa” or “Hey Siri,” for example. It utilizes identifiers such as specialty, visit type, complaints, symptoms, and frequency alongside scribe specific inputs, to automatically generate and organize clinically relevant structured medical notes which can then be inserted into the patient’s electronic health record. Selection options are dynamically filtered as the virtual scribe makes entries such that the most relevant selections are displayed. The clinician keeps the patient-physician relationship.
How much time can clinicians save and how much efficiency can they gain from this technology?
In my experience, I estimate that the Augmedix service has saved me 2-3 hours per day. Now I can complete documentation in 15 minutes.
Could you provide some details on how Emergency Department environments are currently using Augmedix services?
Being able to connect ER doctors to virtual scribes provides a safe way to continue to support physicians. Augmedix is a proven partner in ambulatory and have successfully pivoted to the ED using the same hands-free hardware, connecting to a virtual service.
The Augmedix ED scribes support key facets of documentation, real-time alerts, and Electronic Health Records tasks. This includes tracking and providing reminders for labs, radiology, and EKGs, as well as attaching patient education materials at the end of each encounter. Additionally, Augmedix scribes positively impact clinician workflow and the patient experience by removing barriers between the ED and hospitalists. The ED virtual scribes assist clinicians to optimize workflow by prioritizing documentation requirements for patients being admitted, while still ensuring all ED patient documentation is complete for clinician review and sign-off by the end of their shift.
Augmedix ED virtual scribes support the important work of ensuring patients receive quality and timely care in the emergency department. They capture key details from the natural patient conversation that would otherwise be impossible for clinicians during busy shifts. They have been especially instrumental as COVID-19 has had a major impact on ED workflow. The fact that they are remote ensures complete safety and flexibility, eliminating additional exposure risk for the care team.
The Notebuilder technology uses external datasets. What type of data is accessed and what is it used for?
In addition to historical notes to train the model, we incorporated external dictionaries of medical terminology to deepen the accuracy and ensure documentation is accurate.
Is there anything else that you would like to share about Augmedix?
There is a human behind the technology. We have well-trained scribes for 35 specialties. They understand their specialty well and have necessary expertise to increase accuracy. This helps hospitals so there’s one simple solution for all their clinicians.
Physicians want to make sure the technology service is going to provide them with the same notes, quality, and preferences that they’re used to (not auto-corrected, generic notes). We want to replicate what a human can do with the assistance of technology. Our hybrid approach allows us to go beyond documentation to do things that are important –– such as being aware of other quality metrics so we can be proactive in care. In this way, it’s easy to pivot and be flexible.
Pre-charting is desirable, and AI can’t do that. Our service is flexible. Our model integrates with numerous platforms. You don’t have to adapt the EHR to Augmedix, the converse is true. Augmedix adapts to the EHR.
Thank you for the great interview, readers who wish to learn more should visit Augmedix.
- Removing Objects From Video More Efficiently With Machine Learning
- AI Hardware Technology Imitates Changes in Neural Network Topology
- Stealing Machine Learning Models Through API Output
- NeRF: The Challenge of Editing the Content of Neural Radiance Fields
- Researchers Create Self-Propelled Artificial Cilia