Webinar recap: Improving the Urology Encounter with Ambient AI

In this April 2026 session, two healthcare technology veterans walk through what urologists should expect to gain from ambient AI, along with DeepScribe's urology capabilities and what specialty-specific ambient AI looks like in practice.

DeepScribe Presenters

Blake Swinehart — Director of Strategic Growth
Blake has spent over a decade consulting with healthcare organizations to drive strategic alignment and expansion. He currently focuses on innovation that transforms clinical documentation and workflow efficiency.

Ravneet Riar — Director of Customer Solutions
Rav brings a comprehensive knowledge of clinical workflows, technology implementation, and public health to her role, developing the right specialty solutions with clinicians, and ensuring an exceptional AI experience.

Key Insights

• General-purpose ambient AI wasn't built for urology. The variety of urologic visit types, from new consults to post-op follow-ups to in-office procedures, requires highly specific language (PSA trends, LUTS scoring, urodynamics, cystoscopy documentation) that general AI models don't handle with the precision urology demands.
  
  
• Pre-charting and context awareness reduce work before the encounter begins. When integrated with an EHR, DeepScribe ambient AI pulls all relevant data in advance, including prior notes, new labs, imaging, and pathology. Clinicians arrive at the visit with the context already structured rather than assembled on the fly.
  
  
• Coding happens automatically and continuously. DeepScribe pulls in existing diagnoses, captures new ones from the conversation, and calculates E/M codes based on visit complexity. No need for clinicians to stop and document codes separately.
  
  
• Adoption is the metric that determines ROI. DeepScribe has found that adoption and utilization rates are the foundation for positive outcomes in documentation quality, coding accuracy, and clinician satisfaction. Organizations at low adoption levels rarely see meaningful returns.
  
  
• The note is the starting point, not the end goal. Ambient documentation is the entry point into a broader ambient operating system, in which structured clinical data is connected to revenue cycle, operations, and workflow intelligence across the practice.

Why doesn't general ambient AI work for urology?

Urology is one of the more complex specialties in ambulatory care. A new prostate cancer consult, a post-op cystoscopy follow-up, and a BPH management visit all require different clinical vocabulary, documentation structures, and contextual framing. An ambient AI model trained primarily on general or primary care workflows will show those gaps when it generates a clinical note. 

Blake Swinehart put it directly: general ambient AI likely won’t know the difference between a voiding dysfunction visit and a kidney stone follow-up. For urologists, that difference shows up immediately in note quality.

“General ambient AI just really wasn't designed with a urologist in mind. It was more designed around typical primary care visits, and providers feel that gap every day.”

What urology specifically requires is a model that carries longitudinal context, whether it’s tracking a PSA trajectory, prior imaging, or symptom progression, and applies that context to each encounter automatically. DeepScribe has focused on closing that gap, drawing on urology-specific training data and ongoing feedback from practices across general urology and subspecialties like urologic oncology, and female pelvic medicine.

How does pre-charting work for urology?

Ideally, before a clinician walks into the room, their pre-charting work has already been done, work that’s often on the shoulders of a medical assistant or nurse. An ambient AI platform can take on that prep work, and that’s just what DeepScribe does. When integrated with the EHR, the platform pulls patient data from the EHR—prior notes, new imaging results, recent labs, pathology, visit diagnoses—and organizes it into a structured pre-charting summary available days or even a week in advance.

For urology follow-up visits in particular, this matters. A follow-up for a patient on active surveillance for prostate cancer, or being managed for recurrent kidney stones, arrives with clinical history spread across multiple EHR tabs and document types. DeepScribe consolidates that into a single structured view, organized by imaging, labs, pathology, and prior plan. The provider walks in already updated rather than assembling the picture during the encounter.

"You only have to sit down with the patient and talk about what's happening that day. The groundwork to gather data, look at your past note—all of that is done before you go into the room."

Because that prior data is already incorporated into the note structure, clinicians don't have to verbalize imaging findings or lab values word-for-word during the ambient recording. The conversation can stay focused on what's happening that day.

Watch: 2 minutes on pre-charting for urology
‍Also: DeepScribe SmartPrep pre-visit intelligence
‍

What does context-aware documentation look like in practice?

The live mock encounter featured in the webinar illustrates what context awareness actually produces. Ravneet plays the urologist in a new consult scenario: a 42-year-old patient referred for an elevated PSA of 5.8, with a family history of prostate cancer and new lower urinary tract symptoms. The conversation covers PSA velocity, workup planning (free PSA ratio, transrectal ultrasound, multiparametric MRI), IPSS scoring and, given the patient’s age and family history, an extended discussion about active surveillance.

"Your note feels like your own note. Your note sounds like a continuation of your last note with all of the updates from today."

Within a couple of minutes after ending the recording of the encounter, a SOAP note is generated that captures the chief complaint, a full HPI organized by problem, medications and supplements, family history, social history, review of systems, impression, plan, and after-visit summary. Urology-specific terminology including PSA density, PI-RADS scoring, and fusion biopsy, plus E/M coding at the appropriate complexity level, all appeared in the note without being spelled out verbally during the encounter.

The after-visit summary translates clinical language into patient-facing plain language, giving patients a clear record of what was discussed, what studies were ordered, and what to expect next.

Watch: 1 minute on context awareness in ambient AI for urology

How does DeepScribe handle urology coding?

Coding in urology is operationally significant, particularly for practices managing complex longitudinal patients with multiple active diagnoses. As discussed in the webinar, DeepScribe's coding workflow starts with what's already in the EHR, with established diagnoses pulled forward as context. From there, anything new discussed during the encounter is added, and the model calculates an E/M code based on medical decision-making complexity.

When the platform is EHR-integrated, HCC coding assistance runs alongside this coding work. DeepScribe tracks the HCC codes a patient is eligible for and surfaces them in real time as the clinical conversation addresses the relevant criteria. The documentation captures the full picture of patient complexity without requiring a separate coding review after the fact.

‍Watch: 90 seconds on automated coding features for urology 
‍

What does high ambient AI adoption actually mean for a urology practice?

Blake is direct about this: Technology itself is not what determines whether ambient AI will deliver meaningful returns. Adoption of that technology is. Across DeepScribe's customer base, organizations reaching 80%+ adoption see better outcomes. Organizations at lower adoption levels struggle to justify the investment.

The onboarding approach reflects this and drives high adoption, with an overall DeepScribe adoption rate over 85%. DeepScribe reviews a provider's prior notes before go-live to build a profile that matches their documentation style, structure, terminology, and formatting. The first note a clinician generates already looks and sounds like their own. From there, the model can learn from every edit, and then apply preferences to future visits automatically.

Rav described what this produces in practice: Providers finish clinic work on time, documentation is complete, and the downstream effects reach the MAs, front desk staff, and admin team, who spend less time managing documentation gaps.

"When a provider says that they won't work somewhere that doesn't have DeepScribe, that's a retention signal worth paying attention to."
‍

To learn more about urology-specific ambient AI: 

• Watch the complete webinar now
  ‍
• See more on our website 

‍

‍