ar7hur

Refraining from interactions with VCs (it's hard to resist! "hey {first_name}, we love what you are building at {company}") brought me a lot of clarity. Once the silence settled in, I started to hear all these weak signals from our customers and prospects that were lost in the noise of hype before. I started to make decisions for them, only for them, not for the bold insights of my investor deck. And it’s already paid off in the two best currencies: customer love and ARR.

Sometimes, as a startup, you don't have enough hard data to raise money. So you have to recite what VCs want to hear (agents, anyone?). I'm not criticizing: I've been there, which entrepreneur hasn't?

But if you play this game too long, you lose yourself. You end up drinking your own Kool-Aid. Your McKinsey inner-self will select a GTM strategy just because it's making your differentiation slide better.

Do you guys agree?

We're Alex, Martin and Laurent. We previously founded Wit.ai (W14), which we sold to Facebook in 2015. Since 2019, we've been working on Nabla (https://nabla.com), an intelligent assistant for health practitioners.

When GPT-3 was released in 2020, we investigated it's usage in a medical context[0], to mixed results.

Since then we’ve kept exploring opportunities at the intersection of healthcare and AI, and noticed that doctors spend am awful lot of time on medical documentation (writing clinical notes, updating their EHR, etc.).

Today, we're releasing Nabla Copilot, a Chrome extension generating clinical notes from video consultations, to address this problem.

You can try it out, without installation nor sign up, on our demo page: https://nabla.com/copilot-demo/

Here’s how it works under the hood:

- When a doctor starts a video consultation, our Chrome extension auto-starts itself and listens to the active tab as well as the doctor’s microphone.

- We then transcribe the consultation using a fine-tuned version of Whisper. We've trained Whisper with tens of thousands of hours of medical consultation and medical terms recordings, and we have now reached an error rate which is 3× lower than Google's Speech-To-Text.

- Once we have the transcript, we feed it to a heavily trained GPT-3, which generates a clinical note.

- We finally return the clinical note to the doctor through our Chrome extension, the doctor can copy it to their EHR, and send a version to the patient.

This allows doctors to be fully focused on their consultation, and saves them a lot time.

Next, we want to make this work for in-person consultation.

We also want to extract structured data (in the FHIR standard) from the clinical note, and feed it to the doctor’s EHR so that it is automatically added to the patient's record.

Happy to further discuss technical details in comments!

---

[0]: https://nabla.com/blog/gpt-3/