Show HN: Using GPT-3 and Whisper to save doctors’ time | Hacker News Reader

Show HN: Using GPT-3 and Whisper to save doctors’ time

117 pointsar7hur137 comments

Hey HN,

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/

View on HN

I don't spend much time worrying about AI ethics, but bringing AI close to patient interactions and record-keeping in healthcare seems grossly premature and irresponsible. There are countless brilliant applications of AI and the last one we need is a mediocre automated transcriptionist that distances a doctor from their responsibility to engage with the patient and can bear no accountability for error.

This is a task that perhaps can be supported with AI some day, but there are fields that deserve the application of a mature technology, not the gold rush rush game of integrating today's hottest thing.

The transcription part doesn't bother me. That's just advanced speech-to-text.

The summarization part, though, is dangerous. It's a very quick path to us losing all faith in our own medical records. Any way you slice it, no matter how much you train it, it's still going to be vulnerable to hallucination errors that slip by the reviewing doctor and become part of the patient's medical history.

I understand your concerns. Our objectives are, on the contrary, to reduce the doctor/patient distance that was created by EHRs and all the required administrative processes. We already measure that when AI takes care of this stuff, doctors do better engage with patients.

"I don't spend much time worrying about AI ethics"

Maybe you should re-evaluate whether that's the right choice based on your own comment, especially when ethics and safety experts are being fired by big tech co.

As a patient, this is an application that I would be extremely wary of, personally. I don't want the details of my conversations with my doctor to be sent to a third party in the first place, and I wouldn't trust the results of the transcription to be correct.

My doctor could vet it for accuracy, I suppose, but why? He's already putting his notes in my records anyway.

It takes 40% of your doctor time to put his notes in your medical records. We have been testing with doctors to reduce this time drastically so that they can spend more quality time with you. We also interviewed a lot of patients who take notes to be able to remember what’s been discussed during a consultation. So it could also benefit the patient

Your response completely ignores the concerns raised in the comment you are replying to. "It saves your doctor 40% of their time" is not a reasonable response to "I don't want my private medical details divulged to a third party".

Perhaps avoiding commenting on patient privacy concerns may be a better answer than any evasive answer may ever be.

Their website's privacy policy doesn't say anything about their product. It's also incorrect (they seem to have switched tracking providers without updating their privacy policy). I would quote the sections that are incorrect, but their terms and conditions forbid copying any content from their site. Their T&C also forbids me from linking to their page or terms.

I very much doubt that using this product as advertised is even legal under the GDPR. The company is French and health data, even pseudonymized, is strictly regulated. The "demo" doesn't feature any explicit consent at the very least.

It looks like they're making all privacy risks and concerns the doctors' problems. After all, they're the ones violating the law when they use this product.

At least in US this company would responsible for violating the law. Not only the physician.

I honestly have no idea how these products are even developed. Does nobody think about any laws at all? This is something I really hate about the current state of "hustle" development.

Thanks for looking into our legal documents. The ones related to the Nabla Copilot product are different from the website T&C and privacy policy, and can be found here: https://www.nabla.com/legal-documents/

Our product is GDPR compliant.

I read the French legal documents and your notices regarding GDPR.

What I don't understand from those, and statements made by your team in this thread, is how some claims can be compatible.

- That the product is GDPR compliant.

- That you don't store the PII or health data.

- Yet all data is stored at Google servers.

- Also, you reserve the right to re-use said data. (Which, since this is for R&D purposes, should probably qualify you for the need to ask the CNIL for an authorization as "health data repository"? [0])

- That none of the data is sent outside the EU or to additional 3rd parties.

- Yet it uses a fine-tuned "GPT-3" (a term that to the best of my knowledge exclusively refers to Microsoft/OpenAI's US-based API service, not to on-prem GPT-like LLMs like GPT-J or GPT-NeoX).

All in all, I can feel the enthusiasm but it does feel like this thread would have been so much more reassuring with some proactive comments about the privacy/health data issues, rather than have everyone voice the obvious concern with no prepared answers.

[0] https://www.cnil.fr/fr/la-cnil-adopte-un-referentiel-sur-les...

You're absolutely right, we should have been much more upfront with the privacy/security aspect of the product and add this link to the post: https://www.nabla.com/blog/privacy-security/

I hope this link will clarify our position.

Here are additional answers related to your points. - We do use Google Cloud to host our backend in EU or in the US but also the data for the Care Platform product. For the Copilot product, we don't host any data. They are hosted locally on the practitioner browser. - Our T&C reserves the right to re-use data in the event we will store the data in future versions of Nabla Copilot. In any case, the reuse of data, even health data, is allowed by GDPR for the improvement of the service provided if the data controller (practitioner) authorizes us and if they have informed the patient. - We did not say that "none of the data is sent outside the EU". Actually we say the opposite in the Copilot APD Annexe 1. We specifically mention Google and OpenAI and we comply with GDPR with a data protection agreement with both these companies.

> that they can spend more quality time with you

Maybe its just the doc offices i've been to, but i think this would actually just increase the patient churn in a hospital. There's no way a doc is going to increase their patient time by the 40% saved if the hospital can toss them in front of another patient.

It's definetly not just you... Tho that may still be a plus considering the shortage of doctors in a lot of countries. But the privacy concerns still remain.

I can absolutely see the benefits of this, that's not the question. The question is whether or not the benefit justifies the risk.

Most people doing a video meeting with their doctor are already trusting their information to third parties in some way. I understand that you probably are in a different situation. Perhaps services like these are on the path to your more privacy-oriented doctor having a tool that helps them and is comfortable for knowledgeable patients.

Yes, if you're doing telehealth, you're already taking a risk in terms of privacy. But the service provider doesn't need to actually pay attention to what you and your doctor are talking about to provide the service.

This is different, in that this service requires that a third party pay attention to the content of your discussion in order to work. That's significantly more intrusive.

> I don't want the details of my conversations with my doctor to be sent to a third party in the first place

Then you simply cannot use modern healthcare. Unless you're doctor is cash-pay, off the cuff, they're _legally_ using third-parties to perform their services.

----

If your doctor/healthcare provider is a covered entity, they're bound by HIPAA. That means they're free to establish relationships with Business Associates that are part of delivering healthcare. Those Business Associates must also be HIPAA compliant.

Many practices today have a transcriber with every doc, and that person is their personal scribe. You trust them, right?

In any case HIPAA will protect you. Your provider will have a business associate contract with them, so they're subject to huge fines if they violate that.

This is a demo. In real life this'll be wrapped with a whole lot of legal contracts.

I trust a scribe because they are a person with years of specialised training and experience who can be held accountable.

> in any case, HIPAA will protect you

It protects you insofar as it disincentives honest actors from doing sketchy things with your data. It's punishment for orgs, not protection for patients, like how laws against murder punishes the murderer rather than protecting the victim.

The best thing a patient can do to protect their privacy is to be actively avoid of medical practitioners that, for example, use tools like this which send your private medical consultation transcripts to God-knows-where.

"I trust a scribe because they are a person with years of specialised training and experience who can be held accountable."

They can be held accountable, but it's unclear how much specialized training or experience they actually have. They could have had a 6-month course, an online course, etc.

And of course you're assuming their EMR is actually secure. In a small office you can usually find the passwords you need stuck to the monitor or under the keyboard.

I like the idea, but I would like it even more if it was on-prem. The doctors (at least in EU) will be very wary of having their client meeting essentially recorded by a third party. With this as a cloud SaaS, patient confidentiality is essentially broken since the raw data is available to you while you transcribe it. I understand that you compete with "google speech-to-text" but this is not a feature meant to be used by doctors (even if they "illegally" do).

Obviously the business model is harder with on-prem, but cloud-first for doctor notes is in the long run much harder.

There are so many regulations and certifications to do that there are no chances anyone will have money and time to do all paperwork to host it on-perm. Because of these regulatory reasons most hospitals use so old systems that it surprises no one to see Windows XP with IE6 there. I had the pleasure myself of fixing a Material Angular bug that wasn't displaying form fields validation correctly in IE6 only 2 years ago, and that was for network of hospitals in Canada, US and Germany. Effort to allow them to use anything newer is too big and there are too many documents to review. It's simply cheaper to keep WinXP and IE6 running for as long as possible.

I've read about yesterday someone running LLAMA in a single GPU. Maybe if you optimise the model enough, you can give it to them as a box.

Off prem is also a rats nest of regulations.

Yes, but it's done once, on demand for each customer and service-wide certificates don't expire for a few years or releases, so the cost is lower.

Yep, I worked at a med-tech company and one of our biggest-value adds was that we were HIPAA compliant and part of Direct Trust — 10k customers one certification.

Thanks for your comment.

We plan to offer an on-prem option eventually.

In the meantime: - we offer a GDPR compliant EU-based hosting option - we don't store anything (no audio, no transcript, no note): it's stateless and all erased at the end of each consultation - data is pseudonomyzed as it flows though our systems

Our first customers (large healthcare orgs) have been OK with this so far!

Hi, I have two questions.

First question - you say that you don't store anything (no audio, no transcript, no note), but your legal agreement says that in order to use this service, the doctor asserts to you that they have gotten consent from the patient for you to reuse all data processed through the service for research and development, and to improve the performance, models, and algorithms of this or any other solution you come up with in the future. Why the difference and how do you square A with B?

"Due to the substantial financial, material and human investments made by NABLA within the framework of the Contract for the development and updating of the Solution, NABLA wish to be allowed to reuse the data processed within the framework of the Contract.

The CLIENT, when applicable in the name and on behalf of the DATA CONTROLLER, warrants that the Data Subjects have been informed of their rights and have given their consent for the use of their data within the framework of the Contract when required by applicable laws or the Regulation and authorizes the DATA PROCESSOR to reuse the Data processed within the framework of the Contract, as long as the latter undertakes to comply with the Regulation for all of this Data, for the uses listed below:

- research and development of the Solution,

- improving the performance, models and algorithms developed and trained by NABLA in the context of the Solution or any other solution published by NABLA,"

Second question - you say that you don't store anything (no audio, no transcript, no note) and that it is all erased at the end of each consultation. But do you store any artifacts derived from the audio, transcript, or notes of a consultation, like data processed directly or indirectly from the audio, transcripts, or notes of a consultation that is fed in to an AI model, ML model, or other dataset that you persist after the consultation?

I am in this space, dealing with similar problems with similar organizations. Any DPO that is not listing you as a shared controller will be acting illegally. And they might, its still a grey area, but how many of those assesments can you handle before your CAC becomes too much? Don't get me wrong I really think that it can help a lot on HCLS space if you want to chat more shoot me an email contact_vdk@proton.me

My worry is verification of facts. What if the model summarized incorrect facts, and it gets added to the medical history? How can doctors easily verify and fix things?

Have you had a medical billing experience recently? They don't care about facts, they care about maximizing the bill subject to plausibility constraints.

I fear that LLMs will kick this to a whole new level :(

Thanks for your comment. We clearly show a diff to the doctor before updating patient records. We are very explicit that they should check it (and edit if necessary).

It’s easier to revise than to write a first draft. It’s possible to miss mistakes on review, but relatively, I think it’s more likely that a distracted human will mess something up on the first draft.

That's exactly the thought process that leads to us trusting LLMs in places where they shouldn't be trusted.

A good benchmark could be to compare the accuracy of generated notes to the accuracy of other currently accepted practices like human medical scribes and assistants.

I feel uncomfortable knowing doctor would use anything like this.

How are 2 party consent states handled?

Is this HIPPA compliant?

Imagine a doctor instead mumbled his notes in a non native accent into a dictation machine, and had someone on minimum wage type them up?

Um there is dictation tools which are HIPPA compliant?

Of course, medical dictation is a whole industry. https://www.bls.gov/ooh/Healthcare/Medical-transcriptionists...

That job is known as a medical scribe.

https://www.scribeamerica.com/what-is-a-medical-scribe/

> A Medical Scribe is a revolutionary concept in modern medicine. Traditionally, a physician's job has been focusing solely on direct patient contact and care. However, the advent of the Electronic Health Record (EHR) created an overload of documentation and clerical responsibilities that slows physicians down and pulls them away from actual patient care. To relieve the documentation overload, physicians across the country are turning to Medical Scribe services.

> A Medical Scribe is essentially a personal assistant to the physician; performing documentation in the EHR, gathering information for the patient's visit, and partnering with the physician to deliver the pinnacle of efficient patient care.

HIPAA only applies to healthcare providers I think. Private companies like the extension maker can do whatever they want. At least that’s what someone on the internet told me and maybe it’s wrong.

If the provider wants to use the extension for patient care, the extension maker must be prepared to enter into an agreement to comply with the HIPAA rules.

https://www.hhs.gov/hipaa/for-professionals/covered-entities...

> If a covered entity engages a business associate to help it carry out its health care activities and functions, the covered entity must have a written business associate contract or other arrangement with the business associate that establishes specifically what the business associate has been engaged to do and requires the business associate to comply with the Rules’ requirements to protect the privacy and security of protected health information.

No. HIPPA applies to software as the software company would be considered a business associate.

"If you handle, store or transmit protected health information (PHI) to or from a covered entity then you need to be HIPAA compliant."

Source: https://github.com/truevault/hipaa-compliance-developers-gui...

This is not so clear-cut, though. There is a lot of gray area and doubt about this. HIPAA is not as complete protection as people think, and there are many situations where you'd think HIPAA would obviously apply, but it doesn't.

Business associate only comes into play when you're working with a covered entity. And, covered entities are far less inclusive than most people think.

----

The posted software is absolutely free to be non-HIPAA compliant. They're not a covered entity and without a relationship with a covered entity, they're not a business associate. However, without a relationship with a covered entity, they're also unlikely to generate any meaningful revenue.

Your statement is correct, but not complete.

When a covered entity (a HIPAA-required provider) does business with a private non-covered entity _and_ that transaction involves HIPAA controlled information, they must enter into a Business Associate Agreement (BAA). This effectively forces the private entity to maintain the same HIPAA standard as the provider.

A private company is absolutely free to build non-HIPAA compliant software, but they completely unlikely to get any healthcare providers to actually use it.

This! No mention of HIPAA on their site at all. Would be a total non-starter for any providers not in private practice.

Their website seems to suggest it is a French company with a US office. The issues around HIPAA would not be there and instead replaced by GDPR.

The blog posts also mention French trained ML.

> Cedille is a new open source French language model created by Coteries. It is trained to understand and write French and is also the largest model of its kind for French. Cedille is trained using large databases of publicly available content on the internet filtered for toxic content.

Expanding into the US, yes - they would need to deal with HIPAA, but until they do they likely don't need to.

From their main page (https://www.nabla.com/), the mention HIPAA:

Secure and HIPAA-eligible

    Audio, transcripts, and notes are not stored by Nabla
    HIPAA-eligible and GDPR compliant
    SOC 2 and ISO 27001 certifications in progress

Digging deeper (https://www.nabla.com/blog/privacy-security/):

This data processing is done on Nabla's servers, which are powered by the HIPAA and GDPR compliant Google Cloud Platform (GCP), and on HIPAA-eligible LLM servers.

What does "HIPAA-eligible" mean?

EDIT:

I was very curious about this and did a bit of research. The answer to it is squishy. It seems to be mostly a marketing term. The best definition I found was this:

"A service that is HIPAA eligible is one that is capable of being configured in a way that could meet HIPAA compliance requirements, but you have to know how to do it, it doesn’t happen ‘out of the box.’"

https://www.cleardata.com/articles/hipaa-eligible-hipaa-comp...

So it sounds great but doesn't actually mean that much.

lmao literally a weasel word

"HIPAA-eligible" means they have no idea what they're doing in healthcare.

The second you said that in a pitch to executive leadership, they'd realize you have no idea how to operate in healthcare.

What that really means is:

* they may be HIPAA-compliant (ie: they fulfill the requirements), * they haven't gone through a HIPAA-certification (no third-party cert), * they aren't using services that aren't HIPAA-certifiable

The latter point is important, because there are some services (ie: firebase) that apparently won't be HIPAA compliant. Some services are HIPAA-compliant if configured correctly. AWS has a list. I believe google does too.

There are a bunch of HIPAA guides out there.

So as a demo, it's not a big deal. But if they start selling this they need at least to be HIPAA-compliant with certification on the roadmap.

HIPAA is a self certification. You can claim compliance simply by following the rules. Therefore, you’re either HIPAA compliant or your not. I have never heard anyone describe it as being HIPAA-eligible.

HITRUST is a third party audit with higher standards than HIPAA. That is not a self-attestation.

I’ve spent a bunch of time in this space. Most of the major players offer HIPAA compliant services and sign BAAs. As of now, I don’t believe OpenAI offers a BAA, so this is dead in the water.

Well, HIPAA can be self-certified, but that probably won't stand up in court so most organizations will pay a third-party provider to perform the certification for them. That also lets you see the gaps, because HIPAA is big.

Here's AWS's list of HIPAA-eligible services. HIPAA-eligible is technology provider specific:

https://aws.amazon.com/compliance/hipaa-eligible-services-re...

Here's google's:

https://cloud.google.com/security/compliance/hipaa-complianc...

In general it means that the service may not be HIPAA compliant by default, but can be configured to be HIPAA compliant.

HITRUST is something else and it outside the scope of this discussion IMO. Not sure why you brought that up.

> Well, HIPAA can be self-certified, but that probably won't stand up in court

The is no certification requirement, so there is nothing to ”stand up in court”. Straight from the horse's mouth:

Are we required to “certify” our organization’s compliance with the standards of the Security Rule?

Answer: No, there is no standard or implementation specification that requires a covered entity to “certify” compliance.

https://www.hhs.gov/hipaa/for-professionals/faq/2003/are-we-...

Hi everyone! We’re of course well aware of the importance of HIPAA for all organizations operating in the US. To clear up any confusion: HIPAA-eligible means in our case that we’re ready to sign BAAs.

Either they are offering to sign a HIPAA BAA or they aren’t. If they aren’t, its HIPAA-radioactive. HIPAA-eligible for this kind of service is meaningless; when you have PHI going through a third party system, its BAA or GTFO.

Kindly reminder that HIPAA is a particular law applicable only in a single particular country.

This is the scariest real life attempt to use the bullshit machine I’ve seen so far.

Nothing like “subtly wrong” clinical notes to affect positive medical outcomes. And time pressed medical professionals are certainly well suited to vigilance tasks like correcting machine generated errors. /s.

I sincerely hope this never sees the light of day.

If you think this is scary you should see how it's done today without AI.

Doctor's handwriting is notoriously bad. Now try putting that into an EMR. Good luck with that! You think you're going to get someone who gets paid $2000/hr to type shit up? Yeah, guess again.

We don’t get paid even close to $2k/hr. Where is that number from? Many physicians type notes, others use dragon dictation, or medical assistants/scribes.

I used to type, now I dictate.

Quality of note is provider dependent though. If an LLM can improve quality of notes across the board through simple summary that’s interesting. But the input matters - is the provider actually asking the right questions? Did they look at past relevant history and other notes and put it in their current note?

The LLM should be in a HIPAA compliant environment and have access to the EMR. Then it provides a tidy summary for that. Second it should have relevant questions for the provider to ask during the interview but also a real time/dynamic component which generates relevant follow up questions dependent on patient responses.

Lastly it should put together the old and the new and generate a new note which can then be edited by the provider. Editing notes is much easier than generation of a new high quality note.

Technically interesting but solving the wrong problem.

Doctor's already today spent too little time with their patients to understand diseases holistically enough.

Adding technically between these 2 will make treatments in most of the cases worse, not better.

Wow, lots of haters in this thread. Having doctors spend less time with clerical/administrative work needed to record conversations will free up more time to focus on patients. I assume the generated note text can be easily modified by the doc if necessary, but saves them the bulk of having to type things up and remember everything that was discussed.

Nothing's going to fix that but having more doctors around.

Going from a recorded transcript to a summary note, extracted structured data, and diagnostic codes is a big time saver.

Right now, drs type notes as you speak to them. If this works, they'd have more time to spend actually listening and talking to you.

> solving the wrong problem

You seem quite confident about this, but based on the doctors I know writing notes is a real pain and mostly seen as (important) scutwork. They're only given a set amount of time to see a patient (including notes), and if you can reduce notes they'd actually get more time talking and seeing the patient.

What percent of that scutwork is actually note taking rather than clicking a bunch of checkboxes or combo boxes or otherwise navigating through EHR UI?

The last time I visited a doctor, he had an assistant with him that took all the notes.

> if you can reduce notes they'd actually get more time talking and seeing the patient

You're not thinking like a hospital administrator. Sounds like these doctors can handle 40% more patients to me!

You're both correct. My wife is a psychiatrist. I'm friends with several doctors through her.

Everyone hates notes. However, there are already a bunch of dictation services that help write notes more efficiently. They absolutely work well enough.

Adoption and funding is the primary blocker. When my wife works at facilities that have M-Modal, she's 80%+ faster. However, she works at many, who despite knowing the benefits, simply don't care to spend the time, money, or resources on implementing scribe/dictation services.

Last time I went to the doctor, he charged me $300 to google symptoms on a computer. Now you're saying I get to pay $300 for him to ask ChatGPT? No thanks. We all know that is where this is going - doctor decision support system, i.e. the AI has access to information locked behind expensive journals that the patient does not.

The tone of this reply is excessively snarky.

You clearly have a personal bad experience with your doctor - but that doesn’t mean that this potential product is a terrible idea.

> Last time I went to the doctor, he charged me $300 to google symptoms on a computer.

Would you be upset if someone were to ask you to help with their code and you needed to Google for documentation?

It's not exactly the same thing, but it's a similar situation.

I do tend to agree with you on the rest of your concerns, especially info being used that the patient has no way to verify - but that seems tangential at worst to the product in it's current state.

> It's not exactly the same thing, but it's a similar situation.

It's pretty close to the same thing. Programmers are paid for knowing how to program, not for memorizing APIs, system calls, etc. Being able to look up details is essential because it lightens the cognitive load, allowing deeper thought to be put into the real meat of the problem. No programmer can realistically memorize every technical detail they'll need to know.

It's the same with doctors. They're paid because they know the process of diagnosing illnesses, not for memorizing Gray's Anatomy and every pharmacopoeia. Being able to look stuff up is essential for the same reason it is for programmers.

Not to mention that medicine changes, and you want your doctor to have the latest information about whatever it is that ails you.

Last time I asked a developer to fix a bug, he charged me $300 to google the api docs..!

But this isn't even about the doctor asking GPT, this is transcribing the video call so the journaling becomes easier and the doctor can spend more time with their patients. A good thing, no?

I'll be happy to do that for like $200 as long as I can use ChatGPT

Doctors are liability lightning rods who've gone through a professional hazing process to join an exclusive club that has massively warped American medicine through its lobbying efforts. Unless you're dealing with a specialist they're often not super competent - that's why doctors are being replaced with nurse practitioners and physicians assistants everywhere (to the point that you often won't ever see an actual doctor at a lot of medical facilities) and usually people are completely oblivious to the fact.

Pharmacists in other countries do a lot of the basic shit doctors in the USA do, with less overhead, for cheaper and with a much faster turnaround. Antibiotics, blood pressure/cholesterol meds, antidepressants and other rubber stamp drugs absolutely do not need to be strongly controlled.

Antibiotics don't need to be strongly controlled?!

No. The transcription part is done by Whisper. After that, GPT summarizes a clinical note which may be truthful - or may have hallucinations just as well.

Well if you believe that you can have you fixed by Googling and still go to someone else to do the Googling you obviously have to pay.

No one is obligated to do free Googling.

Or maybe you didn’t go the for the Googling?

Last time I went to the doctor it was free.

I just like to point out the small advantages of living in Europe now and again

I'm not sure you understand the definition of free.

If I had never paid any tax in my life it would have been the same price. If I were a refuge, child, homeless person it would have been the same price. If I had lost my job due to being sick it would have been the same price.

That is my definition of free

I believe this product is to help the physicians transcribe notes and medical chart information from video/audio recordings, not diagnose.

You are not wrong; I did not think about this until you brought it up but it's a great plan to wrap nearly free customers, then get in bed with Health Insurers to reduce cost of care by recording and transcribing all doctor conversations, then alerting to possible fraud or excessive care based on patient input against your huge DB of patient interactions. Historically recording of the patient DR interaction has always been incredibly hard sell to do for the management types. Then insurers mandate chat capture with the visit summary billed to them. Love it!

If my doctor is googling he is doing his job.

He has medical knowledge which allows him to sort through information.

Any developer who doesn't google today is hindering his ability to develop and slowing the development process. Just like the dev cannot hold all the algorithms and language caveats/syntax in his head, a physician cannot hold all the information in his brain. He may not be familiar with symptoms or might be googling to see if there is anything "new" as pertaining to a patients specific issue.

There was a time steroid injections were the go to for tendon injuries, recently science came out that it results in worse long term outcomes and PT should be the first line treatment.

I would find it suspect if my specialist was googling unless I had something really weird going on, and might consider going to a different specialist.

If they're a neurologist and Googling "what does brain do", sure.

If they're searching up safety data on a slightly-over-recommended dose for a particular medication (like, say, a gastric bypass patient who takes an oral med that's not absorbed as much as in a person with a normal stomach), I'd much prefer they be certain than guess.

I still routinely Google some things that "I should know" when doing software development. Despite being already correct 99% of the time, an internet search helps with two things:

* Presenting potentially new information or methods of doing something. My "old" knowledge might still be correct, but a "new" method may have optimizations or benefits.

* Even though, I'm already 99% correct, it's not worth the hassle of being wrong the 1% of the time - especially when the solution is 20 seconds of internet search.

I would want my doctor to Google things. No one knows everything, but I expect an expert to know where to look. If he used Google the same way I do, he'd know which results are reliable. Doctors who finish med school and never consult a resource ever again cannot be good doctors.

Sure, but as an example, I paid a lawyer who presents that they are an expert in real estate. And I later find out they have zero understanding of legal non-conforming permits. It was a key factor I made clear upfront before I hired them.

I shouldn't have to pay for your time so you can educate yourself on the thing you purported to know before I hired you. Especially when the hourly rates are in the hundreds of dollars.

So there is some balance there. And doctors are using software that spits them out shit in real-time and then they send you on your way with some printed info you can readily find on the web.

I just had a knee issue and the doctor thought it may be a torn meniscus. Didn't have me do any range of motion movements. We just chatted and he said I think it is this. Take an xray. Bye.

I go get an xray and results come back the next day over the app. All he says in the app - "xray is normal".

So I'm thinking what the hell. That's where you leave it? So what. Should we do an MRI? What's next. You gave me an anti-inflammatory and your guess was apparently wrong. What a shitty engineer this person would be.

Tech can help in some cases. I like getting lab results back in the app. I like that I can follow up with chat later.

But I fear it's making doctors less effective. And the best ones will be the ones that maintain their traditional craft and nuance even with all the fancy new tools and tech that save them time.

Just like everyone is a React developer these days. Tech and advances can make many people sloppy. And dumber. It can push away great talent due to mandates of the tech or process. And attract new talent that is worse.

The "feed it to a heavily trained GPT-3" part is backwards. The transcribed notes should be stored and retained. The LLM should be applied later, in real time, when someone is referring back to the notes.

That way you still get the benefit of summaries, without polluting the medical records with hallucinated AI nonsense.

Ignore the nay-sayers in this thread. I doubt the ones commenting have spent many days personally observing or personally doing the problem scenario you are solving for.

This is spectacular for telemedicine in the post-covid regulation environment.

The new model is an onsite medical assistant taking patient vitals and setting up the Zoom, the patient then talks to an APRN (Advanced Practice Registered Nurse) in the patient's field (e.g. subacute care, gerontology, etc.).

The APRNs spend 30 minutes reading the file, tele-consult with a patient for a half hour. Medicare pays for the half hour. Then the APRN has to spend ~60 minutes turning that session into the EHR notes. Many just don't, they make an ultra short note, and hope Medicare doesn't challenge it. Patient loses. If too strapped for time, they also may not pre-read, because they only get paid for the video conf time. Patient loses.

Downside I see with this is the APRNs doing telemedicine are making about 1/4 of the rate they should be, and can't afford an expensive tool. The telehealth provider is not going to pay for it, they get paid for the 1/2 hour, the wasted time is not out of their pocket, so they won't spend to make it up. This is going to be on the nurse, who likely can't afford it.

On the plus side: time saved with this tool would go directly to ability to pre-read, or ability to see a patient an hour instead of one per two or three hours. It would also permit a higher caliber of expertise nurse to do tele-medicine visits in off days, because the time saved could bring comp back more in line with the expertise.

Go for it!

> This allows doctors to be fully focused on their consultation

Here's the thing: rigorous medical documentation is not a burden on the practitioner, but rather an integral part of treating someone. You need detailed historical data to diagnose some tricky conditions and treat them correctly.

What's the tool doing to prevent a practitioner from simply blindly copy-pasting the output without verifying it?

> and saves them a lot time.

Which means more billable. Have you considered charging per medical act performed?

Unfortunately I've had interactions with the medical system where the doctors wrote their after meeting notes with severe errors and pieces of fiction, as well as specialists who almost ran out of the room after 5 minutes of an appointment they billed 45 minutes for to the insurance. I doubt such cases are rare and I suspect this sort of product, which I'm a priori neutral about, would change the situation much for the worse.

Is Whisper really up to this? When I've used it, it's...fine. But it also randomly dreams up weird transcriptions that have nothing to do with what's actually being said, and with perfect audio too. Like the audio will be a presentation at a tech conference, and Whisper will predict "check out our site at randomlink.com", when no one said anything close to that.

I'd be pretty concerned if my doctor relied on this to be honest.

Good point, this is why we finetune Whisper with our own medical dataset to improve the performance

I'd prefer to have 0 AI interpreting my session with a doctor. Will there be any way for me to identify which physicians use this service so that I can avoid using them?

Hell, it's hard enough these days to identify which physicians are actually physicians. I recently had an NP introduce herself to me as "Dr. So-and-so."

I did some work at a startup that tried to solve a similar problem and currently finishing medical school. Nobody here is discussing the paperwork and burnout problem that exists in medicine that will make these tools valuable. Fortunately, language models used for summarization based on a transcript are much less likely to hallucinate compared to upscaling/denoising image algorithms. Recent LMs are scoring top marks on LSAT, so they are maintaining accuracy as they get larger. If doctors can save time, this will be useful for the patient in the long-run.

On-prem stuff is more tricky, but useful for privacy. Nvidia is supporting the space with their health AGX series.

> On-prem stuff is more tricky, but useful for privacy.

An on-prem installation, assuming that it doesn't phone home to the mothership, is something I would entertain. But one that requires disclosure of medical information to a service provider would be a lot harder for me to accept.

I used to work in a hospital and one thing that was already tough was when doctors would copy massive (multiple page) notes that were mostly boilerplate, probably stored as some template in the EHR. After seeing the same note a few times, you started to learn which page to scroll to in order to view the relevant section for the patient. I would hope this system only generates clinically-relevant text for that patient, usually only 2-3 paragraphs for a standard consultation.

This is really interesting work, congratulations! A lot of negativity in the comments seems to stem from cynicism around doctors and their billing, but in a country like the UK I know removing essentially "admin tasks" from the GP would be very welcome and give time back to the clinician, allowing them to spend more time focussing on the patient and outcomes.

> 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.

I think the general idea is interesting but this is the wrong benchmark to convince people imo. What is the actual error rate of Google's Speech-to-Text on this your data? How convinced are you that it's properly labeled? That it's 3x lower isn't necessarily impressive. Then a follow up question... What I as a patient would want to know is something else, how does it compare against medical secretaries that usually transcribe documents like these?

I'm a medical student and one of my main takeaways from spending time in clinical environments is that a lot of doctors and other staff seem to either not realise or not care about how much time is wasted due to slow, inefficient processes and technology. I was wondering to myself about what it might take to train a model to recognise medical jargon to help doctors rapidly transcribe audio recordings of their consults/spoken recounts of their notes. Not surprised that somebody is looking to make a commercial product of it, there's definitely space for it.

It looks pretty neat! Just wondering, what other languages are supported? English seems pretty natural as it is where GPT-3 thrives, but have you been able to support other languages as well?

We currently support English and French and we will be supporting more soon

How tied are you to the name, it's almost Nambla.

That's all I see when I see this company's name. Unfortunate choice.

What is OpenAI's privacy policy? How do you ensure the transcription doesn't get stored by OpenAI or used in training their future LLM versions?

I'd recommend looking at a fully self-hosted LLM alternative or ensuring the GPT-3 endpoint you're using is compliant (I've heard Azure has a GPT-3 API that is supposedly more "compliant").

Maybe the future of the first line of healthcare will start to use AI with the assistance of a specialized nurse to follow your health overtime and only fall back to a real doctor when a real problem occurs.

This could elevate a lot of problems we have with our public healthcare(Canada).

Love this and have had discussions w/ doctors and practitioners about this exact problem. I think extracting to FHIR is key, along with making it searchable as a longitudinal data store for a more robust personal record. Clear need. Would love to help.

This is really cool! What surprised you guys about fine-tuning whisper + gpt-3?

Domain specificity is a killer advantage when fine-tuning! For Whisper in particular, fine-tuning with actual medical speech gave us an immediate strong advantage over competing APIs.

Side-note: your email field to sign up for your newsletter is very wonky with both Safari and Firefox (at least for me). It only lets me type one letter at a time, for some reason.

Thanks for the report, it should be fixed now.

Just read the blog post, very interesting. I thought sections about changing the temperature and repetition penalty were particularly neat, thanks for including examples.

One thing that a lot of folks still don't get is that some problems are bad, while others are not-so bad.

Pretext: My wife is a psychiatrist. Via her, I'm am close friends with many doctors. Also, I know it's super easy to be critical, but I'm going to save you an insane amount of effort and heartache.

This is actually a solved problem in medicine, already. Maybe not solved, but absolutely "good enough". M-Modal, Dragon, and Telephone services (yes, call someone and have them translate) all exist. They all translate notes well enough. For most doctors, it's a huge improvement over typing.

The problem is primarily access. Most institutions simply do not want to pay for this. They don't view it as necessary. Further complicating the issue. Unless a physician is in private practice, they have little to no control over their ability to use a tool like this. HIPAA is far too high of a barrier.

You're trying to implement a technical solution to what's ultimately a politics/human problem.

----

I've spent a bunch of time in the space, both as a MedSpouse and the CTO of a HealthTech company. I wish you the best of luck as I'd love to see life be easier for my wife, but this is _extremely_ far from being MVP viable.

* HIPAA - You absolutely, 100%, without a doubt need to be HIPAA compliant. You likely want to be HITRUST compliant as well. You're very, very clearly a Business Associate (HIPAA term) and processing PHI. Without that, you're dead in the water.

* Chrome extension - This is not going to cut it. At baseline, it requires copy and pasting between Chrome and the EHR system. Bad UX.

    * More tactically, most EHR access occurs via remote access, like Citrix. Often from a thin-terminal on the hospital floor or at a med station. These are locked down environments that have a high likely hood of not having Chrome installed _and_ not being allow to install _any_ software (including Chrome extensions).

* Doctors need real-time feedback and the ability to correct. Real-time feedback is critical because carrying a patient takes mental load. Part of the purpose of notes is to create a written record that "closes" a patient for the time being. An async service isn't really helpful since it generates an interrupt after the physician has already moved on.

* Accuracy is important. Accuracy is even more important with _extremely_ similar and _extremely_ uncommon words. For example, can this system differential between prednisone and prednilisone? In fact, can it even identify those words? Can it do it with an accent?

* FHIR is only as much of a standard as tables are to relational databases. The hard part is the thousands of custom fields that are different between health systems.

* Structured data really isn't used or useful in medicine. You might have a medication list that can be structured as data, but actual notes include a lot of prose. I believe part of this is necessary to capture the nuance/minutiae of some healthcare. Part of it stems from the lack of any standard.

* Again, this is all possible already. Tele-notes have been done since the introduction of EHR systems.

* Nobody is going to use a Chrome extension.

Right now I am working on an MVP of a product just like this one for a private psychiatry practice.

> HIPAA - You absolutely, 100%, without a doubt need to be HIPAA compliant

This is incorrect: kindly reminder that HIPAA is a specific law applicable only to a single particular country. The vast majority of the world does not need to be compliant to it. Although similar laws exists elsewhere, they are not at all insurmountable.

> More tactically, most EHR access occurs via remote access

I have never seen a system where EHR access is used via remote desktop. (Although I admit that would be a good security measure). But then again, I'm not in the US.

> can this system differential between prednisone and prednilisone

Whisper is nothing short of magic in this regard. Yes it can - out of the box. Even more: it can easily differentiate lots of medical terms in languages other than english. Even more: it can easily parse compound terms like "hypothalamic–pituitary–gonadal axis" in other languages where first two words sounds very very similar.

I reckon that differentiating prednisone and prednisolone in doctor's writing is a harder problem than differentiating those from voice.

Having said that, your concerns and remarks are very valid. This is why we are not focusing on replacing written records/EHR fill out. There are other ways AI and large language models could deliver help for psych practice and the patient. And safer.

it sounds like you're removing a checklist from doctors.

Will you sign a BAA for HIPAA covered entities?

They list it under their "legal documents" page: https://www.nabla.com/docs/business-associate-agreement

They're saying that they will keep the information private, but I'm wondering about whether GPT will uphold this standard as well.

If I were a provider, I'd probably want to see their BSA with OpenAI.

Yes absolutely.

Looks very interesting! Are you recruiting ?

Yes we are! https://www.nabla.com/careers/

Congrats on the launch Nabla!

That's super cool!

Amazing idea!!

how can this be compliant with HIPAA?

It can be HIPAA compliant by every company in the chain of use of the data signing a BAA with every one of the companies they deal directly with that handles PHI, and then following the terms of HIPAA in dealing with the data.

this is amazing. great great

amazing. you're ROCKKKKKKKKKK

Reply on news.ycombinator.com