---
title: Generative AI and Emergency Medicine
description: Generative AI and Emergency Medicine
---

[Emergency Medicine Insights from CyrenCare](https://blog.cyrencare.com)

# [Generative AI and Emergency Medicine](https://blog.cyrencare.com/generative-ai-and-emergency-medicine)

 Written by [Kathy Chae](https://blog.cyrencare.com/author/kathy-chae) | Jan 2, 2025 11:57:42 PM

As we step into a new year, it’s worth reflecting on 2024—an era where we juggled ongoing boarding challenges, new technology hype, and the ever-evolving conversation around generative AI. This newsletter looks back at 2024—the “best of times and worst of times” for AI in health care—and highlights how we at [CyrenCare](https://www.cyrencare.com/?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs) believe in a more measured, provider-centered way forward.

Illustration by [Paul Noth](https://condenaststore.com/art/paul+noth?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs) "It wants to do our job

 

A Tale of Two Extremes  
*“It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness…”*

[The Fall of a $650M “AI-First” Company](https://www.fiercehealthcare.com/health-tech/primary-care-player-forward-shutters-after-raising-400m-rolling-out-carepods?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs)  
A once-hyped primary care startup—at one point valued at $650 million—pursued a futuristic vision of AI-driven “medical pods” to automate clinical care. In practice, they overestimated current AI capabilities and underestimated the complexity of real-world patient interactions and clinical oversight.

[Headlines of AI Chatbots “Defeating” Doctors](https://www.nytimes.com/2024/11/17/health/chatgpt-ai-doctors-diagnosis.html?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs)  
Simultaneously, media outlets stirred debate with claims that GPT-like models were “better” than doctors at diagnosing illnesses. The reality is far more nuanced, as multiple studies show below.

Conflicting Results: Man vs. Machine  
Various studies in 2024 showed mixed findings and conflicting results, with each context presenting different caveats.

- **Triage**  
  **Use of a Large Language Model to Assess Clinical Acuity of Adults in the Emergency Department **[*Williams et al., JAMA Netw Open 2024 *](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2818387?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs)The LLM accurately identified the patient with higher acuity when given pairs of presenting histories extracted from patients’ first ED documentation. 
- Diagnosis
  
  **Large Language Model Influence on Diagnostic Reasoning **[Goh et al., *JAMA Netw Open* 2024 ](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2825395?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs)In this trial, LLM alone outperformed physicians even when the LLM was available to them, making the headlines of NYT and indicating that further development in human-computer interactions is needed to realize the potential of AI in clinical decision support systems.
  
  **ChatGPT (GPT-4) versus doctors on complex cases of the Swedish family medicine specialist examination: an observational comparative study **[Arvidsson et al., BMJ Open 2024](https://bmjopen.bmj.com/content/14/12/e086148?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs#DC1:) Researchers tested GPT-4 on open-ended primary care questions. GPT fell short compared to physicians—especially in capturing psychosocial details, medication nuances, and individualized care planning.
  
  **Evaluation and mitigation of the limitations of large language models in clinical decision-making **[*Hager et al., Nat Med 2024 *](https://www.nature.com/articles/s41591-024-03097-1?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs)In a framework simulating a realistic clinical setting, LLMs performed significantly worse than physicians in diagnosing patients, follow neither diagnostic nor treatment guidelines, and cannot interpret laboratory results, thus posing a serious risk to the health of patients.
  
  **ChatGPT and Generating a Differential Diagnosis Early in an Emergency Department Presentation  **[*Berg, Hidde ten et al., *](https://www.annemergmed.com/article/S0196-0644(23)00642-X/abstract?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs)*[Annals of Emergency Medicine 2024](https://www.annemergmed.com/article/S0196-0644(23)00642-X/abstract?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs)  *In a retrospective analysis of ED cases, ChatGPT matched experts’ ability to generate differential diagnoses.
- **ED-to-Inpatient Handoff  
  Developing and Evaluating Large Language Model–Generated Emergency Medicine Handoff Notes [*Hartman et al., JAMA Netw Open 2024*](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2827327?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs) Although AI-generated handoff notes looked promising to automated scoring tools, human reviewers deemed them “marginally inferior” in real-world safety and usefulness. The study stressed the need for a 'physician-in-the-loop' processes.**
  
   
  
  **✔️ Key Takeaway**: [Warraich et al., JAMA 2024](https://jamanetwork.com/journals/jama/fullarticle/2825146?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs) LLM performance should be monitored in the environment where it’s actually used, not just on multiple-choice or short-answer tests—aligning with FDA guidance on real-world lifecycle monitoring.

 

CyrenCare's Approach:   
**We’re firm believers in using AI to “do the dishes, not the art.”**

[CyrenCare](https://www.cyrencare.com/?utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-9-ij1bOgtzVASeurylakCGzs5GX0xCxiy2ZgpYXctqtrnyeArFqu6dY-24EVycZyT44MVs) believes in using AI to handle repetitive and lower-stakes tasks so clinicians can focus on higher-level decision-making and one-on-one patient interactions.

- **Direct Patient Voices**  
  Our platform collects structured clinical symptom information *directly from patients* creating a patient generated HPI report, enabling providers to have focused meaningful conversations rather than struggling to ask questions repetitively to collect basic information.
- **Transparent, Rule-based Logic**  
  Our algorithms are **transparent**; ED teams see exactly how data is processed—fostering trust and interpretability. No more black-box mysteries.
- **Augmentation, Not Replacement**  
  Think of it as “doing the dishes.” We capture core patient details and flag red flags, but the **art of medicine** remains with you.
- **Seamless Integration**  
  Designed to **fit ED workflows**, we continually refine our interface based on clinician feedback to ensure simplicity and timesaving.
  
  *“We use LLMs as a lubricant for user experience—collecting symptoms more accurately—but our core system is a rule-based engine. Not every clinical challenge needs a black-box AI solution.”*

[View full post](https://blog.cyrencare.com/generative-ai-and-emergency-medicine)

```json
{
  "@context" : "http://schema.org",
  "@type" : "BlogPosting",
  "author" : {
    "@type" : "Person",
    "name" : "Kathy Chae"
  },
  "dateModified" : "2025-01-03T00:01:08.462Z",
  "datePublished" : "2025-01-02T23:57:42Z",
  "headline" : "Generative AI and Emergency Medicine",
  "image" : {
    "@type" : "ImageObject",
    "height" : 5041,
    "url" : "https://39691667.fs1.hubspotusercontent-na1.net/hubfs/39691667/smarterAI.jpg",
    "width" : 4900
  },
  "mainEntityOfPage" : "https://blog.cyrencare.com/generative-ai-and-emergency-medicine",
  "publisher" : {
    "@type" : "Organization",
    "logo" : {
      "@type" : "ImageObject",
      "height" : 60.0,
      "url" : "https://39691667.fs1.hubspotusercontent-na1.net/hubfs/39691667/CyrenCare-1.png",
      "width" : 309.57983
    },
    "name" : "Emergency Medicine Insights from CyrenCare"
  }
}
```