Moderate evidence
More
AI chatbots read single EEG images to predict recovery after cardiac arrest, small study finds
Study title: Commercially available multimodal LLMs for EEG-based prognostication after cardiac arrest
- Evidence type
- Controlled or validated comparison
- Sample
- n = 33
- Sources
- 1 sourcePeer-reviewed
- Published
- 2 min read
- Truvace record
- TRV-2026-3020cite or check it
Study title: Commercially available multimodal LLMs for EEG-based prognostication after cardiac arrest
Researchers tested whether two commercially available chatbots, ChatGPT 5.4 and Claude Opus 4.7, could predict neurological outcome in comatose patients after cardiac arrest by looking at a single EEG image. The study used 33 patients from a prospective cohort, all recorded on the first day after arrest.
Read the study abstract
Objectives: Machine-learning and deep-learning approaches have been successfully applied to EEG-based prognostication in comatose patients after cardiac arrest (CA), but typically rely on task-specific signal processing, feature extraction, and model development. Here, we investigated whether two commercially available multimodal large language models (LLMs) could predict neurological outcome directly from EEG images. Methods: Thirty-three comatose patients on the first day after CA from a prospective cohort wer…

The gain
In an exploratory study of 33 comatose cardiac arrest patients, ChatGPT 5.4 and Claude Opus 4.7 each reached 93.9% accuracy at predicting favorable neurological outcome from a single 10-second EEG image using few-shot prompting.
In brief
Across five prompting strategies, ChatGPT's AUC ranged from 0.86 to 0.99 and Claude's from 0.74 to 1.00. Both models peaked at 93.9% accuracy with few-shot prompting, where four example EEGs were supplied.
A previously published dedicated deep-learning model reached 100% on the same cohort, so the chatbots did not beat the specialised approach. The authors describe the discrimination as variable but potentially promising and say performance depended substantially on prompt design.
Main points
- The study included 33 comatose patients from a prospective cohort, assessed on the first day after cardiac arrest.
- Each patient was represented by one 10-second EEG segment in a longitudinal bipolar montage, rendered as a PNG image.
- ChatGPT 5.4 and Claude Opus 4.7 were asked to estimate the probability of a favorable outcome.
- Five prompting strategies were tested, including few-shot prompting with four example EEGs from outside the 33 patients.
- AUC ranged from 0.86 to 0.99 for ChatGPT and from 0.74 to 1.00 for Claude.
- Both models reached their best accuracy of 93.9% with few-shot prompting.
- A previously published dedicated deep-learning model achieved 100% on the same cohort.
The full story
Machine-learning and deep-learning methods have been applied to EEG prognostication after cardiac arrest, but they typically need task-specific signal processing, feature extraction and model development. This study asked whether general-purpose commercial multimodal chatbots could skip that work and predict outcome directly from an image of the EEG.
Each of the 33 patients contributed one 10-second EEG segment plus a generic information block describing recording conditions. The chatbots estimated the probability of favorable outcome, and performance was scored with AUC and accuracy.
The spread in results was wide: Claude's AUC ranged from 0.74 to 1.00 depending on the prompt, and ChatGPT's from 0.86 to 0.99. The best accuracy for both models, 93.9%, came with few-shot prompting, while the dedicated deep-learning model scored 100% on the same cohort.
Sources
1 on file- Resuscitationdoi:10.1016/j.resuscitation.2026.111398
Talk it over
Each of the 33 comatose patients was represented by one 10-second EEG image. What can a single short segment tell us about outcome, and what might a clinician want beyond it?
AI chatbots read single EEG images to predict recovery after cardiac arrest, small study findshttps://truvace.com/item/commercially-available-multimodal-llms-for-eeg-based-prognostication-after-cardi
Each of the 33 comatose patients was represented by one 10-second EEG image. What can a single short segment tell us about outcome, and what might a clinician want beyond it?
Answer in the threadClaude's AUC ranged from 0.74 to 1.00 depending on the prompt, while a dedicated deep-learning model reached 100% on the same cohort. How would you weigh the convenience of a commercial chatbot against that spread?
Answer in the threadIf you were a family member of a comatose patient after cardiac arrest, how would you want a chatbot's 93.9% accuracy on 33 patients to be used, if at all, in decisions about care?
Answer in the thread
If you have more time
Look up how doctors currently judge recovery after cardiac arrest. Where could an EEG reading by a chatbot sit alongside those steps, and who would check it?
Answer in the threadThe best results came from few-shot prompting with four example EEGs. What would a larger cohort need to show before you trusted a particular prompt?
Answer in the thread
ace
What readers say