Epilepsy surgery candidate identification with artificial intelligence: An implementation study
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Artificial intelligence, including machine learning algorithms and large language models (LLMs), shows promise in identifying patients for epilepsy surgery evaluation and extracting relevant clinical information for referrals.
Area Of Science
- Neurology
- Artificial Intelligence
- Medical Informatics
Background
- Epilepsy surgery evaluation requires careful patient selection.
- Identifying suitable candidates can be time-consuming.
- AI tools may streamline this process.
Purpose Of The Study
- To evaluate a machine learning algorithm for identifying epilepsy surgery evaluation candidates.
- To assess a large language model's (LLM) performance in extracting key information for referrals.
Main Methods
- AI analyses applied to patients in an epilepsy clinic over 12 months.
- A random forest model stratified patients by surgery candidacy likelihood.
- Top 5% underwent manual review; an LLM extracted referral-relevant data from clinic notes.
Main Results
- 53.3% of manually reviewed patients met criteria for epilepsy surgery evaluation.
- 20% were referred within one month.
- LLM accuracy ranged from 80-100%, with most errors in management plan summarization.
Conclusions
- AI, including machine learning and LLMs, shows potential to aid in identifying patients for epilepsy surgery evaluation.
- AI tools can assist in extracting critical information for surgical referrals.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
