An AI Model for the Brain Is Coming to the ICU
SMRTR summary
A groundbreaking collaboration between Cleveland Clinic and San Francisco startup Piramidal aims to revolutionize intensive care monitoring through artificial intelligence that "reads" brainwaves in real time.
Currently, doctors in ICUs rely on electroencephalogram (EEG) readings that are typically reviewed just once every 12 to 24 hours, with analysis taking up to four hours for a single day's data.
"This type of thing is time-consuming. It is subjective, and it is experience- and expertise-dependent," explains Dr. Imad Najm, director of Cleveland Clinic's Epilepsy Center.
The new AI system, trained on nearly a million hours of brainwave data from tens of thousands of patients, promises to flag neurological abnormalities within seconds rather than hours.
"Our model plays that role of constantly monitoring patients in the ICU and letting the doctors know what's happening with the patient and how their brain health is evolving in real time," says Piramidal's chief product officer Kris Pahuja.
Testing in live ICU environments will begin within months, with developers proceeding cautiously to minimize false readings—especially missed diagnoses, which Najm describes as "a big problem that keeps us awake at night."
SMRTR provides this summary for quick context. The original article belongs to Wired.
Read the original article