Cleveland Clinic Is a partnership with the Piramidal startup based in San Francisco to develop in a large -scale AI model that will be used to monitor patient health health in intensive care units.

Instead of being formed on text, the system is based on electroencephalogram data (EEG), which is collected via electrodes placed on the scalp, then read by computer in a series of corrugated lines. The EEG records electrical activity of the brain and changes in this activity can be indicated for problems. In a context of intensive care, doctors scan EEG data in search of crisis evidence, altered consciousness or to decrease in brain function.

Currently, doctors are counting on continuous EEG monitoring to detect abnormal brain activity in a patient in intensive care, but they cannot monitor each individual patient in real time. Instead, EEG reports are generally generated every 12 hours of ore 24 hours, then analyzed to determine if a patient is an experience in the neurological issue. He can take two to four hours to demonstrate the value of the data of brain waves.

“This type of thing takes time. It is subjective, and it is dependent on experience and expertise,” said Imad Najm, a neurologist and director of the Epilepsy Center of the Cleveland Clinic Neurological Institute.

The system that the Cleveland Clinic and Piramidal develops are designed to interpret continuous flows of EEG data and signaling anomalies in a few seconds so that doctors can intervene earlier.

“Our model plays this role of constantly monitoring patients in the USI and making doctors know what is happening with the patient and how their brain health evolves in real time,” explains Kris Pahuja, director of Piramidal products.

Pahuja and CEO Dimitris Fotis Sakellariou founded Piramidal in 2023, in order to build a brain foundation model – a system that can read and interpret the neural signals largely through different people. Before that, Sakellariou spent 15 years as a neuroengineer and scientist of AI doing EEG research. Pahuja World on product strategy at Google and Spotify. Their startup, which is supported by Y combinator, raised $ 6 million in seed financing last year.

The company has built a USI brain model using publicly aviipilable EEG data, as well as EEG data from the owner of the Cleveland Clinic and other partnerships. Sakellariou says that the model incorporates nearly a million hours of EEG surveillance data from “tens of thousands” of patients, both healthy and unhealthy neurological craftsmen. The brain activity models are extremely variable from one person to another, therefore the construction of a brain model requires amounts of data to capture common models and characteristics.

“The beauty of a foundation model is in the same way that Chantgpt can generalize the text, it can adapt to your tone, it can adapt to your way of writing – the model is able to adapt to the brains of different people,” explains Sakellariou.

Cleveland Clinic and Piramidal Team is currently using retrospective data for patients to refine the model. Over the next six to eight months, they plan to test the model in an environment that is closely controlled with data on living patients and a limited number of beds and doctors. From there, they aim to slowly deploy the software to all of the USI. Maybe the software was going to attract the hospital system to monitor hundreds of patients at a time, says Najm.