The field of artificial intelligence (AI) is rapidly expanding, with applications in various industries. Medical technology is one area where AI models are being utilized extensively, but there are challenges in achieving consistent outcomes. Piramidal, a startup founded by Dimitris Sakellariou and Kris Pahuja, aims to address this issue by developing a foundational model for analyzing brain scan data using electroencephalography (EEG) technology.
EEG technology is widely used in hospitals but is fragmented among different machines, making it challenging to interpret the data. Nurses and doctors often struggle to identify worrisome patterns in brain scan data due to the variations in EEG systems. Piramidal recognizes the need for a software solution that can consistently analyze EEG data regardless of the equipment type or location. This could significantly improve outcomes for patients with brain disorders while reducing the burden on healthcare professionals.
Sakellariou and Pahuja’s background in computational tools in neurology led them to the realization that automated analysis of EEG data could revolutionize patient care. By developing a foundational model for EEG readings, Piramidal aims to streamline brainwave pattern detection and make it accessible across different systems. The founders’ hands-on experience with neurologists in operating rooms provided valuable insights into the practical application of AI technology in healthcare.
One of the main obstacles in developing a universal model for EEG readings is the lack of standardization in EEG systems. Different machines vary in terms of electrode placement and data formats, making it difficult to create a one-size-fits-all solution. Piramidal’s approach involves building a foundational model that can adapt to any setup, regardless of the number of electrodes or machine model.
The founders of Piramidal are confident in the potential of their foundational model for EEG analysis, although their work has not yet been published. They plan to deploy the technology in hospitals early next year, with pilots in intensive care units to test its efficacy. The company’s focus on refining the model for specific applications sets them apart from other AI companies that rely on pre-trained models for various purposes.
Piramidal faces two significant challenges typical of AI companies: funding and data. A recent $6 million seed round provides a starting point for addressing compute costs and expanding their team. Data aggregation and harmonization are ongoing processes to ensure the model’s accuracy and reliability. Partnerships with hospitals offer access to vast amounts of training data, paving the way for further advancements in the technology.
Piramidal’s innovative approach to analyzing brain scan data using AI technology has the potential to revolutionize neurology and healthcare. By developing a foundational model for EEG readings, the company aims to improve patient outcomes and support healthcare professionals in their decision-making process. The future looks promising for Piramidal as they continue to refine their technology and expand its applications in medical settings.