Piramidal, a promising startup founded by Dimitris Sakellariou and Kris Pahuja, aims to revolutionize electroencephalography (EEG) data analysis. The co-founders have identified a critical need for a foundational model that can effectively analyze brain scan data, which is currently fragmented and requires specialized knowledge for interpretation. By developing a software solution that can reliably detect worrisome patterns in EEG data, regardless of the machine type or location, Piramidal hopes to improve outcomes for patients with brain disorders while alleviating the burden on healthcare professionals.
One of the key challenges in EEG data analysis is the lack of standardization across different EEG systems and hospital setups. Each machine may vary in the number of electrodes used and their placement, making it difficult to develop a universal model for interpreting brainwave patterns. Sakellariou and Pahuja recognized the potential for automation in EEG analysis but realized that deploying such technology would require a fundamental shift in how data is processed and interpreted.
The Foundational Model
Piramidal’s founders believe that a foundational model for EEG readings could streamline brainwave pattern detection and eliminate the need for extensive studies and manual data annotation. While the specifics of their model have not been published yet, Sakellariou and Pahuja are confident in its capabilities and are currently preparing to scale the technology for widespread use in hospitals. The goal is to develop a model that can interpret signals from any EEG setup, regardless of the number of electrodes or the type of machine used.
The first production version of Piramidal’s model is scheduled to be deployed in hospitals early next year, with four pilot programs planned to test its effectiveness in different ICU settings. The company’s approach is focused on collaborative development with healthcare providers, ensuring that the model is fine-tuned for specific applications and meets the needs of patients and clinicians. While Piramidal’s ultimate goal is to improve the quality of care through AI-driven analysis, they emphasize the importance of ongoing refinement and adaptation to different clinical scenarios.
To support their ambitious goals, Piramidal recently raised a $6 million seed round led by Adverb Ventures and Lionheart Ventures, with backing from Y Combinator and angel investors. The funding will be used to cover computational costs for model training and expand the company’s team. In terms of data acquisition, Piramidal is leveraging open-source datasets and forming partnerships with hospitals to access valuable training data. The collaboration with healthcare providers is expected to yield thousands of hours of EEG data, which will be instrumental in enhancing the model’s capabilities and potentially surpassing human performance in pattern recognition.
While Piramidal’s foundational model represents a significant advancement in EEG data analysis, the company recognizes the need for continuous innovation and adaptation to evolving healthcare needs. By leveraging AI technology, Piramidal has the potential to transform the way brain scan data is interpreted and improve diagnostic accuracy for patients with neurological disorders. As the company moves forward with its pilot programs and expansion plans, the quest for superhuman capabilities in EEG analysis looms on the horizon, promising a future where AI-driven insights enhance patient outcomes and drive advancements in clinical care.