We demonstrate the development of a non-invasive method for estimating intracranial pressure (ICP) using computed tomography (CT) images and artificial intelligence (AI) techniques. Elevated ICP is a critical condition that requires timely diagnosis and management to prevent severe neurological damage or death. Traditional methods for measuring ICP are invasive and carry risks of complications. This study aims to leverage machine learning algorithms to predict ICP levels from CT scans, providing a safer alternative. A comprehensive dataset of CT images from patients, including demographic and clinical information, was utilized to train and test the AI model. The model was designed to analyze various features in the CT images, such as brain tissue density and anatomical structures, and correlate these with ICP readings obtained through conventional methods. To enhance the model's accuracy, advanced techniques such as image patching and data augmentation were employed. The dataset was divided into multiple patches, increasing the number of samples and reducing overfitting. Data augmentation methods like blurring and rotation were used to further improve the model's robustness. This approach could significantly reduce the risks associated with current ICP measurement techniques and improve patient outcomes. Future work will focus on expanding the dataset, refining the model, and validating the method in clinical settings to ensure its applicability and reliability in real-world scenarios.

Dimitrios Theodoropoulos is a doctoral candidate at the University of Crete's Medical School, focusing on AI applications in medical imaging. He holds a degree in Physics with a specialization in Microelectronics and a master's in Computer Engineering. His research includes AI-driven diagnostic tools for diabetic retinopathy and non-invasive intracranial pressure monitoring. With a background in radiology, he has worked in various medical centers and currently supervises the radiology lab at Asclepius Crete. He is also an advisor for the Greek Cancer Prevention Society and Medpro24, contributing to AI model training for medical applications. His research has been recognized in conferences and scientific journals, contributing to advancements in medical imaging and AI. 

Dimitrios Theodoropoulos is a doctoral candidate at the University of Crete's Medical School, focusing on AI applications in medical imaging. He holds a degree in Physics with a specialization in Microelectronics and a master's in Computer Engineering. His research includes AI-driven diagnostic tools for diabetic retinopathy and non-invasive intracranial pressure monitoring. With a background in radiology, he has worked in various medical centers and currently supervises the radiology lab at Asclepius Crete. He is also an advisor for the Greek Cancer Prevention Society and Medpro24, contributing to AI model training for medical applications. His research has been recognized in conferences and scientific journals, contributing to advancements in medical imaging and AI.