Stereotactic MRI-Guided Radiosurgery Using AI Resting State Networks Recognition

Stereotactic Radiosurgery (SRS) is a noninvasive, high-precision therapeutic modality employed for the treatment of intracranial lesions. It relies on the delivery of intense ionizing radiation via multiple beams, typically administered in a few fractions. While SRS is highly effective, it is not without its challenges, particularly in relation to the dosage received by eloquent brain regions. Elevated radiation doses in these regions can lead to the development of neurocognitive disorders, with potential repercussions on cognitive function, memory, attention, and psychomotor abilities.The primary objective of this study is to harness the capabilities of Artificial Intelligence (AI) in the identification and integration of Resting-State Neural Networks (RSNs) derived from resting-state functional Magnetic Resonance Imaging (rs-fMRI) studies into the existing radiotherapy planning systems. Subsequently, we aim to simulate SRS treatment plans, combining traditional anatomical information with rs-fMRI data. This integrated approach seeks to optimize the therapeutic process by minimizing the radiation doses received by eloquent brain regions, thereby mitigating the risk of neurocognitive complications.