Cutting-Edge Neuroimaging for Functional Brain Network Evaluation in Cancer Patients

Objective

The overall goal of CONNECT is to establish functional connectivity MRI (fcMRI) as a quantitative imaging tool of large-scale brain networks, ready for clinical application. Cancer neuroscience has demonstrated that harmful brain networks drive tumor aggressiveness, yet no imaging tool translates these findings into clinical metrics. With CONNECT, I aim to develop AI-powered fcMRI anomaly detection, mapping and quantification in individual patients based on normative reference data, then establish the structural and neuropathological underpinnings of detected fcMRI anomalies and finally demonstrate their clinical potential in patients with oncologic brain diseases. To achieve this, I will create a reference database of healthy connectivity profiles by leveraging existing fMRI datasets, integrating variability across demographics and MRI protocols to ensure broad applicability. Based on this database, I will develop AI-powered tools for fcMRI anomaly detection at speed and scale. To demonstrate biological validity and clinical utility, CONNECT will establish two cohorts undergoing repeated MRI: patients with glioma and patients at risk for brain metastases from breast or lung cancer. To confirm biological validity, CONNECT will relate detected fcMRI anomalies to structural connectivity changes and transcriptomic findings from connectivity-guided tumor sampling. To demonstrate clinical utility, CONNECT will demonstrate fcMRI’s potential for early detection and prognostication in these patients. CONNECT represents a significant leap forward by transforming fcMRI from a promising research tool into an accessible technology for clinical application, thereby translating cancer neuroscience insights into improved patient care. This approach has great potential to extend to neurodegenerative diseases, amplifying its impact across neurological disorders.