Using machine learning to evaluate large-scale brain networks in patients with brain tumors: Traditional and non-traditional eloquent areas

Abstract

Background Large-scale brain networks and higher cognitive functions are frequently altered in neuro-oncology patients, but comprehensive non-invasive brain mapping is difficult to achieve in the clinical setting. The objective of our study is to evaluate traditional and non-traditional eloquent areas in brain tumor patients using a machine learning platform. Methods We retrospectively included patients who underwent surgery for brain tumor resection at our Institution. Preoperative MRI with T1-weighted and DTI sequences were uploaded into the Quicktome platform. We categorized the integrity of nine large-scale brain networks: language, sensorimotor, visual, ventral attention, central executive, default mode, dorsal attention, salience and limbic. Network integrity was correlated with preoperative clinical data. Results One-hundred patients were included in the study. The most affected network was the central executive network (49%), followed by the default mode network (43%) and dorsal attention network (32%). Patients with preoperative deficits showed a significantly higher number of altered networks before the surgery (3.42 vs 2.19, p<.001), compared to patients without deficits. Furthermore, we found that patients without neurologic deficits had an average 2.19 networks affected and 1.51 networks at risk, with most of them being related to non-traditional eloquent areas (p<.001). Conclusion Our results show that large-scale brain networks are frequently affected in patients with brain tumors, even when presenting without evident neurologic deficits. In our study, the most commonly affected brain networks were related to non-traditional eloquent areas. Integrating non-invasive brain mapping machine-learning techniques into the clinical setting may help elucidate how to preserve higher-order cognitive functions associated with those networks.

Understanding Treatment Preferences and Cognitive Outcomes in Patients with Gliomas

“…The potential to localize the networks, not just of motor, sensation, and speech, but especially of other cognitive functional networks like memory, which is both highly vulnerable to treatment-related de cits and highly valued by patients, needs to be explored. In particular, a renewed focus on patient education and patient autonomy in determining what matters most in glioma treatment is required when considering the impacts of surgical and nonsurgical treatments on cognitive functional networks [17][18][19]. As more is learned about the impacts of treatment on various cognitive functions, it becomes essential to share that information to patients considering treatment, particularly considering the importance of these functions to quality of life.…”

“…This eld has contributed to a revolutionary understanding of functional neuroanatomy by using machine learning analysis of aggregated fMRI data to identify networks of cognitive functions and could transform surgical planning [16]. The human connectome project has identi ed a number of highly preserved brain networks that are consistent between individuals whose disruption in surgery has been shown to result in cognitive de cits and effect quality of life [17][18][19]. Increasingly, quality of life is incorporated into treatment paradigms and clinical trials.…”

Connectomic Networks and Their Impact on Clinical Outcomes in Glioma Treatment: A Review

“…The ability of general processing, critical reasoning, problem-solving, and retaining knowledge and skills through long-term memory can be diminished. In a study of 100 patients by Morell et al, 24 patients with brain tumors and associated neurologic deficits had on average 3.42 affected networks seen on preoperative MRI versus 2.19 affected networks in patients without deficits, and the CEN was most commonly affected (49%), followed by the DMN (43%) and DAN (32%). The frontoparietal cortical areas interconnected by the dorsal perisylvian white matter are important for executive functioning, and damage to the dorsal frontoparietal and intralobar frontal tracts is correlated with worse executive function.…”

“…24 These nontraditional areas are extensive throughout the brain's connectome and are involved in higher cognitive functions; their disruption can significantly worsen quality of life by impairing mentalizing and theory of mind in affected individuals. 5 The DMN is often implicated in these nontraditional eloquent areas, 24 in part because of the multiple interacting subsystems within the DMN. 18 Thus, damage to any of its nodes can result in whole-brain functional connectivity decline beyond solely local effects.…”

“…The risk of having nontraditional regions of eloquence affected by gliomas may be as high as 93%. 24 These nontraditional areas are extensive throughout the brain's connectome and are involved in higher cognitive functions; their disruption can significantly worsen quality of life by impairing mentalizing and theory of mind in affected individuals. 5 The DMN is often implicated in these nontraditional eloquent areas, 24 in part because of the multiple interacting subsystems within the DMN.…”