Abstract

Title: Integrating Diffusion Tensor Imaging and Resting-State fMRI in Presurgical Planning of Intracranial Lesions: Enhancing Functional and Structural Mapping for Safer Resection

Background:
 Accurate presurgical mapping of functional networks and white matter tracts is essential to optimize outcomes in patients undergoing resection of intracranial lesions. While task-based functional MRI (tb-fMRI) and anatomical landmarks have traditionally been used for this purpose, they have notable limitations. Task-based paradigms require patient cooperation and intact function, which may not be possible in all patients. The integration of resting-state fMRI (rs-fMRI), which identifies intrinsic functional networks without requiring active participation, with diffusion tensor imaging (DTI)-based tractography, provides a comprehensive, non-invasive approach for delineating both cortical and subcortical structures. However, despite increasing research interest, this combined approach remains underutilised in clinical settings.

Objective:
 To assess the feasibility and clinical utility of integrating resting-state fMRI and DTI tractography for presurgical planning in patients with intracranial tumours and other space-occupying lesions, and to evaluate its impact on surgical decision-making, extent of resection, and postoperative outcomes.

Methods:
 We conducted a prospective observational study involving 10 patients (age range: 25–64 years) with supratentorial brain tumours or other mass lesions who underwent presurgical imaging at tertiary care hospitals. Imaging protocols included high-resolution T1-weighted MRI, diffusion tensor imaging (50 directions, b = 1000 s/mm²), and resting-state fMRI (8-10 minutes of acquisition, TR = 2–2.5 s, eyes closed). Preprocessing and analysis were performed using VoxelBox Explore (BrainSightAI, Pvt. Limited). Independent component analysis (ICA) was used to delineate key resting-state networks, including the sensorimotor, language, auditory, visual, and five cognitive networks. Probabilistic tractography was performed, using both anatomical landmarks and functionally defined ROIs derived from rs-fMRI networks.

Lesions were segmented, and fused 3D maps were generated to visualise their relationship with functional networks and white matter tracts. Neurosurgeons reviewed these multimodal maps during the surgical planning process. The impact on surgical approach, extent of resection, and neurological outcome was documented.

Results:
 The integration of rs-fMRI and DTI was feasible in all 10 patients. Relevant resting-state networks were successfully identified in all patients, including those with neurologically preserved conditions and those with specific deficits. The use of rs-fMRI and DTI overlayed structural images enabled more accurate delineation of the at-risk tracts. Based on the pre-surgical data, Gross total resection was achieved in 7 patients (64.4%) without new permanent deficits. In the remaining patients, subtotal resection was planned to preserve eloquent function based on imaging findings. There were no cases of unexpected functional deterioration postoperatively. The addition of rs-fMRI-derived maps to DTI improved surgeon confidence and enhanced preoperative discussions with patients regarding risk.

Conclusion:
 The combination of resting-state fMRI and diffusion tractography provides a powerful, non-invasive tool for individualised functional and structural mapping in the presurgical evaluation of brain lesions. It offers significant advantages over traditional task-based paradigms and anatomical seeding, particularly in challenging clinical scenarios. This multimodal approach enhances the surgeon’s ability to maximise resection while minimising functional risk, ultimately improving patient outcomes. Wider clinical adoption of this integrated methodology, supported by emerging AI-based automation tools, could transform the landscape of image-guided neurosurgery.