KI-Traktographie verbessert Bildgebung im Gehirn: Studie

The image shows two axial MRI brain scans overlaid with tractography visualizations of white matter fiber tracts. The left scan displays a more intact brain with extensive fiber tracts extending in multiple directions, visualized in different colors indicating fiber orientation. The right scan, however, exhibits a large region of missing brain tissue, likely due to a surgical intervention such as a hemispherotomy. The fiber tracts in this hemisphere are significantly reduced, with disrupted connections and fewer streamlines, reflecting the structural changes caused by the procedure. The image highlights differences in white matter organization between an intact and a surgically altered brain. Das Bild zeigt zwei axiale MRT-Gehirnscans mit überlagerten Traktographie-Visualisierungen von weißen Nervenbahnen. Der linke Scan zeigt ein weitgehend intaktes Gehirn mit umfangreichen Faserbahnen, die in verschiedene Richtungen verlaufen und durch unterschiedliche Farben zur Darstellung der Faserorientierung kodiert sind. Der rechte Scan weist hingegen eine große Region mit fehlendem Gehirngewebe auf, wahrscheinlich infolge eines chirurgischen Eingriffs wie einer Hemispherotomie. Die Faserbahnen in dieser Hemisphäre sind deutlich reduziert, mit unterbrochenen Verbindungen und weniger Stromlinien, was die durch den Eingriff verursachten strukturellen Veränderungen widerspiegelt. Das Bild verdeutlicht die Unterschiede in der Organisation der weißen Substanz zwischen einem intakten und einem operativ veränderten Gehirn.

How can nerve pathways in the brain be visualized to improve the planning of complex surgeries? A research team from the Lamarr Institute and the University of Bonn, in collaboration with the Translational Neuroimaging Group at the Departments of Neuroradiology and Epileptology at the University Hospital Bonn (UKB), has investigated an AI-powered method that makes these reconstructions more precise. The study, recently published in NeuroImage: Clinical, could ultimately help make neurosurgical procedures safer.

What Is Tractography – and Why Is It Important?

The brain is a highly complex network of nerve cells interconnected by delicate pathways – known as nerve fibers or tracts. These connections are essential for movement, speech, thought, and many other functions. To visualize these structures, researchers use tractography, an imaging technique that calculates the course of nerve pathways based on specialized MRI scans. This information is particularly crucial for planning brain surgeries, such as those performed on epilepsy patients undergoing surgical intervention.

Current tractography methods rely on mathematical models that infer the location of nerve pathways from MRI data. However, these methods often involve uncertainties, especially when the brain has been altered due to disease or surgery. This is where modern AI methods come into play: By leveraging machine learning, these systems can recognize patterns and generate more accurate reconstructions.

Fig. 1. A T1 weighted image of a patient after hemispherotomy. We investigate the ability of learning-based tractography to reconstruct tracts in the preserved hemisphere, despite the strong abnormalities in the affected one, and whether it erroneously reconstructs tracts as intact that have been surgically disconnected. © Grün, Bauer, Rüber, Schultz

AI-Powered Tractography Shows Potential – But Also Challenges

In the study, the researchers tested a widely used AI method called TractSeg, originally trained on healthy brains. The team investigated whether it could also work for epilepsy patients who had undergone a hemispherotomy – a surgical procedure that disconnects the two hemispheres of the brain.

The results showed that TractSeg performed well in many cases but also produced unexpected errors: It reconstructed nerve pathways that should no longer exist due to the surgery – a phenomenon known as „hallucination.“ At the same time, some remaining pathways were either incompletely captured or entirely missing from the reconstruction.

A New Hybrid Approach for More Accurate Reconstructions

To address these issues, the team developed a new hybrid method that combines the advantages of AI with the data fidelity of traditional techniques. This approach ensures that only existing nerve connections are reconstructed. The result: No more hallucinations, better detection of preserved pathways, and overall more accurate reconstructions – even in healthy brains.

Prof. Dr. Thomas Schultz, Principal Investigator in the Life Sciences at the Lamarr Institute and a professor at the Institute for Computer Science at the University of Bonn, highlights the significance of this work:

„Our study demonstrates both the potential and the limitations of AI-powered tractography in clinical applications. Combining AI with traditional methods offers a promising solution for more precise reconstructions, especially when dealing with patient data affected by pathological changes. Our goal is to further refine these approaches and make them applicable for neurosurgery in the long run.“

The image shows fiber tract reconstructions of the inferior longitudinal fasciculus (ILF) in the intact hemisphere of a hemispherotomy patient. The bottom row displays sagittal MRI brain scans with tractography overlays, where the ILF is highlighted in green, blue, and other colors. A red rectangle marks a region of interest. The top row presents zoomed-in views of this region, focusing on the differences between two reconstruction methods. On the left, TractSeg includes streamlines within a lesion, while on the right, the proposed regularized low-rank reconstruction excludes the lesion more accurately. The visualization highlights the tractography differences, with the right-side method maintaining a more precise exclusion of the lesion despite segmenting a larger overall tract volume. — Fig. 2. Reconstructions of the inferior longitudinal fasciculus (ILF) in the intact hemisphere of a hemispherotomy patient. In the close-up views, corresponding to the area within the red rectangle, streamlines have been clipped near the slice to better visualize the fact that TractSeg reconstructs streamlines within a lesion, whereas the proposed regularized low-rank reconstruction more accurately excludes the lesion, despite segmenting an overall larger tract volume. © Grün, Bauer, Rüber, Schultz

Funded by a TRA Research Award from the University of Bonn

This publication is the result of a collaboration funded by the Modeling for Life and Health research award from the University of Bonn’s Transdisciplinary Research Areas (TRA) Modeling and Life & Health. The project was conducted in partnership with PD Dr. Theodor Rüber, Principal Investigator of the Translational Neuroimaging Group and a neurologist at the UKB’s Department of Neuroradiology.

The University of Bonn’s Transdisciplinary Research Areas (TRA) serve as innovation and exploration hubs in research and education, enabling scholars to collaborate across disciplines and faculties, as well as with external partners, to tackle key scientific, technological, and societal challenges. This first joint publication illustrates how the exchange between AI research and neuroscience can help advance medical procedures – offering direct benefits for patients.

Additionally, the project was supported by the BNTrAinee program at the University of Bonn and the Neuro-aCSis Bonn Neuroscience Clinician Scientist Program.

Publication

Grün, Bauer, Rüber, Schultz: Deep Learning Based Tractography With TractSeg in Patients With Hemispherotomy: Evaluation and Refinement, NeuroImage: Clinical. https://doi.org/10.1016/j.nicl.2025.103738

lamarr institute person Schultz Thomas scaled e1663922506873 - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Prof. Dr. Thomas Schultz

Principal Investigator Life Sciences zum Profil

Themen

Life Sciences

KI-Traktographie verbessert die Bildgebung im Gehirn: Neue Studie