Toward Autonomous Histotripsy: Integrating Deep Learning Segmentation With Robotic Control for Glioblastoma

Area of Science:

• Neurosurgery
• Medical Robotics
• Artificial Intelligence

Background:

• Glioblastoma multiforme (GBM) surgery faces challenges with incomplete margin delineation, leading to residual disease or healthy tissue damage.
• Automated tumor delineation and guided ablation are crucial for improving surgical outcomes in GBM treatment.

Purpose of the Study:

• To develop and evaluate an AI- and robotics-enabled closed-loop framework for automated GBM tumor delineation and histotripsy ablation.
• To enhance surgical precision and minimize damage to healthy tissue during GBM resection.

Main Methods:

• Deep learning models were trained on intra-operative ultrasound data for real-time tumor segmentation.
• The segmentation output was integrated with robotic control to guide histotripsy targeting.
• The framework was validated in preclinical mouse GBM models (ex vivo and in vivo).

Main Results:

• AI models demonstrated strong real-time tumor segmentation performance.
• Histotripsy targeting demonstrated good alignment with the intended treatment zones in both ex vivo and in vivo experiments.
• Minimal undershooting was observed during histotripsy ablation, indicating precise targeting.

Conclusions:

• The study supports the feasibility of combining AI-based ultrasound segmentation with robotic guidance for histotripsy targeting in preclinical GBM models.
• This integrated approach represents a significant step toward increased automation in GBM treatment.
• The framework shows promise for improving the safety and efficacy of GBM surgery.