Image-Free Tumor Segmentation of Soft Tissue Using a Minimally Invasive Robotic Palpation System

Area of Science:

• Robotics
• Surgical Technology
• Medical Imaging

Background:

• Tumor segmentation is vital for surgical planning and effective cancer treatment.
• Traditional methods like CT, MRI, and surgeon palpation have limitations in minimally invasive robotic surgery (MIS), including registration errors and subjectivity.
• Accurate tumor localization is essential for precise tumor resection.

Purpose of the Study:

• To introduce a novel robotic palpation system for image-free tumor segmentation in MIS.
• To enable precise tumor shape differentiation and stiffness mapping through robotic palpation.
• To overcome the limitations of traditional imaging and palpation methods in MIS.

Main Methods:

• Development of a robotic palpation system capable of collecting surface shape information.
• Implementation of a process for tissue palpation at specific depths based on surface curvature.
• Visualization of tissue stiffness maps for image-free tumor segmentation.
• Validation using planar and curved phantom models and ex-vivo porcine liver models.

Main Results:

• The robotic system achieved precise tumor shape differentiation and stiffness mapping.
• Evaluation on phantom models demonstrated high sensitivity (0.9634-0.9729) and specificity (0.9646-0.9878) for tumor segmentation.
• Ex-vivo validation confirmed the system's efficacy in real biological tissues.

Conclusions:

• The proposed robotic palpation system offers a precise and image-free solution for tumor segmentation in MIS.
• This technology has the potential to enhance surgical planning and improve tumor resection outcomes.
• Robotic palpation represents a significant advancement in overcoming current challenges in MIS tumor localization.