3D liver segmentation in preoperative CT images using a level-sets active surface method.

Area of Science:

• Medical Imaging
• Computational Anatomy
• Radiology

Background:

• Accurate segmentation of abdominal organs, particularly the liver, is crucial for surgical planning and disease assessment.
• Manual segmentation of liver volumes from CT scans is time-consuming and subject to inter-observer variability.

Purpose of the Study:

• To develop and validate an automated active surface method for segmenting complete liver volumes from preoperative CT abdominal images.
• To evaluate the accuracy and efficiency of the proposed method compared to manual segmentations by radiologists.

Main Methods:

• An active surface segmentation algorithm was developed, minimizing an energy function that incorporates image intensity, gradient, and curvature information.
• The implementation utilized a level set technique with a multi-resolution strategy for computational efficiency.
• The method requires only a single user-defined seed point for liver initialization.

Main Results:

• The active surface method successfully segmented complete liver volumes across a validation set of preoperative CT scans.
• Segmentation accuracy was numerically evaluated against manual delineations by experienced radiologists.
• Radiologists assessed the segmented volumes, indicating the potential for reduced manual correction needs.

Conclusions:

• The proposed active surface method offers an efficient and accurate approach for automated liver volume segmentation from CT images.
• This technique has the potential to streamline preoperative planning and improve consistency in liver volume assessment.
• Further validation in diverse clinical scenarios is warranted to establish its widespread applicability.