Related Experiment Video
Updated: Sep 2, 2025
06:39
Three-dimensional Location Approach with Silk Thread Guided Laparoscopic Segmentectomy for Liver Tumor
Published on: May 23, 2025
129
Automated Three-Dimensional Liver Reconstruction with Artificial Intelligence for Virtual Hepatectomy
Takeshi Takamoto1, Daisuke Ban2, Satoshi Nara2
1Department of Hepatobiliary and Pancreatic Surgery, National Cancer Center Hospital, 5-1-1 Tsukiji, Chuo-ku, Tokyo, 104-0045, Japan. ttakamot@ncc.go.jp.
Summary
Artificial intelligence (AI) significantly speeds up 3D liver reconstruction for virtual hepatectomy, improving planning for liver tumor resections. This AI-assisted method offers reliable volumetric analysis and faster processing times compared to manual techniques.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Medicine
- Surgical Planning
Background:
- A novel artificial intelligence (AI) tool has been developed utilizing a deep learning algorithm trained on healthy liver computed tomography (CT) images.
- This AI system is designed to aid in the three-dimensional (3D) reconstruction of the liver for virtual hepatectomy procedures.
Purpose of the Study:
- To validate a new AI-assisted simulation for liver tumor surgery planning.
- To evaluate the speed and accuracy of 3D liver reconstruction and segmental volumetry using the AI tool.
Main Methods:
- Three-dimensional (3D) reconstruction of liver CT scans from 144 patients with liver tumors was performed.
- Two methods were compared: a manual tracking algorithm and the AI-assisted method using Synapse 3D software.
- Processing time and volumetric accuracy of whole liver, tumor-containing, and tumor-free segments were analyzed.
Main Results:
- AI-assisted reconstruction was significantly faster, with a median processing time of 2.1 minutes compared to 35.0 minutes for the manual method (p < 0.001).
- Volumetric measurements showed comparable accuracy between AI-assisted and manual methods, with low mean absolute deviations for total liver and segmental volumes.
- Median total liver volume was 1035 mL (AI) vs. 1120 mL (manual); tumor-containing segment volume ratio was 9.4% (AI) vs. 9.9% (manual).
Conclusions:
- AI-assisted 3D liver models provide reliable functional volumetric analysis for virtual hepatectomy.
- The validated AI-assisted method is suitable for practical planning of liver tumor resections.
- The AI tool enhances efficiency and accuracy in pre-surgical liver assessment.
