Artificial Intelligence-based Liver Volume Measurement Using Preoperative and Postoperative CT Images

Area of Science:

• Medical Imaging
• Artificial Intelligence in Medicine
• Surgical Planning

Background:

• Accurate liver volumetry is essential for successful hepatectomy and patient recovery.
• Current methods for liver volume assessment can be time-consuming and subjective.
• Post-hepatectomy liver regeneration monitoring is critical for predicting outcomes.

Purpose of the Study:

• To develop and validate a deep learning system for automated liver volumetry.
• To assess the AI model's accuracy across multiple postoperative time points.
• To evaluate the potential of AI in tracking liver regeneration after hepatectomy.

Main Methods:

• A 3D U-Net deep learning model was trained on CT images.
• Data included preoperative, 7-day, and 3-month postoperative scans.
• Five-fold cross-validation was used to assess model performance.

Main Results:

• The AI model achieved a high mean Dice Similarity Coefficient (DSC) of 94.31%.
• Excellent volumetric accuracy was observed, with a mean difference of ~1% across time points.
• The model demonstrated consistent performance in liver volume measurement preoperatively and postoperatively.

Conclusions:

• The developed AI model provides accurate and automated liver volumetry for hepatectomy patients.
• This technology can significantly enhance surgical planning and postoperative patient monitoring.
• Further validation with larger datasets is recommended to confirm clinical and prognostic utility.