Accuracy Evaluation of Artificial Intelligence Assisted Liver Tumor Ablation Path Planning

Accuracy Evaluation of Artificial Intelligence Assisted Liver Tumor Ablation Path Planning

CT-guided microwave ablation is one of the important treatment methods for primary or metastatic liver tumors. It has the advantages of minimally invasive, safe, and economical. For liver tumors with a diameter of less than 3 cm, it can be comparable to surgical resection. However, different doctors have different levels of experience, leading to a high risk of residues after ablation and even complications. The key is to fail to plan the needle placement reasonably. The past research on path planning was mostly conducted by the engineering team, so that the actual clinical needs were not considered comprehensively, and the sample size of verified cases was small, which limited the application of existing models in the clinical application. In previous study, we used artificial intelligence to segment the upper abdominal organs and construct the constraint function to establish the ablation path planning model, and initially verified the effectiveness of the path planning model in a small number of cases. Now it is planned to expand the number of patients enrolled to evaluate the accuracy of the ablation path planning model that we have established.

Inclusion Criteria:

1. 18-70 years old patients with liver tumors, not limited to primary liver cancer or secondary liver metastases;
2. The number of tumors is less than 5;
3. The diameter of the tumor is less than 3 cm;
4. No local treatment of intrahepatic tumors has been done in the past.

Exclusion Criteria:

1. Diffuse liver tumor;
2. Multiple lesions in the liver> 3cm, which cannot be completely ablated;
3. Patients with lymph node metastasis and distant metastasis outside the liver;
4. Other situations where the researcher believes that it is necessary to withdraw from the study