Clinically oriented automatic 2D liver tumor segmentation: LCMambaNet with a state-space model and liver cancer–specific attention

Key Points

• LCMambaNet achieves efficient liver tumor segmentation, enhancing clinical application and diagnosis.
• With a focus on small lesions, the model shows strong potential for aiding early diagnosis and treatment decisions.
• The approach balances computational efficiency with accuracy, addressing existing model limitations in tumor segmentation.
• This AI-based tool may enable more precise monitoring and treatment strategies for patients with liver tumors.

Abstract