Identification of Liver Lobes Based on Anatomic Knowledge: With CT during Arterial Portography

Patients with certain types of malignant hepatic tumors have been shown to have an improved long-term survival rate, if they undergo the hepatic resection of lesions with sufficient tumor-free margins. Preoperatively, it is important that certain vital information, which includes the number and sizes of lesions, the segmental extent and locations, and the relationship of the tumors to the hepatic vasculature, be captured for determining the resectability and planning the type of resection. In general, this information was provided by the experienced radiologists based on the examination of a sequence of CTAP image slices of patients. In this paper, anatomical knowledge in depicting the liver segments was realized into a computerized method, in order that the liver segmental description can be put into a computer system. The anatomical features used in the system include the positions of gallbladder, IVC, portal veins, hepatic veins, and ligaments, which are usually adopted by doctors in delineating the liver segments. Image processing techniques are applied to extract these anatomical features. The proposed method makes the idea of using computer for help in identifying focal lesions to specific hepatic segments possible, which would provide radiologists and surgeons with significant assistant in depicting the liver anatomical segments.