IntroductionLiver surgery is widely used as a treatment modality for various liver pathologies. Despite significant improvement in clinical care, operative strategies and technology over the last few decades, liver surgery is still risky and optimal preoperative planning and anatomical assessment are necessary to minimize risks of serious complications. 3D printing technology is rapidly expanding and its applications in medicine are growing, but its applications in liver surgery are still limited. This article describes development of models of hepatic structures specific to a patient diagnosed with an operable hepatic malignancy.
MethodsAnatomy data was segmented and extracted form CT and MRI liver of a single patient with a resectable liver tumour. The digital data of the extracted anatomical surfaces was then edited and smoothed resulting in a set of digital 3D models of the hepatic vein, portal vein with tumour, biliary tree with gallbladder and hepatic artery. These were then 3D printed.
ResultsThe final models of the liver structures and tumour is provide good anatomical detail and representation of the spatial relationships between the liver tumour and adjacent hepatic structures. It can be easily manipulated and explored from different angles.
ConclusionsA graspable, patient specific, 3D printed models of liver structures could provide an improved understanding of the complex liver anatomy, better navigation in difficult areas and allow surgeons to anticipate anatomical issues that might arrive during the operation. Further research into adequate imaging, liver specific volumetric software, and segmentation algorithms are worth considering to optimize this application.
Methods
Data Extraction and SegmentationRetrospectively collected radiology image data from a patient with an operable malignant hepatic tumour consisted of a standard CT angiogram of abdomen and pelvis and MRI of liver performed using a standard hepatic imaging protocol with gadolinium contrast. The CT slides were 3mm thick and MRI slides were 8.99 mm thick, both yielding anisotropic voxels when viewed as 3D. The data of both scans was stored in Digital Imaging and Communications in Medicine (DICOM) files. Amira 4.5.4. visualisation software (FEI, Hillsboro, USA) was used to view and segment the data. All scans were interrogated in three planes and pixels containing image data for hepatic and portal veins, hepatic artery, biliary structures and tumour were manually selected. (Figure 1). Due to varying image quality between the two radiology modalities used, MRI data was used to segment the biliary tree, portal vein, hepatic veins and tumour, whilst the CT was used to collect data for the hepatic artery. Segmentation was completed with a combination of manual, and "region growing"techniques, where the latter was used for large regions of similar density signal.
Surgical InnovationMadurska, Poyade, Eason, Rea, Watson/ Development of patient specific 3D printed liver 5 model for preoperative planning
Surface extraction and model pr...
