Objective To assess the value of precise hepatectomy in treatment of primary hepatocellular carcinoma. Methods Three-dimensional (3D) models from MR image were reconstructed by 3D-Doctor software in 32 patients with primary hepatocellular carcinoma scheduled for liver resection between July 2007 and Sept 2009. From these 3D models, the vena cava, portal vein, hepatic vein, and short hepatic vein images were reconstructed, total liver volume, tumor volume, functional liver volume and ratio of functional liver volume to standard liver volume (SFLVR) were calculated. The patients were followed-up for 1-27 months, with an average of 12 months. Results The anatomic detail of liver veins and its relationship with the tumor could be displayed clearly in liver 3D models. By the 3D models, total liver volume was calculated as （1 353±419）ml, tumor volume as （287±248） ml, functional liver volume as （830±289） ml, and SFLVR as （71±22）%. Of 32 patients with hepatocellular carcinoma, right hemihepatectomy was performed in 8 cases, left hemihepatectomy in 2, and segmental or limited resection in 22. All operations were completed successfully. Postoperative complications included pulmonary infection in 1 case, bile leak in 1, moderate ascites (500-3 000 ml) in 8, and massive ascites (gt;3 000 ml) in 2 including one patient developed hepatic failure. Six and 12-month survival rates were 100% and 87%. Three, 6, and 12-month disease-free survival rates were 78%, 72%, and 72%. Conclusions Precise hepatectomy technique provides an accurate picture of liver veins anatomy and its relationship with the tumor， and allows the procedure to be simulated preoperatively for adequate and safe hepatectomy.
Objective To investigate the diagnostic value of CT scanning and MR imaging on acute cholecystitis. Methods The CT or MR imaging data of 21 patients with proved acute cholecystitis were retrospectively reviewed. Eleven patients were examined with contrast-enhanced multi-detector-row spiral CT scanning and other 10 cases underwent contrast-enhanced MR imaging. Results Nineteen patients showed obscure gallbladder outlines （90.5%）. The gallbladder wall demonstrated even thickening in 15 patients （71.4%） and irregular thickening in 6 cases （28.6%）. All patients showed inhomogeneous enhancement of the gallbladder wall （100%）. The bile was hyper-dense or hyper-intense on T1W image in 11 cases （52.4%）. Ten cases had free peri-cholecystic effusion （47.6%）, and 16 cases had peri-cholecystic adhesive changes or fat swelling （76.2%）. Patchy or linear-like transient enhancement of the adjacent hepatic bed in the arterial phase was seen in 16 cases （76.2%）. Twelve patients developed pleural effusion, or ascites, or both （57.1%）. Gallbladder perforation complicated with peritonitis was seen in one case, micro-abscess formation and pneumocholecystitis was observed in another case, and one case had gallbladder diverticulum. Conclusion Wall blurring, pericholecystic adhesion or fat edema, and transient enhancement of adjacent hepatic bed in the arterial phase are the imaging findings specifically associated with acute cholecystitis, which are readily appreciated on contrast-enhanced multi-phasic CT and MR scanning.
When applying deep learning algorithms to magnetic resonance (MR) image segmentation, a large number of annotated images are required as data support. However, the specificity of MR images makes it difficult and costly to acquire large amounts of annotated image data. To reduce the dependence of MR image segmentation on a large amount of annotated data, this paper proposes a meta-learning U-shaped network (Meta-UNet) for few-shot MR image segmentation. Meta-UNet can use a small amount of annotated image data to complete the task of MR image segmentation and obtain good segmentation results. Meta-UNet improves U-Net by introducing dilated convolution, which can increase the receptive field of the model to improve the sensitivity to targets of different scales. We introduce the attention mechanism to improve the adaptability of the model to different scales. We introduce the meta-learning mechanism, and employ a composite loss function for well-supervised and effective bootstrapping of model training. We use the proposed Meta-UNet model to train on different segmentation tasks, and then use the trained model to evaluate on a new segmentation task, where the Meta-UNet model achieves high-precision segmentation of target images. Meta-UNet has a certain improvement in mean Dice similarity coefficient (DSC) compared with voxel morph network (VoxelMorph), data augmentation using learned transformations (DataAug) and label transfer network (LT-Net). Experiments show that the proposed method can effectively perform MR image segmentation using a small number of samples. It provides a reliable aid for clinical diagnosis and treatment.
The registration of preoperative magnetic resonance (MR) images and intraoperative ultrasound (US) images is very important in the planning of brain tumor surgery and during surgery. Considering that the two-modality images have different intensity range and resolution, and the US images are degraded by lots of speckle noises, a self-similarity context (SSC) descriptor based on local neighborhood information was adopted to define the similarity measure. The ultrasound images were considered as the reference, the corners were extracted as the key points using three-dimensional differential operators, and the dense displacement sampling discrete optimization algorithm was adopted for registration. The whole registration process was divided into two stages including the affine registration and the elastic registration. In the affine registration stage, the image was decomposed using multi-resolution scheme, and in the elastic registration stage, the displacement vectors of key points were regularized using the minimum convolution and mean field reasoning strategies. The registration experiment was performed on the preoperative MR images and intraoperative US images of 22 patients. The overall error after affine registration was (1.57 ± 0.30) mm, and the average computation time of each pair of images was only 1.36 s; while the overall error after elastic registration was further reduced to (1.40 ± 0.28) mm, and the average registration time was 1.53 s. The experimental results show that the proposed method has prominent registration accuracy and high computational efficiency.