Recent advances in 3D fully convolutional networks (FCN) have made it feasible to produce dense voxel-wise predictions of volumetric images. In this work, we show that a multi-class 3D FCN trained on manually labeled CT scans of several anatomical structures (ranging from the large organs to thin vessels) can achieve competitive segmentation results, while avoiding the need for handcrafting features or training class-specific models. To this end, we propose a two-stage, coarse-to-fine approach that will first use a 3D FCN to roughly define a candidate region, which will then be used as input to a second 3D FCN. This reduces the number of voxels the second FCN has to classify to ∼10% and allows it to focus on more detailed segmentation of the organs and vessels. We utilize training and validation sets consisting of 331 clinical CT images and test our models on a completely unseen data collection acquired at a different hospital that includes 150 CT scans, targeting three anatomical organs (liver, spleen, and pancreas). In challenging organs such as the pancreas, our cascaded approach improves the mean Dice score from 68.5 to 82.2%, achieving the highest reported average score on this dataset. We compare with a 2D FCN method on a separate dataset of 240 CT scans with 18 classes and achieve a significantly higher performance in small organs and vessels. Furthermore, we explore fine-tuning our models to different datasets. Our experiments illustrate the promise and robustness of current 3D FCN based semantic segmentation of medical images, achieving state-of-the-art results..

Recent advances in deep learning, like 3D fully convolutional networks (FCNs), have improved the state-of-the-art in dense semantic segmentation of medical images. However, most network architectures require severely downsampling or cropping the images to meet the memory limitations of today's GPU cards while still considering enough context in the images for accurate segmentation. In this work, we propose a novel approach that utilizes auto-context to perform semantic segmentation at higher resolutions in a multi-scale pyramid of stacked 3D FCNs. We train and validate our models on a dataset of manually annotated abdominal organs and vessels from 377 clinical CT images used in gastric surgery, and achieve promising results with close to 90% Dice score on average. For additional evaluation, we perform separate testing on datasets from different sources and achieve competitive results, illustrating the robustness of the model and approach.

This paper presents an automated multi-organ segmentation method for 3D abdominal CT images based on a spatially-divided probabilistic atlases. Most previous abdominal organ segmentation methods are ineffective to deal with the large differences among patients in organ shape and position in local areas. In this paper, we propose an automated multi-organ segmentation method based on a spatiallydivided probabilistic atlas, and solve this problem by introducing a scale hierarchical probabilistic atlas. The algorithm consists of image-space division and a multi-scale weighting scheme. The generated spatial-divided probabilistic atlas efficiently reduces the inter-subject variance in organ shape and position either in global or local regions. Our proposed method was evaluated using 100 abdominal CT volumes with manually traced ground truth data. Experimental results showed that it can segment the liver, spleen, pancreas, and kidneys with Dice similarity indices of 95.1%, 91.4%, 69.1%, and 90.1%, respectively.

The authors identified the left FAT and confirmed that it was associated with language functions. This tract should be recognized by clinicians to preserve language function during brain tumor surgery, especially for tumors located in the deep frontal lobe on the language-dominant side.

This paper presents a new spatial fully connected tubular network for 3D tubular-structure segmentation. Automatic and complete segmentation of intricate tubular structures remains an unsolved challenge in the medical image analysis. Airways and vasculature pose high demands on medical image analysis as they are elongated fine structures with calibers ranging from several tens of voxels to voxel-level resolution, branching in deeply multiscale fashion, and with complex topological and spatial relationships. Most machine/deep learning approaches are based on intensity features and ignore spatial consistency across the network that are otherwise distinct in tubular structures. In this work, we introduce 3D slice-by-slice convolutional layers in a U-Net architecture to capture the spatial information of elongated structures. Furthermore, we present a novel loss function, coined radial distance loss, specifically designed for tubular structures. The commonly used methods of cross-entropy loss and generalized Dice loss are sensitive to volumetric variation. However, in tiny tubular structure segmentation, topological errors are as important as volumetric errors. The proposed radial distance loss places higher weight to the centerline, and this weight decreases along the the radial direction. Radial distance loss can help networks focus more attention on tiny structures than on thicker tubular structures. We perform experiments on bronchus segmentation on 3D CT images. The experimental results show that compared to the baseline U-Net, our proposed network achieved improvement about 24% and 30% in Dice index and centerline over ratio.

PBT-IACT for stage III-IVB tongue cancer has an acceptable toxicity profile and showed good treatment results. This protocol should be considered as a treatment option for locally advanced tongue cancer.

Retrograde superselective intra-arterial chemoradiotherapy for tongue cancer provided good overall survival and locoregional control.

BackgroundThe purpose of this study is to retrospectively evaluate the incidence of lung toxicities after proton beam therapy (PBT) in patients with idiopathic pulmonary fibrosis (IPF).MethodsPatients diagnosed with primary lung cancer or lung metastasis who were treated with PBT between January 2009 and May 2015 were recruited from our database retrospectively. Cases of pneumonitis (excluding infection-related pneumonitis) were evaluated using the Common Terminology Criteria for Adverse Events version 4.0, and the Fletcher-Hugh-Jones classification of respiratory status was used to evaluate pretreatment and posttreatment respiratory function.ResultsSixteen IPF patients received PBT for lung tumors, 15 received PBT for primary lung cancer, and one patient received PBT for metastasis from lung cancer. The cohort was composed of 14 men and 2 women, with a median age of 76 years (range: 63–89 years). The median follow-up time was 12 months (range: 4–39 months). The median dose of PBT was 80.0 Gy relative biological dose effectiveness (RBE) (range: 66.0–86.4 Gy [RBE]). The cumulative incidence of pneumonitis was 19.8 % (95 % confidence interval [CI]: 0–40.0 %), including one case of grade 5 pneumonitis. Reduced respiratory function was observed after PBT in seven patients, including one patient with pleural dissemination; five of these patients required home oxygen therapy.ConclusionsThis study suggests that PBT can be performed more safely in IPF patients than surgery or X-ray irradiation. Although PBT has become a treatment choice for lung tumors of patients with IPF, the adverse events warrant serious attention.