HDC-Net: Hierarchical Decoupled Convolution Network for Brain Tumor Segmentation

Luo, Zhengrong; Jia, Zhongdao; Yuan, Zhimin; Peng, Jialin

doi:10.1109/jbhi.2020.2998146

Cited by 85 publications

(46 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All these methods were trained using the BraTS2018 training dataset and tested on the BraTS2018 validation dataset. The compared methods can be roughly categorized into 2D slicebased ones (17)(18)(19)(20) and 3D patch-based ones (22)(23)(24)(25)(26). Typically, 3D patch-based methods perform better than 2D slice-based Although a single 3D model such as HDC-Net can deliver promising results, most existing methods employ ensemble strategies to further improve the accuracy.…”

Section: Comparison With Other Methods On the Brats2018 Validation Datasetmentioning

confidence: 99%

“…The multi-modal Brain Tumor Segmentation (BraTS) challenge has released a large amount of pre-operative multi-modal MR images and the corresponding manual annotations of brain tumor (3)(4)(5)(6). Benefited from this large dataset, convolutional neural network (CNN) has quickly dominated the fully-automated brain tumor segmentation field (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29).…”

Section: Related Workmentioning

confidence: 99%

“…achieves the best Dice/H95 of ET (81.5%/2.42 mm) using the singe HDC-Net(25), but their Dice/H95 of TC (84.3%/8.76 mm) are relatively worse than other methods.…”

mentioning

confidence: 90%

See 2 more Smart Citations

Brain Tumor Segmentation From Multi-Modal MR Images via Ensembling UNets

et al. 2021

View full text Add to dashboard Cite

Glioma is a type of severe brain tumor, and its accurate segmentation is useful in surgery planning and progression evaluation. Based on different biological properties, the glioma can be divided into three partially-overlapping regions of interest, including whole tumor (WT), tumor core (TC), and enhancing tumor (ET). Recently, UNet has identified its effectiveness in automatically segmenting brain tumor from multi-modal magnetic resonance (MR) images. In this work, instead of network architecture, we focus on making use of prior knowledge (brain parcellation), training and testing strategy (joint 3D+2D), ensemble and post-processing to improve the brain tumor segmentation performance. We explore the accuracy of three UNets with different inputs, and then ensemble the corresponding three outputs, followed by post-processing to achieve the final segmentation. Similar to most existing works, the first UNet uses 3D patches of multi-modal MR images as the input. The second UNet uses brain parcellation as an additional input. And the third UNet is inputted by 2D slices of multi-modal MR images, brain parcellation, and probability maps of WT, TC, and ET obtained from the second UNet. Then, we sequentially unify the WT segmentation from the third UNet and the fused TC and ET segmentation from the first and the second UNets as the complete tumor segmentation. Finally, we adopt a post-processing strategy by labeling small ET as non-enhancing tumor to correct some false-positive ET segmentation. On one publicly-available challenge validation dataset (BraTS2018), the proposed segmentation pipeline yielded average Dice scores of 91.03/86.44/80.58% and average 95% Hausdorff distances of 3.76/6.73/2.51 mm for WT/TC/ET, exhibiting superior segmentation performance over other state-of-the-art methods. We then evaluated the proposed method on the BraTS2020 training data through five-fold cross validation, with similar performance having also been observed. The proposed method was finally evaluated on 10 in-house data, the effectiveness of which has been established qualitatively by professional radiologists.

show abstract

Section: Comparison With Other Methods On the Brats2018 Validation Datasetmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Brain Tumor Segmentation From Multi-Modal MR Images via Ensembling UNets

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Task curriculum learning consists of tackling easy but related tasks first to provide auxiliary information for more complicated tasks, which will be solved later. Task curriculum learning is highly related to multi-stage learning in segmentation [2], [143], [144], where more easier tasks such as location or coarse segmentation are first solved with a simple method. After that, the more complex pixel-level segmentation is addressed.…”

Section: Curriculum Learningmentioning

confidence: 99%

“…Typical data augmentation methods not only include data warping methods [62] such as random affine and elastic transformations, random cropping [50], random erasing [63], [64], intensity transformation and adversarial data augmentation [65], [66], but also include methods that can synthesize more diverse and realistic labeled examples, such as mixing images [67]- [70], feature space augmentation [71], and generative adversarial networks [18], [72]- [74]. While general transformation augmentation methods such as random affine transformations, elastic transformations, and intensity transformations are easy to implement and have shown performance improvements in abundant applications [2], [34], [75], they do not take advantage of the knowledge in unlabeled training data. Recently, there is a growing interest in developing augmentation that can simulate real variations of the data, and thus task-driven approach [76]- [78] is a promising direction.…”

Section: B Data Augmentationmentioning

confidence: 99%

Medical Image Segmentation With Limited Supervision: A Review of Deep Network Models

Wang

2021

IEEE Access

Self Cite

View full text Add to dashboard Cite

Despite the remarkable performance of deep learning methods on various tasks, most cutting-edge models rely heavily on large-scale annotated training examples, which are often unavailable for clinical and health care tasks. The labeling costs for medical images are very high, especially in medical image segmentation, which typically requires intensive pixel/voxel-wise labeling. Therefore, the strong capability of learning and generalizing from limited supervision, including a limited amount of annotations, sparse annotations, and inaccurate annotations, is crucial for the successful application of deep learning models in medical image segmentation. However, due to its intrinsic difficulty, segmentation with limited supervision is challenging and specific model design and/or learning strategies are needed. In this paper, we provide a systematic and up-to-date review of the solutions above, with summaries and comments about the methodologies. We also highlight several problems in this field, discussed future directions observing further investigations.

show abstract

Segmentation of brain tumor subregions with depthwise separable dense U‐NET (DSDU‐NET)

Nehru

Prabhu

2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Brain tumors are one of the most dangerous medical conditions. The dispersed extremities and nonuniform structure of the tumors are the basis of why techniques of traditional segmentation have grown to be inefficient. Magnetic resonance imaging (MRI) is one of the most widely used scanning procedures for tumors. However, to ameliorate the survival rate, the detection of tumors alone does not suffice, and there are other effective procedures. One of the most pivotal procedures for diagnosing the condition is the process of the brain tumor segmentation is a laborious and cumbersome process. As a result, a deep learning (DL) based solution is used to extract tumor subregions like enhancing tumor (ET), tumor core (TC), and whole tumor (WT). The proposed model involves novel DSDU‐Net: Depth‐wise separable dense U‐NET (DSDU‐NET) that precise outputs are acquired by retaining the low‐level features. In the preprocessing stage, a methodology called multi‐scale patch extraction is used to segregate tumor regions, and a grouping of Gaussian filter, unsharp masking, and histogram equalization is carried out on the data set to get significantly better performance. The proposed DSDU‐NET network performed better in terms of performance, specificity, sensitivity, dice similarity index, and Hausdorff distance for segmented image sub‐regions when validated on BraTS 2018 and 2019 data sources when particularly in comparison to other models.

show abstract

HDC-Net: Hierarchical Decoupled Convolution Network for Brain Tumor Segmentation

Cited by 85 publications

References 41 publications

Brain Tumor Segmentation From Multi-Modal MR Images via Ensembling UNets

Brain Tumor Segmentation From Multi-Modal MR Images via Ensembling UNets

Medical Image Segmentation With Limited Supervision: A Review of Deep Network Models

Segmentation of brain tumor subregions with depthwise separable dense U‐NET (DSDU‐NET)

Contact Info

Product

Resources

About