LambdaUNet: 2.5D Stroke Lesion Segmentation of Diffusion-Weighted MR Images

Ou, Yanglan; Yuan, Ye; Huang, Xiaolei; Wong, Kelvin; Volpi, John; Wang, James; Wong, Stephen T.C.

doi:10.1007/978-3-030-87193-2_69

Cited by 14 publications

(10 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, 3D models can converge faster during training because they can use the contextual information in the 3D image volume to segment each structure. 10 Conversely, 2.5D models can only use the contextual information in a few slices of the image, 11 and 2D models can only use the contextual information in one slice only. 12 Since the 3D approach provides more contextual information for each segmentation target, the complex shape of structures such as the hippocampus can be learned faster, and, as a result, convergence of 3D models can become faster.…”

Section: Discussionmentioning

confidence: 99%

“…Second, there are multiple ways to develop a 2.5D auto-segmentation model. 11,36,37 While we did not implement all the different versions of 2.5D models, we believe that our implementation of 2.5D models (using five consecutive image slices as input channels) is the best approach to segment the neuroanatomy on brain images. Third, our results about the relative deployment speed of 3D models as compared to 2.5D or 2D models might change as computational resources change.…”

Section: Discussionmentioning

confidence: 99%

“…Compared to two-dimensional auto-segmentation tasks, the three-dimensional (3D) nature of brain MRIs makes auto-segmentation considerably more challenging (Figure 1). There have been three proposed approaches to handling auto-segmentation of 3D images: 1) analyze and segment one two-dimensional slice of the volume individually one at a time (2D), 10 2) analyze five consecutive two-dimensional slices at a time generate a segmentation the middle slice (2.5D), 11 and 3) analyze and segment the image volume in three-dimensional space (3D). 10 Although each method has shown some promise in medical image segmentation, a comprehensive comparison and benchmarking of these approaches for auto-segmentation of brain MRIs is lacking.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Comparing 3D, 2.5D, and 2D Approaches to Brain Image Segmentation

Avesta¹,

Hossain

Lin

et al. 2022

Preprint

View full text Add to dashboard Cite

Deep-learning methods for auto-segmenting brain images either segment one slice of the image (2D), five consecutive slices of the image (2.5D), or an entire volume of the image (3D). Whether one approach is superior for auto-segmenting brain images is not known. We compared these three approaches (3D, 2.5D, and 2D) across three auto-segmentation models (capsule networks, UNets, and nnUNets) to segment brain structures. We used 3430 brain MRIs, acquired in a multi-institutional study, to train and test our models. We used the following performance metrics: segmentation accuracy, performance with limited training data, required computational memory, and computational speed during training and deployment. 3D, 2.5D, and 2D approaches respectively gave the highest to lowest Dice scores across all models. 3D models maintained higher Dice scores when the training set size was decreased from 3199 MRIs down to 60 MRIs. 3D models converged 20% to 40% faster during training and were 30% to 50% faster during deployment. However, 3D models require 20 times more computational memory compared to 2.5D or 2D models. This study showed that 3D models are more accurate, maintain better performance with limited training data, and are faster to train and deploy compared. However, 3D models require more computational memory compared to 2.5D or 2D models.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Comparing 3D, 2.5D, and 2D Approaches to Brain Image Segmentation

Avesta¹,

Hossain

Lin

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Yanglan et al (Ou et al, 2021) aimed to segmentation of stroke lesions using the diffusion-weighted images (DWI). In this study, 2.5D approach was considered due to the volumetric nature and interslice discontinuities of images.…”

Section: Related Workmentioning

confidence: 99%

İskemik İnme Lezyon Segmentasyonunda Segmentasyon Ağlarının ve Kayıp Fonksiyonlarının Etkilerinin Analizi

BAYRAM

DERİN

GURKAN

et al. 2022

EJOSAT

View full text Add to dashboard Cite

Stroke was the cause of one out of every six deaths from cerebrovascular disease in 2020. A stroke occurs in the United States (US) every 40 seconds. Every 3.5 minutes, people die of a stroke. More than total 795,000 stroke cases occur yearly in the US. This study aims to detect the ischemic stroke lesion that occurs in the brain. The Ischemic Stroke Lesion Segmentation (ISLES) 2017 data set, which includes 82 Magnetic Resonance images of 43 patients, was used. The UNet, Attention UNet, Residual UNet, Attention Residual UNet, and Residual UNet++ segmentation networks were tested. Moreover, Cross Entropy, Dice, IoU, Tversky, Focal Tversky, and their compound forms were analyzed. The IoU loss function tested on Attention UNet achieved the best performance with the dice score of 0.766, the IoU score of 0.621, the sensitivity of 0.730, the specificity of 0.997, the precision of 0.805, and the accuracy of 0.993.

show abstract

“…The success of Vision Transformers has attracted significant attention in the domain of medical image segmentation. For example, Swin-Unet [1], LambdaUNet [12], and U-NetR [5] replace convolutional layers with Transformers in a U-Net-like architecture. Other models like TransUNet [2], U-Net Transformer [13], and TransFuse [18] adopt a hybrid approach where they use Transformers to capture global context and convolutional layers to extract local features.…”

Section: Introductionmentioning

confidence: 99%

Patcher: Patch Transformers with Mixture of Experts for Precise Medical Image Segmentation

Ou¹,

Yuan²,

Huang³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

We present a new encoder-decoder Vision Transformer architecture, Patcher, for medical image segmentation. Unlike standard Vision Transformers, it employs Patcher blocks that segment an image into large patches, each of which is further divided into small patches. Transformers are applied to the small patches within a large patch, which constrains the receptive field of each pixel. We intentionally make the large patches overlap to enhance intra-patch communication. The encoder employs a cascade of Patcher blocks with increasing receptive fields to extract features from local to global levels. This design allows Patcher to benefit from both the coarse-to-fine feature extraction common in CNNs and the superior spatial relationship modeling of Transformers. We also propose a new mixture-of-experts (MoE) based decoder, which treats the feature maps from the encoder as experts and selects a suitable set of expert features to predict the label for each pixel. The use of MoE enables better specializations of the expert features and reduces interference between them during inference. Extensive experiments demonstrate that Patcher outperforms state-of-the-art Transformer-and CNN-based approaches significantly on stroke lesion segmentation and polyp segmentation. Code for Patcher will be released with publication to facilitate future research.

show abstract

LambdaUNet: 2.5D Stroke Lesion Segmentation of Diffusion-Weighted MR Images

Cited by 14 publications

References 12 publications

Comparing 3D, 2.5D, and 2D Approaches to Brain Image Segmentation

Comparing 3D, 2.5D, and 2D Approaches to Brain Image Segmentation

İskemik İnme Lezyon Segmentasyonunda Segmentasyon Ağlarının ve Kayıp Fonksiyonlarının Etkilerinin Analizi

Patcher: Patch Transformers with Mixture of Experts for Precise Medical Image Segmentation

Contact Info

Product

Resources

About