A Tversky Loss-Based Convolutional Neural Network for Liver Vessels Segmentation

Altini, Nicola; Prencipe, Berardino; Brunetti, Antonio; Brunetti, Giuseppe; Triggiani, Vito; Carnimeo, Leonarda; Marino, F.; Guerriero, Andrea; Villani, Laura; Scardapane, Arnaldo; Cascarano, Giacomo Donato

doi:10.1007/978-3-030-60799-9_30

Cited by 11 publications

(8 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For major details, surveys about U-shaped architectures and semantic segmentation approaches can be considered [25,26]. Applications of DL and semantic segmentation in the medical imaging include glomeruli segmentation from kidney biopsies [27,28], autosomal dominant polycystic kidney disease segmentation from magnetic resonance images [29], liver and vessels segmentation from CT scans [30] among others. In order to ensure the reproducibility of the algorithms introduced in Sections 4.1 and 4.2, and the visualization tool presented in Section 4.3, the code has been made publicly available on GitHub (https://github.com/Nicolik/Segm_Ident_Vertebrae_ CNN_kmeans_knn, last accessed: 6 June 2021).…”

Section: Methodsmentioning

confidence: 99%

“…The binary segmentation stage is performed by exploiting the V-Net architecture proposed by Milletari et al [24]. The Dice loss function formulation adopted is the same considered in the work of Altini et al [30]. V-Net is an encoder-decoder architecture.…”

Section: Spine Segmentationmentioning

confidence: 99%

“…Compared to U-Net, it is worth noting that V-Net exploits down-convolutions, with stride 2 × 2 × 2 and kernel-size 2 × 2 × 2, instead of a fixed downsampling realized by 2 × 2 × 2 maxpooling, adding learnable parameters also in this stage. As differences between original V-Net architecture and that implemented for this paper, there is the adoption of ReLU non-linearities instead of PReLU ones [32] and the adding of a Batch-Normalization (BN) layer after each convolutional layer, as also done in Altini et al [30] and Shen et al [33]. The network has been trained for 500 epochs with data augmentation (random noise with µ = 0, σ 2 ∈ [0, 0.1], p = 0.1).…”

Section: Spine Segmentationmentioning

confidence: 99%

See 2 more Smart Citations

Segmentation and Identification of Vertebrae in CT Scans Using CNN, k-Means Clustering and k-NN

et al. 2021

Self Cite

View full text Add to dashboard Cite

The accurate segmentation and identification of vertebrae presents the foundations for spine analysis including fractures, malfunctions and other visual insights. The large-scale vertebrae segmentation challenge (VerSe), organized as a competition at the Medical Image Computing and Computer Assisted Intervention (MICCAI), is aimed at vertebrae segmentation and labeling. In this paper, we propose a framework that addresses the tasks of vertebrae segmentation and identification by exploiting both deep learning and classical machine learning methodologies. The proposed solution comprises two phases: a binary fully automated segmentation of the whole spine, which exploits a 3D convolutional neural network, and a semi-automated procedure that allows locating vertebrae centroids using traditional machine learning algorithms. Unlike other approaches, the proposed method comes with the added advantage of no requirement for single vertebrae-level annotations to be trained. A dataset of 214 CT scans has been extracted from VerSe’20 challenge data, for training, validating and testing the proposed approach. In addition, to evaluate the robustness of the segmentation and labeling algorithms, 12 CT scans from subjects affected by severe, moderate and mild scoliosis have been collected from a local medical clinic. On the designated test set from Verse’20 data, the binary spine segmentation stage allowed to obtain a binary Dice coefficient of 89.17%, whilst the vertebrae identification one reached an average multi-class Dice coefficient of 90.09%. In order to ensure the reproducibility of the algorithms hereby developed, the code has been made publicly available.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Spine Segmentationmentioning

confidence: 99%

Section: Spine Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Segmentation and Identification of Vertebrae in CT Scans Using CNN, k-Means Clustering and k-NN

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this work, in order to realize the segmentation of lung parenchyma and lesions, we trained two CNN models based on the V-Net architecture, but considering the 2.5D and 2D variants with the same Dice loss formulation provided by Altini et al [36,37]. For the task of lesion segmentation, two classes were considered: GGO and LC, as also detailed in Section 2.…”

Section: Semantic Segmentationmentioning

confidence: 99%

“…The most common ones are the maximum symmetric surface distance (MSSD or symmetric Hausdorff distance) and the average symmetric surface distance (ASSD). Interested readers could refer to [36,[49][50][51] for further exploration.…”

Section: Segmentation Evaluation Metricsmentioning

confidence: 99%

Lung Segmentation and Characterization in COVID-19 Patients for Assessing Pulmonary Thromboembolism: An Approach Based on Deep Learning and Radiomics

et al. 2021

Self Cite

View full text Add to dashboard Cite

The COVID-19 pandemic is inevitably changing the world in a dramatic way, and the role of computed tomography (CT) scans can be pivotal for the prognosis of COVID-19 patients. Since the start of the pandemic, great care has been given to the relationship between interstitial pneumonia caused by the infection and the onset of thromboembolic phenomena. In this preliminary study, we collected n = 20 CT scans from the Polyclinic of Bari, all from patients positive with COVID-19, nine of which developed pulmonary thromboembolism (PTE). For eight CT scans, we obtained masks of the lesions caused by the infection, annotated by expert radiologists; whereas for the other four CT scans, we obtained masks of the lungs (including both healthy parenchyma and lesions). We developed a deep learning-based segmentation model that utilizes convolutional neural networks (CNNs) in order to accurately segment the lung and lesions. By considering the images from publicly available datasets, we also realized a training set composed of 32 CT scans and a validation set of 10 CT scans. The results obtained from the segmentation task are promising, allowing to reach a Dice coefficient higher than 97%, posing the basis for analysis concerning the assessment of PTE onset. We characterized the segmented region in order to individuate radiomic features that can be useful for the prognosis of PTE. Out of 919 extracted radiomic features, we found that 109 present different distributions according to the Mann–Whitney U test with corrected p-values less than 0.01. Lastly, nine uncorrelated features were retained that can be exploited to realize a prognostic signature.

show abstract

Uncertainty-Guided Pixel-Level Contrastive Learning for Biomarker Segmentation in OCT Images

Bai

Liu

et al. 2021

Intelligent Computing Theories and Application

View full text Add to dashboard Cite

A Tversky Loss-Based Convolutional Neural Network for Liver Vessels Segmentation

Cited by 11 publications

References 36 publications

Segmentation and Identification of Vertebrae in CT Scans Using CNN, k-Means Clustering and k-NN

Segmentation and Identification of Vertebrae in CT Scans Using CNN, k-Means Clustering and k-NN

Lung Segmentation and Characterization in COVID-19 Patients for Assessing Pulmonary Thromboembolism: An Approach Based on Deep Learning and Radiomics

Uncertainty-Guided Pixel-Level Contrastive Learning for Biomarker Segmentation in OCT Images

Contact Info

Product

Resources

About