Encoder-Decoder Architectures for Clinically Relevant Coronary Artery Segmentation

Silva, João Lourenço; Menezes, Miguel Nobre; Rodrigues, Tiago; Silva, Beatríz; Pinto, Fausto J.; Oliveira, Arlindo L.

doi:10.1007/978-3-031-17531-2_6

Cited by 7 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The application of deep learning has considerably improved ICA segmentation performance. [17][18][19] U-Net architecture 20 has been applied to segmentation of the three major vessels using a large database of ICA images, exhibiting a dice similarity coefficient (DSC) of 91.7% 14 and a nested encoder-decoder architecture has been proposed for ICA images. 13 Furthermore, segmentation architecture has been integrated with a preprocessing network as trained filters for boundary sharpening and contrast enhancement of ICA.…”

Section: Deep Learning-based Segmentationmentioning

confidence: 99%

“…The application of deep learning has considerably improved ICA segmentation performance 17–19 . U‐Net architecture 20 has been applied to segmentation of the three major vessels using a large database of ICA images, exhibiting a dice similarity coefficient (DSC) of 91.7% 14 and a nested encoder‐decoder architecture has been proposed for ICA images 13 .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Selective ensemble methods for deep learning segmentation of major vessels in invasive coronary angiography

et al. 2023

View full text Add to dashboard Cite

BackgroundInvasive coronary angiography (ICA) is a primary imaging modality that visualizes the lumen area of coronary arteries for diagnosis and interventional guidance. In the current practice of quantitative coronary analysis (QCA), semi‐automatic segmentation tools require labor‐intensive and time‐consuming manual correction, limiting their application in the catheterization room.PurposeThis study aims to propose rank‐based selective ensemble methods that improve the segmentation performance and reduce morphological errors that limit fully automated quantification of coronary artery using deep‐learning segmentation of ICA.MethodsTwo selective ensemble methods proposed in this work integrated the weighted ensemble approach with per‐image quality estimation. The segmentation outcomes from five base models with different loss functions were ranked either by mask morphology or estimated dice similarity coefficient (DSC). The final output was determined by imposing different weights according to the ranks. The ranking criteria based on mask morphology were formulated from empirical insight to avoid frequent types of segmentation errors (MSEN), while the estimation of DSCs was performed by comparing the pseudo‐ground truth generated from a meta‐learner (ESEN). Five‐fold cross‐validation was performed with the internal dataset of 7426 coronary angiograms from 2924 patients, and prediction model was externally validated with 556 images of 226 patients.ResultsThe selective ensemble methods improved the segmentation performance with DSCs up to 93.07% and provided a better delineation of coronary lesion with local DSCs of up to 93.93%, outperforming all individual models. Proposed methods also minimized the chances of mask disconnection in the most narrowed regions to 2.10%. The robustness of the proposed methods was also evident in the external validation. Inference time for major vessel segmentation was approximately one‐sixth of a second.ConclusionProposed methods successfully reduced morphological errors in the predicted masks and were able to enhance the robustness of the automatic segmentation. The results suggest better applicability of real‐time QCA‐based diagnostic methods in routine clinical settings.

show abstract

Section: Deep Learning-based Segmentationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Selective ensemble methods for deep learning segmentation of major vessels in invasive coronary angiography

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Hence, as the computational complexity of the decoder grows, EfficientNet can build a more computationally efficient system. When using a complex decoder like U-Net

, which needs to perform a lot of operations on the extracted features, EfficientNet has a more significant advantage over other encoders [ 11 ]. Furthermore, to avoid the difficulty of finding suitable regularization parameters when training from scratch, we initialize the EfficientNet network with ImageNet weights to train the model faster and more accurately.…”

Section: Methodsmentioning

confidence: 99%

Brain tumor segmentation based on the U-NET+⁣+ network with efficientnet encoder

Chen,

Quan,

Long

et al. 2024

THC

View full text Add to dashboard Cite

BACKGROUND: Brain tumor is a highly destructive, aggressive, and fatal disease. The presence of brain tumors can disrupt the brain’s ability to control body movements, consciousness, sensations, thoughts, speech, and memory. Brain tumors are often accompanied by symptoms like epilepsy, headaches, and sensory loss, leading to varying degrees of cognitive impairment in affected patients. OBJECTIVE: The study goal is to develop an effective method to detect and segment brain tumor with high accurancy. METHODS: This paper proposes a novel U-Net+⁣+ network using EfficientNet as the encoder to segment brain tumors based on MRI images. We adjust the original U-Net+⁣+ model by removing the dense skip connections between sub-networks to simplify computational complexity and improve model efficiency, while the connections of feature maps at the same resolution level are retained to bridge the semantic gap. RESULTS: The proposed segmentation model is trained and tested on Kaggle’s LGG brain tumor dataset, which obtains a satisfying performance with a Dice coefficient of 0.9180. CONCLUSION: This paper conducts research on brain tumor segmentation, using the U-Net+⁣+ network with EfficientNet as an encoder to segment brain tumors based on MRI images. We adjust the original U-Net+⁣+ model to simplify calculations and maintains rich semantic spatial features at the same time. Multiple loss functions are compared in this study and their effectiveness are discussed. The experimental results shows the model achieves a high segmention result with Dice coefficient of 0.9180.

show abstract

“…We have previously trained AI models for CAG segmentation based on manual CAG annotation of patients undergoing invasive physiology assessment (FFR and/or other indexes) or PCI, with an original sample of 416 images. 18,20 Recently, we published the results of our validation study with an additional dataset of 117 images. 19 Briefly, consecutive patients who had undergone (PCI) and/or invasive physiology assessment in four centers from across Portugal were selected.…”

Section: Previous Work and Study Populationmentioning

confidence: 99%

Segmentation of X‐ray coronary angiography with an artificial intelligence deep learning model: Impact in operator visual assessment of coronary stenosis severity

Nobre Menezes,

Silva,

Silva

et al. 2023

Cathet Cardio Intervent

Self Cite

View full text Add to dashboard Cite

BackgroundVisual assessment of the percentage diameter stenosis (%DSVE) of lesions is essential in coronary angiography (CAG) interpretation. We have previously developed an artificial intelligence (AI) model capable of accurate CAG segmentation. We aim to compare operators’ %DSVE in angiography versus AI‐segmented images.MethodsQuantitative coronary analysis (QCA) %DS (%DSQCA) was previously performed in our published validation dataset. Operators were asked to estimate %DSVE of lesions in angiography versus AI‐segmented images in separate sessions and differences were assessed using angiography %DSQCA as reference.ResultsA total of 123 lesions were included. %DSVE was significantly higher in both the angiography (77% ± 20% vs. 56% ± 13%, p < 0.001) and segmentation groups (59% ± 20% vs. 56% ± 13%, p < 0.001), with a much smaller absolute %DS difference in the latter. For lesions with %DSQCA of 50%–70% (60% ± 5%), an even higher discrepancy was found (angiography: 83% ± 13% vs. 60% ± 5%, p < 0.001; segmentation: 63% ± 15% vs. 60% ± 5%, p < 0.001). Similar, less pronounced, findings were observed for %DSQCA < 50% lesions, but not %DSQCA > 70% lesions. Agreement between %DSQCA/%DSVE across %DSQCA strata (<50%, 50%–70%, >70%) was approximately twice in the segmentation group (60.4% vs. 30.1%; p < 0.001). %DSVE inter‐operator differences were smaller with segmentation.Conclusion%DSVE was much less discrepant with segmentation versus angiography. Overestimation of %DSQCA < 70% lesions with angiography was especially common. Segmentation may reduce %DSVE overestimation and thus unwarranted revascularization.

show abstract

Encoder-Decoder Architectures for Clinically Relevant Coronary Artery Segmentation

Cited by 7 publications

References 36 publications

Selective ensemble methods for deep learning segmentation of major vessels in invasive coronary angiography

Selective ensemble methods for deep learning segmentation of major vessels in invasive coronary angiography

Brain tumor segmentation based on the U-NET+⁣+ network with efficientnet encoder

Segmentation of X‐ray coronary angiography with an artificial intelligence deep learning model: Impact in operator visual assessment of coronary stenosis severity

Contact Info

Product

Resources

About