Carotid Ultrasound Boundary Study (CUBS): Technical considerations on an open multi-center analysis of computerized measurement systems for intima-media thickness measurement on common carotid artery longitudinal B-mode ultrasound scans

Meiburger, Kristen M.; Marzola, Francesco; Zahnd, Guillaume; Faita, Francesco; Loizou, Christos P.; Lainé, Nolann; Carvalho, Catarina; Steinman, David A.; Gibello, Lorenzo; Bruno, Rosa Maria; Clarenbach, Ricarda; Francesconi, Martina; Nicolaides, Andrew; Liebgott, Hervé; Campilho, Aurélio; Ghotbi, Reza; Kyriacou, Efthyvoulos; Navab, Nassir; Griffin, Maura; Panayiotou, Andrie G.; Gherardini, Rachele; Varetto, Gianfranco; Bianchini, Elisabetta; Pattichis, Constantinos S.; Ghiadoni, Lorenzo; Rouco, José; Orkisz, Maciej; Molinari, Filippo

doi:10.1016/j.compbiomed.2022.105333

Cited by 18 publications

(16 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These errors were compared with the inter-and intra-observer variabilities, A2 vs. A1 and A1' vs. A1, respectively. Table I summarizes the segmentation errors separately calculated for images from the CUBS 1 and CUBS 2 databases, which allows a comparison with other methods evaluated on these databases in previous studies [5], [6]. Thus, the previous benchmark on the CUBS 1 database was a MAD value on IMT equal to 114 ± 117 µm, while on the CUBS 2 database the best conventional method achieved 139 ± 119 µm and another Unet-based method obtained 178 ± 120 µm.…”

Section: Resultsmentioning

confidence: 98%

“…The latter achieved errors smaller than the interobserver variability, both in terms of estimated thickness and contour location. Compared to the methods already evaluated on the same data [5], [6], caroSegDeep established a new benchmark. As the proposed method is based on supervised learning, it has the potential to increase its performance by using larger and more diverse data for training.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Segmenting the carotid-artery wall in ultrasound image sequences with a dual-resolution U-net

Lainé

Zahnd

Liebgott

et al. 2022

2022 IEEE International Ultrasonics Symposium (IUS)

Self Cite

View full text Add to dashboard Cite

Thickening of intima-media complex in the common carotid artery is a biomarker of atherosclerosis. To automatically measure this thickness, we propose a region-based segmentation method, involving a supervised deep-learning approach based on the dilated U-net architecture, named caroSegDeep. It was trained and evaluated using 5-fold cross-validation on two open-access databases containing a total of 2676 annotated images. Compared with the methods already evaluated on these databases, caroSegDeep established a new benchmark and achieved a mean absolute error twice smaller than the interobserver variability.

show abstract

Section: Resultsmentioning

confidence: 98%

Section: Discussionmentioning

confidence: 99%

Segmenting the carotid-artery wall in ultrasound image sequences with a dual-resolution U-net

Lainé

Zahnd

Liebgott

et al. 2022

2022 IEEE International Ultrasonics Symposium (IUS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Two previously published and freely downloadable datasets [6], [8], were used as the training and validation sets. The resulting dataset consisted of 2576 B-mode longitudinal ultrasound images of the carotid artery.…”

Section: Dataset Descriptionmentioning

confidence: 99%

Can multiple segmentation methods enhance deep learning networks generalization? A novel hybrid learning paradigm

Marzola

Meiburger

Molinari

et al. 2023

Medical Imaging 2023: Computer-Aided Diagnosis

Self Cite

View full text Add to dashboard Cite

Deep learning methods are the state-of-the-art for medical imaging segmentation tasks. Still, numerous segmentation algorithms based on heuristic-based methods have been proposed with exceptional results. To validate segmentation algorithms, manual annotations are typically considered as ground truth. However, manual annotations often suffer from inter/intra-operator variability and can also be occasionally inaccurate, especially when considering time-consuming and precise tasks. A sample case is the manual delineation of the lumen-intima (LI) and media-adventitia (MA) borders for intima-media thickness (IMT) measurement in B-mode ultrasound images. In this work, a novel hybrid learning paradigm which combines manual segmentations with the automatic segmentation of a dynamic programming technique for ground truth determination is presented. A profile consensus strategy is proposed to construct the hybrid ground truth. Two open-source datasets (n=2576) were employed for training four deep learning networks using the hybrid learning paradigm and three single source training targets as a comparison. The pipeline was fixed across the four tests and included a Faster R-CNN detection network to locate the carotid artery and then subsequent division into patches which were segmented using a UNet. The validation of the results was performed on an external test set comparing the predictions of the four different models to the annotations of three independent manual operators. The hybrid learning paradigm showed the best overall segmentation results (Dice=0.907±0.037, p<0.001) and demonstrated an exceptional correlation between the mean of three operators and the automatic measure (ICC(2,1)=0.958), demonstrating how the incorporation of heuristic-based segmentation methods within the learning paradigm of a deep neural network can enhance and improve final segmentation performance results.

show abstract

“…We used a subset (n = 769) of in vivo images from the CUBS 1 [10] and CUBS 2 [9] databases. As the images also contain text and other graphical inlays, we clipped them out and cropped each sample down to only a region of interest (ROI).…”

Section: Gan-based Post Processing Phase a Datasetmentioning

confidence: 99%

Generation of realistic simulated B-mode image texture with a GAN

Lainé

Zahnd

Bernard

et al. 2022

2022 IEEE International Ultrasonics Symposium (IUS)

Self Cite

View full text Add to dashboard Cite

The intima-media complex of the common carotid artery is considered the sentinel of a silent killer disease called atherosclerosis. Morphological biomarkers such as the intimamedia thickness are already exploitable, but dynamic biomarkers, which reflect tissue deformation over the cardiac cycle, remain to be validated. Recent motion estimation methods seek to quantify compression, shear, and elongation coefficients, but their clinical applicability has not yet been well defined, and their actual accuracy is difficult to assess due to the absence of ground truth. This lack of reference also is the main limitation to explore fully supervised deep learning methods that have shown great potential in other applications. With this in mind, we propose a simulation pipeline to produce realistic in silico sequences, by combining a physics-based simulator with a post-processing based on a generative adversarial network.

show abstract

Carotid Ultrasound Boundary Study (CUBS): Technical considerations on an open multi-center analysis of computerized measurement systems for intima-media thickness measurement on common carotid artery longitudinal B-mode ultrasound scans

Cited by 18 publications

References 46 publications

Segmenting the carotid-artery wall in ultrasound image sequences with a dual-resolution U-net

Segmenting the carotid-artery wall in ultrasound image sequences with a dual-resolution U-net

Can multiple segmentation methods enhance deep learning networks generalization? A novel hybrid learning paradigm

Generation of realistic simulated B-mode image texture with a GAN

Contact Info

Product

Resources

About