Hybrid deep learning segmentation models for atherosclerotic plaque in internal carotid artery B-mode ultrasound

• Published in: Computers in biology and medicine Vol. 136; p. 104721
• Main Authors: Jain, Pankaj K., Sharma, Neeraj, Giannopoulos, Argiris A., Saba, Luca, Nicolaides, Andrew, Suri, Jasjit S.
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.09.2021; Elsevier Limited
• Subjects: Area; Arteriosclerosis; Artificial intelligence; Atherosclerosis; Automation; Cancer; Cardiovascular disease; Carotid arteries; Carotid artery; Cross-entropy; Deep learning; Dice similarity coefficient; Entropy; Health risks; Hybrid deep learning; Internal carotid artery; Magnetic resonance imaging; On-line systems; Performance evaluation; Plaque segmentation; Risk groups; Segmentation; Stroke; Ultrasonic imaging; Ultrasound; Hybrid deep learning; Area; Internal carotid artery; Cross-entropy; Plaque segmentation; Performance; Ultrasound; Dice similarity coefficient
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2021.104721

The automated and accurate carotid plaque segmentation in B-mode ultrasound (US) is an essential part of stroke risk stratification. Previous segmented methods used AtheroEdge™ 2.0 (AtheroPoint™, Roseville, CA) for the common carotid artery (CCA). This study focuses on automated plaque segmentation in the internal carotid artery (ICA) using solo deep learning (SDL) and hybrid deep learning (HDL) models. The methodology consists of a novel design of 10 types of SDL/HDL models (AtheroEdge™ 3.0 systems (AtheroPoint™, Roseville, CA) with a depth of four layers each. Five of the models use cross-entropy (CE)-loss, and the other five models use Dice similarity coefficient (DSC)-loss functions derived from UNet, UNet+, SegNet, SegNet-UNet, and SegNet-UNet+. The K10 protocol (Train:Test:90%:10%) was applied for all 10 models for training and predicting (segmenting) the plaque region, which was then quantified to compute the plaque area in mm2. Further, the data augmentation effect was analyzed. The database consisted of 970 ICA B-mode US scans taken from 99 moderate to high-risk patients. Using the difference area threshold of 10 mm2 between ground truth (GT) and artificial intelligence (AI), the area under the curve (AUC) values were 0.91, 0.911, 0.908, 0.905, and 0.898, all with a p-value of <0.001 (for CE-loss models) and 0.883, 0.889, 0.905, 0.889, and 0.907, all with a p-value of <0.001 (for DSC-loss models). The correlations between the AI-based plaque area and GT plaque area were 0.98, 0.96, 0.97, 0.98, and 0.97, all with a p-value of <0.001 (for CE-loss models) and 0.98, 0.98, 0.97, 0.98, and 0.98 (for DSC-loss models). Overall, the online system performs plaque segmentation in less than 1 s. We validate our hypothesis that HDL and SDL models demonstrate comparable performance. SegNet-UNet was the best-performing hybrid architecture. [Display omitted] •Carotid plaque and wall segmentation and quantification using ten types of solo and hybrid deep learning models.•Ten deep learning models: six SDL models and four HDL models.•DL-models with CE-loss are better than the DL-models with DSC-loss.•The mean correlation coefficient between the AI and GT plaque area were 0.972 (for CE-loss) and 0.978 (for DSC-loss models) having p-value<0.00.