Segmentation of coronary calcifications with a domain knowledge-based lightweight 3D convolutional neural network

• Published in: Computers in biology and medicine Vol. 196; no. Pt B; p. 110798
• Main Authors: Santos, Rui, Castro, Rui, Baeza, Rúben, Nunes, Fábio, Filipe, Vítor M., Renna, Francesco, Paredes, Hugo, Fontes-Carvalho, Ricardo, Pedrosa, João
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.09.2025
• Subjects: 3D CNN; Calcinosis - diagnostic imaging; Calcium scoring; Cardiac CT; Convolutional Neural Networks; Coronary Artery Disease - diagnostic imaging; Coronary calcification; Coronary Vessels - diagnostic imaging; Dilated convolution; Female; Humans; Imaging, Three-Dimensional - methods; Internal Medicine; Lightweight; Male; Middle Aged; Other; Tomography, X-Ray Computed; Vascular Calcification - diagnostic imaging; 3D CNN; Coronary calcification; Dilated convolution; Cardiac CT; Lightweight; Calcium scoring
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2025.110798

Cardiovascular diseases are the leading cause of death in the world, with coronary artery disease being the most prevalent. Coronary artery calcifications are critical biomarkers for cardiovascular disease, and their quantification via non-contrast computed tomography is a widely accepted and heavily employed technique for risk assessment. Manual segmentation of these calcifications is a time-consuming task, subject to variability. State-of-the-art methods often employ convolutional neural networks for an automated approach. However, there is a lack of studies that perform these segmentations with 3D architectures that can gather important and necessary anatomical context to distinguish the different coronary arteries. This paper proposes a novel and automated approach that uses a lightweight three-dimensional convolutional neural network to perform efficient and accurate segmentations and calcium scoring. Results show that this method achieves Dice score coefficients of 0.93 ± 0.02, 0.93 ± 0.03, 0.84 ± 0.02, 0.63 ± 0.06 and 0.89 ± 0.03 for the foreground, left anterior descending artery (LAD), left circumflex artery (LCX), left main artery (LM) and right coronary artery (RCA) calcifications, respectively, outperforming other state-of-the-art architectures. An external cohort validation also showed the generalization of this method’s performance and how it can be applied in different clinical scenarios. In conclusion, the proposed lightweight 3D convolutional neural network demonstrates high efficiency and accuracy, outperforming state-of-the-art methods and showcasing robust generalization potential. •3D lightweight CNN with a domain knowledge-based starting point to accurately segment coronary calcifications.•Methodology that outperforms state-of-the-art architectures.•External validation dataset that demonstrates model’s generalization to unseen cases.