Multi-Scale Deep Information and Adaptive Attention Mechanism Based Coronary Reconstruction of Superior Mesenteric Artery

Abstract

Vascular images contain a lot of key information, such as length, diameter and distribution. Thus reconstruction of vessels such as the Superior Mesenteric Artery is critical for the diagnosis of some abdominal diseases. However automatic segmentation of abdominal vessels is extremely challenging due to the multi-scale nature of vessels, boundary-blurring, low contrast, artifact disturbance and vascular cracks in Maximum Intensity Projection images. In this work, we propose a dual attention guided method where an adaptive adjustment field is applied to deal with multi-scale vessel information, and a channel feature fusion module is used to refine the extraction of thin vessels, reducing the interference and background noise. In particular, we propose a novel structure that accepts multiple sequential images as input, and successfully introduces spatial-temporal features by contextual information. A further IterUnet step is introduced to connect tiny cracks caused using CT scans. Comparing our proposed model with other state-of-the-art models, our model yields better segmentation and achieves an average F1 metric of 0.812.

PE-Net: a parallel framework for 3D inferior mesenteric artery segmentation

“…The CT scans used in the experiment were obtained from a Siemens dual-source CT scanner (Somatom Force, Siemens Healthcare, Forchheim, Germany). The specific CT acquisition parameters were consistent with our previous work (Zhang et al, 2023).…”

“… Overview of the structure of the proposed PE-Net: Two encoding layers consisting of CNN and TAGT, a skip connection layer consisting of EFC and FFB, and one decoding layer. The parallelly encoded features are fused using the Channel Attention-based method, as introduced in our previous work ( Zhang et al, 2023 )—a block named Feature Fusion Module (FFM). Before the operation, the features generated by TAGT are broadcasted to align their channel dimensions with the convolutional features.…”