DSIFNet: Implicit feature network for nasal cavity and vestibule segmentation from 3D head CT

• Published in: Computerized medical imaging and graphics 2024-12, Vol.118, p.102462, Article 102462
• Main Authors: Lu, Yi, Gao, Hongjian, Qiu, Jikuan, Qiu, Zihan, Liu, Junxiu, Bai, Xiangzhi
• Format: Article
• Language: English
• Subjects: Deep supervision; Head - diagnostic imaging; Humans; Imaging, Three-Dimensional - methods; Implicit Feature function; Nasal Cavity - anatomy & histology; Nasal Cavity - diagnostic imaging; Nasal cavity segmentation; Nasal vestibule segmentation; Neural Networks, Computer; Tomography, X-Ray Computed - methods
• ISSN: 0895-6111; 1879-0771; 1879-0771
• DOI: 10.1016/j.compmedimag.2024.102462

This study is dedicated to accurately segment the nasal cavity and its intricate internal anatomy from head CT images, which is critical for understanding nasal physiology, diagnosing diseases, and planning surgeries. Nasal cavity and it’s anatomical structures such as the sinuses, and vestibule exhibit significant scale differences, with complex shapes and variable microstructures. These features require the segmentation method to have strong cross-scale feature extraction capabilities. To effectively address this challenge, we propose an image segmentation network named the Deeply Supervised Implicit Feature Network (DSIFNet). This network uniquely incorporates an Implicit Feature Function Module Guided by Local and Global Positional Information (LGPI-IFF), enabling effective fusion of features across scales and enhancing the network's ability to recognize details and overall structures. Additionally, we introduce a deep supervision mechanism based on implicit feature functions in the network's decoding phase, optimizing the utilization of multi-scale feature information, thus improving segmentation precision and detail representation. Furthermore, we constructed a dataset comprising 7116 CT volumes (including 1,292,508 slices) and implemented PixPro-based self-supervised pretraining to utilize unlabeled data for enhanced feature extraction. Our tests on nasal cavity and vestibule segmentation, conducted on a dataset comprising 128 head CT volumes (including 34,006 slices), demonstrate the robustness and superior performance of proposed method, achieving leading results across multiple segmentation metrics. •Innovative Network Design: Design of DSIFNet for precise and automatic segmentation of nasal cavity and vestibule from 3D CT.•Advanced Feature Extraction: Incorporation of the LGPI-IFF to enhance cross-scale feature extraction and fusion.•Novel Deep Supervision Framework: The deep supervision framework based on LGPI-IFF optimizes multi-scale feature utilization.•Extensive Dataset: Construction of a dataset with 7116 CT for pretraining and 128 annotated CT for training and evaluation.•SOTA Performance: Achieved high accuracy and robustness in nasal cavity and vestibule segmentation via cross-validation.