An artificial-intelligence lung imaging analysis system (ALIAS) for population-based nodule computing in CT scans. (April 2021)

• Record Type: Journal Article
• Authors: Chen, Liyun; Gu, Dongdong; Chen, Yanbo; Shao, Ying; Cao, Xiaohuan; Liu, Guocai; Gao, Yaozong; Wang, Qian; Shen, Dinggang
• Is Part Of: Computerized medical imaging and graphics. Volume 89(2021)
• Publication Date: 2021-04
• Subjects: Computed tomography (CT) -- Lung nodule detection -- Lung nodule segmentation -- Lung nodule atlas -- Statistical analysis; Diagnostic imaging -- Periodicals; Imaging systems in medicine -- Periodicals; Diagnosis, Radioscopic -- Data processing -- Periodicals; Diagnostic Imaging -- Periodicals; Imagerie pour le diagnostic -- Périodiques; Diagnostic imaging; Periodicals; Electronic journals; Electronic journals; 616.0754
• DOI: 10.1016/j.compmedimag.2021.101899
• Languages: English
• ISSNs: 0895-6111

Abstract

Highlights: Present an artificial-intelligence lung image analysis system for lung nodule detection and segmentation. Build up population-based spatial atlas of locations, sizes for all detected nodules, and for benign and malignant nodules. Results on our testing datasets show benign and malignant nodules have significant differences in HU histogram and radiomics features. Abstract: Computed tomography (CT) screening is essential for early lung cancer detection. With the development of artificial intelligence techniques, it is particularly desirable to explore the ability of current state-of-the-art methods and to analyze nodule features in terms of a large population. In this paper, we present an artificial-intelligence lung image analysis system (ALIAS) for nodule detection and segmentation. And after segmenting the nodules, the locations, sizes, as well as imaging features are computed at the population level for studying the differences between benign and malignant nodules. The results provide better understanding of the underlying imaging features and their ability for early lung cancer diagnosis.