Search

Objectives: With the popularization of chest computed tomography (CT) screening, there are more sub-centimeter (≤ 1 cm) pulmonary nodules (SCPNs) requiring further diagnostic workup. This area represents an important opportunity to optimize the SCPN management algorithm avoiding "one-size fits all" approach. One critical problem is how to learn the discriminative multi-view characteristics and the unique context of each SCPN. Methods: Here, we propose a multi-view coupled self-attention module (MVCS) to capture the global spatial context of the CT image through modeling the association order of space and dimension. Compared with existing self-attention methods, MVCS uses less memory consumption and computational complexity, unearths dimension correlations that previous methods have not found, and is easy to integrate with other frameworks. Results: In total, a public dataset LUNA16 from LIDC-IDRI, 1319 SCPNs from 1069 patients presenting to a major referral center, and 160 SCPNs from 137 patients from three other major centers were analyzed to pre-train, train, and validate the model. Experimental results showed that performance outperforms the state-of-the-art models in terms of accuracy and stability and is comparable to that of human experts in classifying precancerous lesions and invasive adenocarcinoma. We also provide a fusion MVCS network (MVCSN) by combining the CT image with the clinical characteristics and radiographic features of patients. Conclusion: This tool may ultimately aid in expediting resection of the malignant SCPNs and avoid over-diagnosis of the benign ones, resulting in improved management outcomes. Clinical relevance statement: In the diagnosis of sub-centimeter lung adenocarcinoma, fusion MVCSN can help doctors improve work efficiency and guide their treatment decisions to a certain extent. Key Points: • Advances in computed tomography (CT) not only increase the number of nodules detected, but also the nodules that are identified are smaller, such as sub-centimeter pulmonary nodules (SCPNs). • We propose a multi-view coupled self-attention module (MVCS), which could model spatial and dimensional correlations sequentially for learning global spatial contexts, which is better than other attention mechanisms. • MVCS uses fewer huge memory consumption and computational complexity than the existing self-attention methods when dealing with 3D medical image data. Additionally, it reaches promising accuracy for SCPNs' malignancy evaluation and has lower training cost than other models. [ABSTRACT FROM AUTHOR]

Objective: In recent years, the lung cancer incidence has grown and the population is younger. We intend to find out the true detection rate of pulmonary nodules and the incidence of lung cancer in the population and search for the risk factors. Method: Hospital employees ≥40 years old who underwent low‐dose computed tomography (CT) lung cancer screening from January 2019 to March 2022 were selected to record CT‐imaging characteristics, pathology, staging, and questionnaires to investigate past history, smoking history, diet, mental health, etc. PM2.5 and radiation intake in radiation‐related occupation received monitoring in hospital. Result: The detection rate of suspicious pulmonary nodules was 9.1% (233/2552), and the incidence rate of lung cancer (including adenocarcinoma in situ) was 4.0% (103/2552). Morbidity among doctors, nurses, technicians, administers, and logistics was no difference (p = 0.184), but higher in women than in men (4.7% vs 2.4% p = 0.002). The invasiveness increased with age and CT density of nodules (p = 0.018). The relationship between lung cancer morbidity and PM2.5 was not clear (p = 0.543); and no lung cancer has been found in employees related ionizing radiation. Conclusion: The high screening rate has brought about a high incidence of lung cancer. At present, the risk factor analysis of lung cancer based on small samples cannot find the direct cause. Most of the ground glass opacity (GGO)s detected by LDCT screening are indolent, but there are also rapidly progressive lung cancer. A predictive model to identify active and indolent GGO is necessary. [ABSTRACT FROM AUTHOR]