1
|
Feng B, Chen X, Chen Y, Yu T, Duan X, Liu K, Li K, Liu Z, Lin H, Li S, Chen X, Ke Y, Li Z, Cui E, Long W, Liu X. Identifying Solitary Granulomatous Nodules from Solid Lung Adenocarcinoma: Exploring Robust Image Features with Cross-Domain Transfer Learning. Cancers (Basel) 2023; 15:cancers15030892. [PMID: 36765850 PMCID: PMC9913209 DOI: 10.3390/cancers15030892] [Citation(s) in RCA: 6] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/02/2022] [Revised: 01/17/2023] [Accepted: 01/18/2023] [Indexed: 02/04/2023] Open
Abstract
PURPOSE This study aimed to find suitable source domain data in cross-domain transfer learning to extract robust image features. Then, a model was built to preoperatively distinguish lung granulomatous nodules (LGNs) from lung adenocarcinoma (LAC) in solitary pulmonary solid nodules (SPSNs). METHODS Data from 841 patients with SPSNs from five centres were collected retrospectively. First, adaptive cross-domain transfer learning was used to construct transfer learning signatures (TLS) under different source domain data and conduct a comparative analysis. The Wasserstein distance was used to assess the similarity between the source domain and target domain data in cross-domain transfer learning. Second, a cross-domain transfer learning radiomics model (TLRM) combining the best performing TLS, clinical factors and subjective CT findings was constructed. Finally, the performance of the model was validated through multicentre validation cohorts. RESULTS Relative to other source domain data, TLS based on lung whole slide images as source domain data (TLS-LW) had the best performance in all validation cohorts (AUC range: 0.8228-0.8984). Meanwhile, the Wasserstein distance of TLS-LW was 1.7108, which was minimal. Finally, TLS-LW, age, spiculated sign and lobulated shape were used to build the TLRM. In all validation cohorts, The AUC ranges were 0.9074-0.9442. Compared with other models, decision curve analysis and integrated discrimination improvement showed that TLRM had better performance. CONCLUSIONS The TLRM could assist physicians in preoperatively differentiating LGN from LAC in SPSNs. Furthermore, compared with other images, cross-domain transfer learning can extract robust image features when using lung whole slide images as source domain data and has a better effect.
Collapse
Affiliation(s)
- Bao Feng
- Department of Radiology, Jiangmen Central Hospital, Jiangmen 529000, China
- School of Electronic Information and Automation, Guilin University of Aerospace Technology, Guilin 541004, China
| | - Xiangmeng Chen
- Department of Radiology, Jiangmen Central Hospital, Jiangmen 529000, China
| | - Yehang Chen
- School of Electronic Information and Automation, Guilin University of Aerospace Technology, Guilin 541004, China
| | - Tianyou Yu
- School of Automation Science and Engineering, South China University of Technology, Guangzhou 510641, China
| | - Xiaobei Duan
- Department of Nuclear Medicine, Jiangmen Central Hospital, Jiangmen 529000, China
| | - Kunfeng Liu
- Department of Radiology, Sun Yat-sen University Cancer Center, Guangzhou 510060, China
| | - Kunwei Li
- Department of Radiology, Fifth Affiliated Hospital Sun Yat-sen University, Zhuhai 519000, China
| | - Zaiyi Liu
- Department of Radiology, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Guangzhou 510080, China
| | - Huan Lin
- Department of Radiology, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Guangzhou 510080, China
| | - Sheng Li
- Department of Radiology, Sun Yat-sen University Cancer Center, Guangzhou 510060, China
| | - Xiaodong Chen
- Department of Radiology, Affiliated Hospital of Guangdong Medical University, Zhanjiang 524000, China
| | - Yuting Ke
- Department of Radiology, Affiliated Hospital of Guangdong Medical University, Zhanjiang 524000, China
| | - Zhi Li
- School of Electronic Information and Automation, Guilin University of Aerospace Technology, Guilin 541004, China
| | - Enming Cui
- Department of Radiology, Jiangmen Central Hospital, Jiangmen 529000, China
| | - Wansheng Long
- Department of Radiology, Jiangmen Central Hospital, Jiangmen 529000, China
- Correspondence: (W.L.); (X.L.); Tel.: +86-0750-3165528 (W.L.); +86-138-0923-8549 (X.L.)
| | - Xueguo Liu
- Department of Radiology, The Seventh Affiliated Hospital of Sun Yat-sen University, Shenzhen 518000, China
- Correspondence: (W.L.); (X.L.); Tel.: +86-0750-3165528 (W.L.); +86-138-0923-8549 (X.L.)
| |
Collapse
|
2
|
Artificial Intelligence assisted discrimination between pulmonary tuberculous nodules and solid lung cancer nodules. CLINICAL EHEALTH 2022. [DOI: 10.1016/j.ceh.2022.12.001] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/15/2022] Open
|
3
|
Pan XL, Liao ZL, Yao H, Yan WJ, Wen DY, Wang Y, Li ZL. Prognostic value of ground glass opacity on computed tomography in pathological stage I pulmonary adenocarcinoma: A meta-analysis. World J Clin Cases 2021; 9:10222-10232. [PMID: 34904092 PMCID: PMC8638064 DOI: 10.12998/wjcc.v9.i33.10222] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/07/2021] [Revised: 08/18/2021] [Accepted: 09/10/2021] [Indexed: 02/06/2023] Open
Abstract
BACKGROUND The clinical role of ground glass opacity (GGO) on computed tomography (CT) in stage I pulmonary adenocarcinoma patients currently remains unclear.
AIM To explore the prognostic value of GGO on CT in lung adenocarcinoma patients who were pathologically diagnosed with tumor-node-metastasis stage I.
METHODS A comprehensive and systematic search was conducted through the PubMed, EMBASE and Web of Science databases up to April 3, 2021. The hazard ratio (HR) and corresponding 95% confidence interval (CI) were combined to assess the association between the presence of GGO and prognosis, representing overall survival and disease-free survival. Subgroup analysis based on the ratio of GGO was also conducted. STATA 12.0 software was used for statistical analysis.
RESULTS A total of 12 studies involving 4467 patients were included. The pooled results indicated that the GGO predicted favorable overall survival (HR = 0.44, 95%CI: 0.34-0.59, P < 0.001) and disease-free survival (HR = 0.35, 95%CI: 0.18-0.70, P = 0.003). Subgroup analysis based on the ratio of GGO further demonstrated that the proportion of GGO was a good prognostic indicator in pathological stage I pulmonary adenocarcinoma patients, and patients with a higher ratio of GGO showed better prognosis than patients with a lower GGO ratio did.
CONCLUSION This meta-analysis manifested that the presence of GGO on CT predicted favorable prognosis in tumor-node-metastasis stage I lung adenocarcinoma. Patients with a higher GGO ratio were more likely to have a better prognosis than patients with a lower GGO ratio.
Collapse
Affiliation(s)
- Xue-Lin Pan
- Department of Radiology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province, China
| | - Zi-Ling Liao
- Department of Radiology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province, China
| | - Hui Yao
- Department of Radiology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province, China
| | - Wei-Jie Yan
- Department of Radiology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province, China
| | - De-Ying Wen
- Department of Radiology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province, China
| | - Yan Wang
- Department of Thoracic Surgery, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province, China
| | - Zhen-Lin Li
- Department of Radiology, West China Hospital, Sichuan University, Chengdu 610041, Sichuan Province, China
| |
Collapse
|