Identification and validation of an explainable transformer-based model for predicting the prognosis of patients with Non-small cell lung cancer DOI Creative Commons
Peng Zhang,

Ruizi Ren,

Zhanliang Ren

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 10, 2024

Abstract Background Currently, there is a lack of practical and explainable prognostic models for NSCLC in clinical settings. This study aims to construct an model using the Transformer framework deep learning. Methods 119751 patients from Surveillance, Epidemiology End Results (SEER) database were used train Transformer-based predict overall survival (OS) at 12, 24, 60 months. Additionally, SHapley Additive exPlanation (SHAP) method was employed interpret constructed model, show casing importance various indicators on patient different time points. Result The dependent AUC values 12 months, 24 months 0.853, 0.860 0.871 i n training cohort, 0.863, 0.881, 0.899 validation 0.850, 0.851 0.869 testing respectively. Moreover, risk scoring system based Kaplan Meier (KM) curves can accurately divide into three groups. Ultimately, demonstrates differences short-term long-term survival. Conclusion distant metastasis tumor has significant effect patients. surgery long term more than short

Language: Английский

Distant metastatic patterns in young and old non-small cell lung cancer patients: a dose‒response analysis based on SEER population DOI Creative Commons

Yingxian Dong,

Sicheng Zhou,

Jue Li

et al.

Heliyon, Journal Year: 2024, Volume and Issue: 10(17), P. e36657 - e36657

Published: Aug. 23, 2024

BackgroundFew research has explored the risk of distant metastasis dynamically changes with age in non-small cell lung cancer patients. The objective this study was to explore factors developing changing age.MethodsThis retrospective cohort based on a large population patients from SEER database. Logistic regression applied identify for metastasis. clinicopathological features were compared between young group (≤50 years old) and old (>50 old). Dose-response analyses conducted age.ResultsA total 18,711 studied study. According univariate multivariate logistic analyses, ten found be Young have greater incidence each pattern However, survival time younger longer. indicated that risks pleural or pericardial overall gradually decreased at ages, but they intend increase older ages.ConclusionsAge, sex, ethnicity, histology, T category, N differentiation grade, primary site tumor, ipsilateral metastases are associated NSCLC may decrease increasing than 70 years, climb years.

Language: Английский

Citations

0
Le TNM : la 9e édition pour l’oncologie thoracique est née ! DOI
Apostolos C. Agrafiotis, Bogdan Grigoriu, Paul Van Schil

et al.

Revue des Maladies Respiratoires Actualités, Journal Year: 2024, Volume and Issue: 16(2), P. 2S49 - 2S62

Published: Oct. 1, 2024

Citations

0
Different approach to M descriptor for future staging of oligometastatic disease in SCLC: A cross-sectional survival analysis DOI
Melahat Uzel Şener, Pınar Akın Kabalak, Suna Kavurgacı

et al.

Clinical & Translational Oncology, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 5, 2024

Language: Английский

Citations

0
Clinicopathological features and outcomes of rare lung adenocarcinoma metastasis to the thyroid gland: A single‐center, 11‐year experience DOI Creative Commons
Xuehan Gao, Zhen Cao, Xiayao Diao

et al.

Thoracic Cancer, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 18, 2024

Abstract Background Metastasis to the thyroid gland from lung adenocarcinoma is rare and challenging diagnose due similar histopathological features. This study aimed analyze clinicopathological characteristics of treatment strategies for metastasis based on 11 years institutional experience. Methods A retrospective included patients with at our center 2010 2023. Clinicopathological features clinical outcomes were analyzed. Results Among 9714 patients, nine (five females, 55.6%) diagnosed metastasis, presenting primarily cough symptoms. Most (88.9%) had synchronous tumors, whereas a minority (11.1%) metachronous tumors. The median time primary tumor diagnosis was 4.8 months. developed bilateral metastases (88.9%). Diagnosis through fine‐needle aspiration (FNA), one case misdiagnosed as papillary carcinoma. Immunohistochemical staining revealed transcription factor‐1 (TTF‐1) novel aspartic proteinase pepsin family (Napsin‐A) positivity paired box 8 (PAX8) negativity. Genetic testing found epidermal growth factor receptor mutations in 71.4% patients. individualized comprehensive therapy surgery, chemotherapy, immunotherapy, targeted supportive therapy. overall survival 56.0 months, progression‐free 12.7 Kaplan–Meier (K–M) analysis suggested improved no advanced symptoms ( p = 0.03) therapies 0.05). Conclusions Lung disease, an incidence 0.1% among Early after symptom onset personalized may improve prognosis. Despite rapid disease progression, favorable can be achieved treatment.

Language: Английский

Citations

0
Identification and validation of an explainable transformer-based model for predicting the prognosis of patients with Non-small cell lung cancer DOI Creative Commons
Peng Zhang,

Ruizi Ren,

Zhanliang Ren

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 10, 2024

Abstract Background Currently, there is a lack of practical and explainable prognostic models for NSCLC in clinical settings. This study aims to construct an model using the Transformer framework deep learning. Methods 119751 patients from Surveillance, Epidemiology End Results (SEER) database were used train Transformer-based predict overall survival (OS) at 12, 24, 60 months. Additionally, SHapley Additive exPlanation (SHAP) method was employed interpret constructed model, show casing importance various indicators on patient different time points. Result The dependent AUC values 12 months, 24 months 0.853, 0.860 0.871 i n training cohort, 0.863, 0.881, 0.899 validation 0.850, 0.851 0.869 testing respectively. Moreover, risk scoring system based Kaplan Meier (KM) curves can accurately divide into three groups. Ultimately, demonstrates differences short-term long-term survival. Conclusion distant metastasis tumor has significant effect patients. surgery long term more than short

Language: Английский

Citations

0