Deep learning radiomics and mediastinal adipose tissue-based nomogram for preoperative prediction of postoperative‌ brain metastasis risk in non-small cell lung cancer

Abstract Background and objectives Brain metastasis (BM) significantly affects the prognosis of non-small cell lung cancer (NSCLC) patients. Increasing evidence suggests that adipose tissue influences cancer progression and metastasis. This study aimed to develop a predictive nomogram integrating me...

Full description

Saved in:
Bibliographic Details
Main Authors: Ye Niu, Hao-Bo Jia, Xue-Meng Li, Wen-Juan Huang, Ping-Ping Liu, Le Liu, Zeng-Yao Liu, Qiu-Jun Wang, Yuan-Zhou Li, Shi-Di Miao, Rui-Tao Wang, Ze-Xun Duan
Format: Article
Language:English
Published: BMC 2025-07-01
Series:BMC Cancer
Subjects:
Online Access:https://doi.org/10.1186/s12885-025-14466-5
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849334723523903488
author Ye Niu
Hao-Bo Jia
Xue-Meng Li
Wen-Juan Huang
Ping-Ping Liu
Le Liu
Zeng-Yao Liu
Qiu-Jun Wang
Yuan-Zhou Li
Shi-Di Miao
Rui-Tao Wang
Ze-Xun Duan
author_facet Ye Niu
Hao-Bo Jia
Xue-Meng Li
Wen-Juan Huang
Ping-Ping Liu
Le Liu
Zeng-Yao Liu
Qiu-Jun Wang
Yuan-Zhou Li
Shi-Di Miao
Rui-Tao Wang
Ze-Xun Duan
author_sort Ye Niu
collection DOAJ
description Abstract Background and objectives Brain metastasis (BM) significantly affects the prognosis of non-small cell lung cancer (NSCLC) patients. Increasing evidence suggests that adipose tissue influences cancer progression and metastasis. This study aimed to develop a predictive nomogram integrating mediastinal fat area (MFA) and deep learning (DL)-derived tumor characteristics to stratify postoperative‌ BM risk in NSCLC patients. Materials and methods A retrospective cohort of 585 surgically resected NSCLC patients was analyzed. Preoperative computed tomography (CT) scans were utilized to quantify MFA using ImageJ software (radiologist-validated measurements). Concurrently, a DL algorithm extracted tumor radiomic features, generating a deep learning brain metastasis score (DLBMS). Multivariate logistic regression identified independent BM predictors, which were incorporated into a nomogram. Model performance was assessed via area under the receiver operating characteristic curve (AUC), calibration plots, integrated discrimination improvement (IDI), net reclassification improvement (NRI), and decision curve analysis (DCA). Results Multivariate analysis identified N stage, EGFR mutation status, MFA, and DLBMS as independent predictors of BM. The nomogram achieved superior discriminative capacity (AUC: 0.947 in the test set), significantly outperforming conventional models. MFA contributed substantially to predictive accuracy, with IDI and NRI values confirming its incremental utility (IDI: 0.123, P < 0.001; NRI: 0.386, P = 0.023). Calibration analysis demonstrated strong concordance between predicted and observed BM probabilities, while DCA confirmed clinical net benefit across risk thresholds. Conclusion This DL-enhanced nomogram, incorporating MFA and tumor radiomics, represents a robust and clinically useful tool for preoperative prediction of postoperative BM risk in NSCLC. The integration of adipose tissue metrics with advanced imaging analytics advances personalized prognostic assessment in NSCLC patients.
format Article
id doaj-art-db315b040d874ef5be18d9fd770da0a7
institution Kabale University
issn 1471-2407
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series BMC Cancer
spelling doaj-art-db315b040d874ef5be18d9fd770da0a72025-08-20T03:45:30ZengBMCBMC Cancer1471-24072025-07-012511910.1186/s12885-025-14466-5Deep learning radiomics and mediastinal adipose tissue-based nomogram for preoperative prediction of postoperative‌ brain metastasis risk in non-small cell lung cancerYe Niu0Hao-Bo Jia1Xue-Meng Li2Wen-Juan Huang3Ping-Ping Liu4Le Liu5Zeng-Yao Liu6Qiu-Jun Wang7Yuan-Zhou Li8Shi-Di Miao9Rui-Tao Wang10Ze-Xun Duan11Department of Internal Medicine, Harbin Medical University Cancer Hospital, Harbin Medical UniversityThe School of Computer Science and Engineering, University of Electronic Science and Technology of ChinaDepartment of Internal Medicine, Harbin Medical University Cancer Hospital, Harbin Medical UniversityDepartment of Internal Medicine, Harbin Medical University Cancer Hospital, Harbin Medical UniversityDepartment of Internal Medicine, Harbin Medical University Cancer Hospital, Harbin Medical UniversityDepartment of Internal Medicine, Harbin Medical University Cancer Hospital, Harbin Medical UniversityDepartment of Internal Medicine, Harbin Medical University Cancer Hospital, Harbin Medical UniversityDepartment of General Practice, Second Affiliated Hospital of Harbin Medical University, Harbin Medical UniversityDepartment of Radiology, Harbin Medical University Cancer Hospital, Harbin Medical UniversitySchool of Computer Science and Technology, Harbin University of Science and TechnologyDepartment of Internal Medicine, Harbin Medical University Cancer Hospital, Harbin Medical UniversityDepartment of Emergency, Harbin Medical University Cancer Hospital, Harbin Medical UniversityAbstract Background and objectives Brain metastasis (BM) significantly affects the prognosis of non-small cell lung cancer (NSCLC) patients. Increasing evidence suggests that adipose tissue influences cancer progression and metastasis. This study aimed to develop a predictive nomogram integrating mediastinal fat area (MFA) and deep learning (DL)-derived tumor characteristics to stratify postoperative‌ BM risk in NSCLC patients. Materials and methods A retrospective cohort of 585 surgically resected NSCLC patients was analyzed. Preoperative computed tomography (CT) scans were utilized to quantify MFA using ImageJ software (radiologist-validated measurements). Concurrently, a DL algorithm extracted tumor radiomic features, generating a deep learning brain metastasis score (DLBMS). Multivariate logistic regression identified independent BM predictors, which were incorporated into a nomogram. Model performance was assessed via area under the receiver operating characteristic curve (AUC), calibration plots, integrated discrimination improvement (IDI), net reclassification improvement (NRI), and decision curve analysis (DCA). Results Multivariate analysis identified N stage, EGFR mutation status, MFA, and DLBMS as independent predictors of BM. The nomogram achieved superior discriminative capacity (AUC: 0.947 in the test set), significantly outperforming conventional models. MFA contributed substantially to predictive accuracy, with IDI and NRI values confirming its incremental utility (IDI: 0.123, P < 0.001; NRI: 0.386, P = 0.023). Calibration analysis demonstrated strong concordance between predicted and observed BM probabilities, while DCA confirmed clinical net benefit across risk thresholds. Conclusion This DL-enhanced nomogram, incorporating MFA and tumor radiomics, represents a robust and clinically useful tool for preoperative prediction of postoperative BM risk in NSCLC. The integration of adipose tissue metrics with advanced imaging analytics advances personalized prognostic assessment in NSCLC patients.https://doi.org/10.1186/s12885-025-14466-5Brain metastasesNon-small-cell lung cancerDeep learningMediastinal fat areaNomogram
spellingShingle Ye Niu
Hao-Bo Jia
Xue-Meng Li
Wen-Juan Huang
Ping-Ping Liu
Le Liu
Zeng-Yao Liu
Qiu-Jun Wang
Yuan-Zhou Li
Shi-Di Miao
Rui-Tao Wang
Ze-Xun Duan
Deep learning radiomics and mediastinal adipose tissue-based nomogram for preoperative prediction of postoperative‌ brain metastasis risk in non-small cell lung cancer
BMC Cancer
Brain metastases
Non-small-cell lung cancer
Deep learning
Mediastinal fat area
Nomogram
title Deep learning radiomics and mediastinal adipose tissue-based nomogram for preoperative prediction of postoperative‌ brain metastasis risk in non-small cell lung cancer
title_full Deep learning radiomics and mediastinal adipose tissue-based nomogram for preoperative prediction of postoperative‌ brain metastasis risk in non-small cell lung cancer
title_fullStr Deep learning radiomics and mediastinal adipose tissue-based nomogram for preoperative prediction of postoperative‌ brain metastasis risk in non-small cell lung cancer
title_full_unstemmed Deep learning radiomics and mediastinal adipose tissue-based nomogram for preoperative prediction of postoperative‌ brain metastasis risk in non-small cell lung cancer
title_short Deep learning radiomics and mediastinal adipose tissue-based nomogram for preoperative prediction of postoperative‌ brain metastasis risk in non-small cell lung cancer
title_sort deep learning radiomics and mediastinal adipose tissue based nomogram for preoperative prediction of postoperative‌ brain metastasis risk in non small cell lung cancer
topic Brain metastases
Non-small-cell lung cancer
Deep learning
Mediastinal fat area
Nomogram
url https://doi.org/10.1186/s12885-025-14466-5
work_keys_str_mv AT yeniu deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT haobojia deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT xuemengli deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT wenjuanhuang deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT pingpingliu deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT leliu deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT zengyaoliu deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT qiujunwang deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT yuanzhouli deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT shidimiao deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT ruitaowang deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer
AT zexunduan deeplearningradiomicsandmediastinaladiposetissuebasednomogramforpreoperativepredictionofpostoperativebrainmetastasisriskinnonsmallcelllungcancer