CT radiomics-histopathological correlation for mediastinal lymph node staging in non-small cell lung cancer

Abstract

Background Non-small cell lung cancer (NSCLC) staging relies on accurate assessment of mediastinal lymph nodes. This study investigates the utility of radiomic features derived from contrast-enhanced thoracic CT scans in predicting malignancy in clinically positive (cN+) mediastinal lymph nodes. Methods A retrospective cohort of 110 NSCLC patients with cN + who underwent surgical resection was analyzed. 3D segmentations of up to three lymph nodes per patient were performed. Radiomic features, encompassing heterogeneity measures, were extracted. A radiomics model was constructed using a random forest and XGboost algorithm. To ensure robustness and minimize bias, 100 iterations of data splitting were conducted to create distinct training and test sets for reproducibility and statistical reliability. Results The radiomics model achieved an area under the curve (AUC) of 0.703 for Forest model and 063 for XGboost model. Three recurrent features in the radiomic signatures, “RootMeanSquared”, “Grey Level Co-Occurrence Matrix Imc2”, and “Grey Level Run Length Matrix RunEntropy”, highlighted the importance of nodal heterogeneity features in the model. Conclusion Radiomic features extracted from contrast-enhanced thoracic CT scans could predict malignancy in cN+ mediastinal lymph nodes of NSCLC patients (AUC = 0.703). Three features of heterogeneity were recurrent in radiomic signature, emphasizing the potential importance of incorporating nodal heterogeneity criteria in addition to size assessment.

Surgical nodal staging in cervical cancer, a narrative review from the French cervical cancer guidelines group
Yohann Dabi, Alejandra Martinez, Aymeline Lacorre +5
Journal of Gynecology Obstetrics and Human Reproduction23 Oct 2026Peer-reviewedCC BY 4.0Full textGeneral biology Epidemiology & public healthHAL
Radiomics- based biomarkers for immunotherapy in non-small cell lung cancer: advances, challenges, and future perspectives in biological interpretation and clinical translation
Ran Zhang, Xiaoqing Ji, Shuran Yang +4
Frontiers in Immunology2 Nov 2026Peer-reviewedCC BY 4.0ImmunologyFrontiersCrossref
Correlation-aware discovery of co-occurring mutational signatures in cancer
H. Jin, B. Geiger, D. Glodzik +2
bioRxiv14 Sep 2026PreprintAll rights reservedFull textBioinformatics & computational biologybioRxiv
Learning Interpretable Tumor Microenvironment Representations by Fitting Pan-Cancer Cell State-Niche Correlation
Xiao Xiao, Jiashu He, Shiyang Zhang +1
arXiv26 Sep 2026PreprintCC BY 4.0Full textGenetics & genomicsarXiv
High‐Throughput Geocoding to Assess Short‐Term Air Pollution in Correlation With Mortality in a French Cancer Patient Cohort
Loice Pokam, Joséphine Bocquet, Nadia Hégarat +10
International Journal of Cancer7 Jun 2026Peer-reviewedCC BY 4.0Full textGeneral biology Epidemiology & public healthHAL

Cite

BibTeX RIS

BibTeX

@article{legrand2026radiomicshistopathological,
  title = {{CT radiomics-histopathological correlation for mediastinal lymph node staging in non-small cell lung cancer}},
  author = {Judith Legrand and Antoine Decoux and Loïc Duron and Cécile Di-Rocco and Manuel Gargiulo and Pascal Thomas and Armelle Arnoux and Kathia Chaumoitre and Jean-Yves Gaubert and Laure Fournier and Paul Habert},
  journal = {PLoS ONE},
  year = {2026},
  month = aug,
  doi = {10.1371/journal.pone.0355650},
  url = {https://amu.hal.science/hal-05737884v1},
  note = {CC BY 4.0},
}

RIS

TY  - JOUR
TI  - CT radiomics-histopathological correlation for mediastinal lymph node staging in non-small cell lung cancer
AU  - Judith Legrand
AU  - Antoine Decoux
AU  - Loïc Duron
AU  - Cécile Di-Rocco
AU  - Manuel Gargiulo
AU  - Pascal Thomas
AU  - Armelle Arnoux
AU  - Kathia Chaumoitre
AU  - Jean-Yves Gaubert
AU  - Laure Fournier
AU  - Paul Habert
PY  - 2026
DA  - 2026/08/31
JO  - PLoS ONE
DO  - 10.1371/journal.pone.0355650
UR  - https://amu.hal.science/hal-05737884v1
AB  - Background Non-small cell lung cancer (NSCLC) staging relies on accurate assessment of mediastinal lymph nodes. This study investigates the utility of radiomic features derived from contrast-enhanced thoracic CT scans in predicting malignancy in clinically positive (cN+) mediastinal lymph nodes. Methods A retrospective cohort of 110 NSCLC patients with cN + who underwent surgical resection was analyzed. 3D segmentations of up to three lymph nodes per patient were performed. Radiomic features, encompassing heterogeneity measures, were extracted. A radiomics model was constructed using a random forest and XGboost algorithm. To ensure robustness and minimize bias, 100 iterations of data splitting were conducted to create distinct training and test sets for reproducibility and statistical reliability. Results The radiomics model achieved an area under the curve (AUC) of 0.703 for Forest model and 063 for XGboost model. Three recurrent features in the radiomic signatures, “RootMeanSquared”, “Grey Level Co-Occurrence Matrix Imc2”, and “Grey Level Run Length Matrix RunEntropy”, highlighted the importance of nodal heterogeneity features in the model. Conclusion Radiomic features extracted from contrast-enhanced thoracic CT scans could predict malignancy in cN+ mediastinal lymph nodes of NSCLC patients (AUC = 0.703). Three features of heterogeneity were recurrent in radiomic signature, emphasizing the potential importance of incorporating nodal heterogeneity criteria in addition to size assessment.
SN  - 1932-6203
SN  - 1932-6203
ER  -

Abstract

Related

Cite