Cross-cancer knowledge transfer for lymph node metastasis prediction: Applicability of an axillary lymph node-based deep learning model
- 1Hallym University Sacred Heart Hospital
Abstract
Evaluation of lymph node (LN) metastasis is critical for cancer staging and treatment planning. Since radiologic hallmarks of metastatic LNs are shared across primary cancers, a computed tomography (CT)-based deep learning (DL) classifier developed from one anatomy may transfer to others. This study built a DenseNet-121 ensemble model using clinical-implication-applied preprocessing and evaluated its cross-anatomical generalizability to LNs from diverse primary cancers. A DenseNet-121 ensemble model combining fixed-size cropping (FSC) and adjustable square cropping (ASC) was trained from ImageNet initialization on an axillary lymph node (ALN) dataset (n = 1,128 from 523 breast cancer patients). The model utilized clinical-implication-applied perinodal cropping with bicubic filtering and standardized resizing. To test cross-anatomical generalizability, an external dataset of 256 CT images across four anatomical regions was evaluated: cervical, mediastinal, upper-abdominal, and lower-abdominal LNs. Perinodal free margins were systematically varied from 10% to 40% to optimize the balance between context preservation and surround interference. Bicubic filtering and standardized resizing yielded the highest internal performance on the original ALN dataset (FSC-ASC ensemble area under the receiver operating characteristic curve [AUROC] 0.958, 95% confidence interval [CI] 0.920-0.987 at 40% margin). The model demonstrated site-dependent, above-chance generalizability across all external regions. Cervical LNs achieved the highest performance at a 40% margin (AUROC 0.939, 95% CI 0.875-0.989). Mediastinal LNs peaked at a 10% margin (AUROC 0.796, 95% CI 0.672-0.903), with wider margins degrading performance due to adjacent structural interference. Upper-abdominal and lower-abdominal LNs reached optimal performance at a 30% margin (AUROC 0.814 for upper; AUROC 0.917 for lower). A single ALN-trained DL model can accurately predict LN metastasis across cervical, mediastinal, and abdominal stations through site-specific perinodal margin tuning. These findings support the feasibility of cross-cancer knowledge transfer and the clinical utility of generalizable backbones for pan-cancer nodal assessment.
Related
- An integrated geometric quantification and shape analysis framework for axillary lymph node metastasis in breast cancer patients
- MINT: Predicting Enzyme-Cofactor Specificity by Multimodal Knowledge Transfer from Protein-Ligand Interactions
- Improved vision transformer for sports image classification via transfer learning and knowledge distillation
- Prediction transfer does not imply reliability transfer in single-cell foundation models
- Climate-Driven Poverty Prediction in Nusa Tenggara Timur Using Transfer Learning-Based CNN-LSTM and ERA5-Land
BibTeX
@misc{kang2026crosscancer,
title = {{Cross-cancer knowledge transfer for lymph node metastasis prediction: Applicability of an axillary lymph node-based deep learning model}},
author = {J. Kang and J. W. Kang and L. M. Kwon and B.-J. Cho and B. J. Kim},
howpublished = {bioRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.04.764191},
url = {https://www.biorxiv.org/content/10.64898/2026.11.04.764191v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Cross-cancer knowledge transfer for lymph node metastasis prediction: Applicability of an axillary lymph node-based deep learning model
AU - J. Kang
AU - J. W. Kang
AU - L. M. Kwon
AU - B.-J. Cho
AU - B. J. Kim
PY - 2026
DA - 2026/11/04
PB - bioRxiv
DO - 10.64898/2026.11.04.764191
UR - https://www.biorxiv.org/content/10.64898/2026.11.04.764191v1
AB - Evaluation of lymph node (LN) metastasis is critical for cancer staging and treatment planning. Since radiologic hallmarks of metastatic LNs are shared across primary cancers, a computed tomography (CT)-based deep learning (DL) classifier developed from one anatomy may transfer to others. This study built a DenseNet-121 ensemble model using clinical-implication-applied preprocessing and evaluated its cross-anatomical generalizability to LNs from diverse primary cancers. A DenseNet-121 ensemble model combining fixed-size cropping (FSC) and adjustable square cropping (ASC) was trained from ImageNet initialization on an axillary lymph node (ALN) dataset (n = 1,128 from 523 breast cancer patients). The model utilized clinical-implication-applied perinodal cropping with bicubic filtering and standardized resizing. To test cross-anatomical generalizability, an external dataset of 256 CT images across four anatomical regions was evaluated: cervical, mediastinal, upper-abdominal, and lower-abdominal LNs. Perinodal free margins were systematically varied from 10% to 40% to optimize the balance between context preservation and surround interference. Bicubic filtering and standardized resizing yielded the highest internal performance on the original ALN dataset (FSC-ASC ensemble area under the receiver operating characteristic curve [AUROC] 0.958, 95% confidence interval [CI] 0.920-0.987 at 40% margin). The model demonstrated site-dependent, above-chance generalizability across all external regions. Cervical LNs achieved the highest performance at a 40% margin (AUROC 0.939, 95% CI 0.875-0.989). Mediastinal LNs peaked at a 10% margin (AUROC 0.796, 95% CI 0.672-0.903), with wider margins degrading performance due to adjacent structural interference. Upper-abdominal and lower-abdominal LNs reached optimal performance at a 30% margin (AUROC 0.814 for upper; AUROC 0.917 for lower). A single ALN-trained DL model can accurately predict LN metastasis across cervical, mediastinal, and abdominal stations through site-specific perinodal margin tuning. These findings support the feasibility of cross-cancer knowledge transfer and the clinical utility of generalizable backbones for pan-cancer nodal assessment.
ER -