Predicting occult lymph node metastasis in solid-predominantly invasive lung adenocarcinoma across multiple centers using radiomics-deep learning fusion model
Abstract Background In solid-predominantly invasive lung adenocarcinoma (SPILAC), occult lymph node metastasis (OLNM) is pivotal for determining treatment strategies. This study seeks to develop and validate a fusion model combining radiomics and deep learning to predict OLNM preoperatively in SPILA...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2024-01-01
|
| Series: | Cancer Imaging |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40644-024-00654-2 |
| _version_ | 1797355538425577472 |
|---|---|
| author | Weiwei Tian Qinqin Yan Xinyu Huang Rui Feng Fei Shan Daoying Geng Zhiyong Zhang |
| author_facet | Weiwei Tian Qinqin Yan Xinyu Huang Rui Feng Fei Shan Daoying Geng Zhiyong Zhang |
| author_sort | Weiwei Tian |
| collection | DOAJ |
| description | Abstract Background In solid-predominantly invasive lung adenocarcinoma (SPILAC), occult lymph node metastasis (OLNM) is pivotal for determining treatment strategies. This study seeks to develop and validate a fusion model combining radiomics and deep learning to predict OLNM preoperatively in SPILAC patients across multiple centers. Methods In this study, 1325 cT1a-bN0M0 SPILAC patients from six hospitals were retrospectively analyzed and divided into pathological nodal positive (pN+) and negative (pN-) groups. Three predictive models for OLNM were developed: a radiomics model employing decision trees and support vector machines; a deep learning model using ResNet-18, ResNet-34, ResNet-50, DenseNet-121, and Swin Transformer, initialized randomly or pre-trained on large-scale medical data; and a fusion model integrating both approaches using addition and concatenation techniques. The model performance was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC). Results All patients were assigned to four groups: training set (n = 470), internal validation set (n = 202), and independent test set 1 (n = 227) and 2 (n = 426). Among the 1325 patients, 478 (36%) had OLNM (pN+). The fusion model, combining radiomics with pre-trained ResNet-18 features via concatenation, outperformed others with an average AUC (aAUC) of 0.754 across validation and test sets, compared to aAUCs of 0.715 for the radiomics model and 0.676 for the deep learning model. Conclusion The radiomics-deep learning fusion model showed promising ability to generalize in predicting OLNM from CT scans, potentially aiding personalized treatment for SPILAC patients across multiple centers. |
| first_indexed | 2024-03-08T14:13:36Z |
| format | Article |
| id | doaj.art-d81fbab52c0a45c09b6611bfb3144adc |
| institution | Directory Open Access Journal |
| issn | 1470-7330 |
| language | English |
| last_indexed | 2024-03-08T14:13:36Z |
| publishDate | 2024-01-01 |
| publisher | BMC |
| record_format | Article |
| series | Cancer Imaging |
| spelling | doaj.art-d81fbab52c0a45c09b6611bfb3144adc2024-01-14T12:34:59ZengBMCCancer Imaging1470-73302024-01-012411910.1186/s40644-024-00654-2Predicting occult lymph node metastasis in solid-predominantly invasive lung adenocarcinoma across multiple centers using radiomics-deep learning fusion modelWeiwei Tian0Qinqin Yan1Xinyu Huang2Rui Feng3Fei Shan4Daoying Geng5Zhiyong Zhang6Academy for Engineering and Technology, Fudan UniversityDepartment of Radiology, Ruijin Hospital, Shanghai Jiao Tong University of MedicineSchool of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing, Fudan UniversitySchool of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing, Fudan UniversityDepartment of Radiology, Shanghai Public Health Clinical Center, Fudan UniversityAcademy for Engineering and Technology, Fudan UniversityDepartment of Radiology, Shanghai Public Health Clinical Center, Fudan UniversityAbstract Background In solid-predominantly invasive lung adenocarcinoma (SPILAC), occult lymph node metastasis (OLNM) is pivotal for determining treatment strategies. This study seeks to develop and validate a fusion model combining radiomics and deep learning to predict OLNM preoperatively in SPILAC patients across multiple centers. Methods In this study, 1325 cT1a-bN0M0 SPILAC patients from six hospitals were retrospectively analyzed and divided into pathological nodal positive (pN+) and negative (pN-) groups. Three predictive models for OLNM were developed: a radiomics model employing decision trees and support vector machines; a deep learning model using ResNet-18, ResNet-34, ResNet-50, DenseNet-121, and Swin Transformer, initialized randomly or pre-trained on large-scale medical data; and a fusion model integrating both approaches using addition and concatenation techniques. The model performance was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC). Results All patients were assigned to four groups: training set (n = 470), internal validation set (n = 202), and independent test set 1 (n = 227) and 2 (n = 426). Among the 1325 patients, 478 (36%) had OLNM (pN+). The fusion model, combining radiomics with pre-trained ResNet-18 features via concatenation, outperformed others with an average AUC (aAUC) of 0.754 across validation and test sets, compared to aAUCs of 0.715 for the radiomics model and 0.676 for the deep learning model. Conclusion The radiomics-deep learning fusion model showed promising ability to generalize in predicting OLNM from CT scans, potentially aiding personalized treatment for SPILAC patients across multiple centers.https://doi.org/10.1186/s40644-024-00654-2Deep learningFeature fusionOccult lymph node metastasisRadiomicsSolid-predominantly invasive lung adenocarcinoma |
| spellingShingle | Weiwei Tian Qinqin Yan Xinyu Huang Rui Feng Fei Shan Daoying Geng Zhiyong Zhang Predicting occult lymph node metastasis in solid-predominantly invasive lung adenocarcinoma across multiple centers using radiomics-deep learning fusion model Cancer Imaging Deep learning Feature fusion Occult lymph node metastasis Radiomics Solid-predominantly invasive lung adenocarcinoma |
| title | Predicting occult lymph node metastasis in solid-predominantly invasive lung adenocarcinoma across multiple centers using radiomics-deep learning fusion model |
| title_full | Predicting occult lymph node metastasis in solid-predominantly invasive lung adenocarcinoma across multiple centers using radiomics-deep learning fusion model |
| title_fullStr | Predicting occult lymph node metastasis in solid-predominantly invasive lung adenocarcinoma across multiple centers using radiomics-deep learning fusion model |
| title_full_unstemmed | Predicting occult lymph node metastasis in solid-predominantly invasive lung adenocarcinoma across multiple centers using radiomics-deep learning fusion model |
| title_short | Predicting occult lymph node metastasis in solid-predominantly invasive lung adenocarcinoma across multiple centers using radiomics-deep learning fusion model |
| title_sort | predicting occult lymph node metastasis in solid predominantly invasive lung adenocarcinoma across multiple centers using radiomics deep learning fusion model |
| topic | Deep learning Feature fusion Occult lymph node metastasis Radiomics Solid-predominantly invasive lung adenocarcinoma |
| url | https://doi.org/10.1186/s40644-024-00654-2 |
| work_keys_str_mv | AT weiweitian predictingoccultlymphnodemetastasisinsolidpredominantlyinvasivelungadenocarcinomaacrossmultiplecentersusingradiomicsdeeplearningfusionmodel AT qinqinyan predictingoccultlymphnodemetastasisinsolidpredominantlyinvasivelungadenocarcinomaacrossmultiplecentersusingradiomicsdeeplearningfusionmodel AT xinyuhuang predictingoccultlymphnodemetastasisinsolidpredominantlyinvasivelungadenocarcinomaacrossmultiplecentersusingradiomicsdeeplearningfusionmodel AT ruifeng predictingoccultlymphnodemetastasisinsolidpredominantlyinvasivelungadenocarcinomaacrossmultiplecentersusingradiomicsdeeplearningfusionmodel AT feishan predictingoccultlymphnodemetastasisinsolidpredominantlyinvasivelungadenocarcinomaacrossmultiplecentersusingradiomicsdeeplearningfusionmodel AT daoyinggeng predictingoccultlymphnodemetastasisinsolidpredominantlyinvasivelungadenocarcinomaacrossmultiplecentersusingradiomicsdeeplearningfusionmodel AT zhiyongzhang predictingoccultlymphnodemetastasisinsolidpredominantlyinvasivelungadenocarcinomaacrossmultiplecentersusingradiomicsdeeplearningfusionmodel |