Development of a multi-feature-combined model: proof-of-concept with application to local failure prediction of post-SBRT or surgery early-stage NSCLC patients
ObjectiveTo develop a Multi-Feature-Combined (MFC) model for proof-of-concept in predicting local failure (LR) in NSCLC patients after surgery or SBRT using pre-treatment CT images. This MFC model combines handcrafted radiomic features, deep radiomic features, and patient demographic information in...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-09-01
|
| Series: | Frontiers in Oncology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2023.1185771/full |
| _version_ | 1797687159501619200 |
|---|---|
| author | Zhenyu Yang Zhenyu Yang Zhenyu Yang Chunhao Wang Yuqi Wang Kyle J. Lafata Kyle J. Lafata Kyle J. Lafata Haozhao Zhang Bradley G. Ackerson Christopher Kelsey Betty Tong Fang-Fang Yin Fang-Fang Yin |
| author_facet | Zhenyu Yang Zhenyu Yang Zhenyu Yang Chunhao Wang Yuqi Wang Kyle J. Lafata Kyle J. Lafata Kyle J. Lafata Haozhao Zhang Bradley G. Ackerson Christopher Kelsey Betty Tong Fang-Fang Yin Fang-Fang Yin |
| author_sort | Zhenyu Yang |
| collection | DOAJ |
| description | ObjectiveTo develop a Multi-Feature-Combined (MFC) model for proof-of-concept in predicting local failure (LR) in NSCLC patients after surgery or SBRT using pre-treatment CT images. This MFC model combines handcrafted radiomic features, deep radiomic features, and patient demographic information in an integrated machine learning workflow.MethodsThe MFC model comprised three key steps. (1) Extraction of 92 handcrafted radiomic features from the GTV segmented on pre-treatment CT images. (2) Extraction of 512 deep radiomic features from pre-trained U-Net encoder. (3) The extracted handcrafted radiomic features, deep radiomic features, along with 4 patient demographic information (i.e., gender, age, tumor volume, and Charlson comorbidity index), were concatenated as a multi-dimensional input to the classifiers for LR prediction. Two NSCLC patient cohorts from our institution were investigated: (1) the surgery cohort includes 83 patients with segmentectomy or wedge resection (7 LR), and (2) the SBRT cohort includes 84 patients with lung SBRT (9 LR). The MFC model was developed and evaluated independently for both cohorts, and was subsequently compared against the prediction models based on only handcrafted radiomic features (R models), patient demographic information (PI models), and deep learning modeling (DL models). ROC with AUC was adopted to evaluate model performance with leave-one-out cross-validation (LOOCV) and 100-fold Monte Carlo random validation (MCRV). The t-test was performed to identify the statistically significant differences.ResultsIn LOOCV, the AUC range (surgery/SBRT) of the MFC model was 0.858-0.895/0.868-0.913, which was higher than the three other models: 0.356-0.480/0.322-0.650 for PI models, 0.559-0.618/0.639-0.682 for R models, and 0.809/0.843 for DL models. In 100-fold MCRV, the MFC model again showed the highest AUC results (surgery/SBRT): 0.742-0.825/0.888-0.920, which were significantly higher than PI models: 0.464-0.564/0.538-0.628, R models: 0.557-0.652/0.551-0.732, and DL models: 0.702/0.791.ConclusionWe successfully developed an MFC model that combines feature information from multiple sources for proof-of-concept prediction of LR in patients with surgical and SBRT early-stage NSCLC. Initial results suggested that incorporating pre-treatment patient information from multiple sources improves the ability to predict the risk of local failure. |
| first_indexed | 2024-03-12T01:14:16Z |
| format | Article |
| id | doaj.art-fa799e1fc18c41ffb6aad2191e07972a |
| institution | Directory Open Access Journal |
| issn | 2234-943X |
| language | English |
| last_indexed | 2024-03-12T01:14:16Z |
| publishDate | 2023-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Oncology |
| spelling | doaj.art-fa799e1fc18c41ffb6aad2191e07972a2023-09-13T20:18:02ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2023-09-011310.3389/fonc.2023.11857711185771Development of a multi-feature-combined model: proof-of-concept with application to local failure prediction of post-SBRT or surgery early-stage NSCLC patientsZhenyu Yang0Zhenyu Yang1Zhenyu Yang2Chunhao Wang3Yuqi Wang4Kyle J. Lafata5Kyle J. Lafata6Kyle J. Lafata7Haozhao Zhang8Bradley G. Ackerson9Christopher Kelsey10Betty Tong11Fang-Fang Yin12Fang-Fang Yin13Department of Radiation Oncology, Duke University, Durham, NC, United StatesMedical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu, ChinaMedical Physics Graduate Program, Duke University, Durham, NC, United StatesDepartment of Radiation Oncology, Duke University, Durham, NC, United StatesMedical Physics Graduate Program, Duke University, Durham, NC, United StatesDepartment of Radiation Oncology, Duke University, Durham, NC, United StatesDepartment of Electrical and Computer Engineering, Duke University, Durham, NC, United StatesDepartment of Radiology, Duke University, Durham, NC, United StatesMedical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu, ChinaDepartment of Radiation Oncology, Duke University, Durham, NC, United StatesDepartment of Radiation Oncology, Duke University, Durham, NC, United StatesDepartment of Surgery, Duke University, Durham, NC, United StatesDepartment of Radiation Oncology, Duke University, Durham, NC, United StatesMedical Physics Graduate Program, Duke Kunshan University, Kunshan, Jiangsu, ChinaObjectiveTo develop a Multi-Feature-Combined (MFC) model for proof-of-concept in predicting local failure (LR) in NSCLC patients after surgery or SBRT using pre-treatment CT images. This MFC model combines handcrafted radiomic features, deep radiomic features, and patient demographic information in an integrated machine learning workflow.MethodsThe MFC model comprised three key steps. (1) Extraction of 92 handcrafted radiomic features from the GTV segmented on pre-treatment CT images. (2) Extraction of 512 deep radiomic features from pre-trained U-Net encoder. (3) The extracted handcrafted radiomic features, deep radiomic features, along with 4 patient demographic information (i.e., gender, age, tumor volume, and Charlson comorbidity index), were concatenated as a multi-dimensional input to the classifiers for LR prediction. Two NSCLC patient cohorts from our institution were investigated: (1) the surgery cohort includes 83 patients with segmentectomy or wedge resection (7 LR), and (2) the SBRT cohort includes 84 patients with lung SBRT (9 LR). The MFC model was developed and evaluated independently for both cohorts, and was subsequently compared against the prediction models based on only handcrafted radiomic features (R models), patient demographic information (PI models), and deep learning modeling (DL models). ROC with AUC was adopted to evaluate model performance with leave-one-out cross-validation (LOOCV) and 100-fold Monte Carlo random validation (MCRV). The t-test was performed to identify the statistically significant differences.ResultsIn LOOCV, the AUC range (surgery/SBRT) of the MFC model was 0.858-0.895/0.868-0.913, which was higher than the three other models: 0.356-0.480/0.322-0.650 for PI models, 0.559-0.618/0.639-0.682 for R models, and 0.809/0.843 for DL models. In 100-fold MCRV, the MFC model again showed the highest AUC results (surgery/SBRT): 0.742-0.825/0.888-0.920, which were significantly higher than PI models: 0.464-0.564/0.538-0.628, R models: 0.557-0.652/0.551-0.732, and DL models: 0.702/0.791.ConclusionWe successfully developed an MFC model that combines feature information from multiple sources for proof-of-concept prediction of LR in patients with surgical and SBRT early-stage NSCLC. Initial results suggested that incorporating pre-treatment patient information from multiple sources improves the ability to predict the risk of local failure.https://www.frontiersin.org/articles/10.3389/fonc.2023.1185771/fullearly-stage lung NSCLClocal failuresurgerySBRTmachine learning (ML)deep learning |
| spellingShingle | Zhenyu Yang Zhenyu Yang Zhenyu Yang Chunhao Wang Yuqi Wang Kyle J. Lafata Kyle J. Lafata Kyle J. Lafata Haozhao Zhang Bradley G. Ackerson Christopher Kelsey Betty Tong Fang-Fang Yin Fang-Fang Yin Development of a multi-feature-combined model: proof-of-concept with application to local failure prediction of post-SBRT or surgery early-stage NSCLC patients Frontiers in Oncology early-stage lung NSCLC local failure surgery SBRT machine learning (ML) deep learning |
| title | Development of a multi-feature-combined model: proof-of-concept with application to local failure prediction of post-SBRT or surgery early-stage NSCLC patients |
| title_full | Development of a multi-feature-combined model: proof-of-concept with application to local failure prediction of post-SBRT or surgery early-stage NSCLC patients |
| title_fullStr | Development of a multi-feature-combined model: proof-of-concept with application to local failure prediction of post-SBRT or surgery early-stage NSCLC patients |
| title_full_unstemmed | Development of a multi-feature-combined model: proof-of-concept with application to local failure prediction of post-SBRT or surgery early-stage NSCLC patients |
| title_short | Development of a multi-feature-combined model: proof-of-concept with application to local failure prediction of post-SBRT or surgery early-stage NSCLC patients |
| title_sort | development of a multi feature combined model proof of concept with application to local failure prediction of post sbrt or surgery early stage nsclc patients |
| topic | early-stage lung NSCLC local failure surgery SBRT machine learning (ML) deep learning |
| url | https://www.frontiersin.org/articles/10.3389/fonc.2023.1185771/full |
| work_keys_str_mv | AT zhenyuyang developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT zhenyuyang developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT zhenyuyang developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT chunhaowang developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT yuqiwang developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT kylejlafata developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT kylejlafata developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT kylejlafata developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT haozhaozhang developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT bradleygackerson developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT christopherkelsey developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT bettytong developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT fangfangyin developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients AT fangfangyin developmentofamultifeaturecombinedmodelproofofconceptwithapplicationtolocalfailurepredictionofpostsbrtorsurgeryearlystagensclcpatients |