Imbalanced Data Correction Based PET/CT Radiomics Model for Predicting Lymph Node Metastasis in Clinical Stage T1 Lung Adenocarcinoma

Abstract

ObjectivesTo develop and validate the imbalanced data correction based PET/CT radiomics model for predicting lymph node metastasis (LNM) in clinical stage T1 lung adenocarcinoma (LUAD).MethodsA total of 183 patients (148/35 non-metastasis/LNM) with pathologically confirmed LUAD were retrospectively included. The cohorts were divided into training vs. validation cohort in a ratio of 7:3. A total of 487 radiomics features were extracted from PET and CT components separately for radiomics model construction. Four clinical features and seven PET/CT radiological features were extracted for traditional model construction. To balance the distribution of majority (non-metastasis) class and minority (LNM) class, the imbalance-adjustment strategies using ten data re-sampling methods were adopted. Three multivariate models (denoted as Traditional, Radiomics, and Combined) were constructed using multivariable logistic regression analysis, where the combined model incorporated all of the significant clinical, radiological, and radiomics features. One hundred times repeated Monte Carlo cross-validation was used to assess the application order of feature selection and imbalance-adjustment strategies in the machine learning pipeline. Prediction performance of each model was evaluated using the area under the receiver operating characteristic curve (AUC) and Geometric mean score (G-mean).ResultsA total of 2 clinical parameters, 2 radiological features, 3 PET, and 5 CT radiomics features were significantly associated with LNM. The combined model with Edited Nearest Neighbors (ENN) re-sampling methods showed strong prediction performance than traditional model or radiomics model with the AUC of 0.94 (95%CI = 0.86–0.97) vs. 0.89 (95%CI = 0.79–0.93), 0.92 (95%CI = 0.85–0.97), and G-mean of 0.88 vs. 0.82, 0.80 in the training cohort, and the AUC of 0.75 (95%CI = 0.57–0.91) vs. 0.68 (95%CI = 0.36–0.83), 0.71 (95%CI = 0.48–0.83) and G-mean of 0.76 vs. 0.64, 0.51 in the validation cohort. The combination of performing feature selection before data re-sampling obtains a better result than the reverse combination (AUC 0.76 ± 0.06 vs. 0.70 ± 0.07, p<0.001).ConclusionsThe combined model (consisting of age, histological type, C/T ratio, MATV, and radiomics signature) integrated with ENN re-sampling methods had strong lymph node metastasis prediction performance for imbalance cohorts in clinical stage T1 LUAD. Radiomics signatures extracted from PET/CT images could provide complementary prediction information compared with traditional model.

CT-based radiomics in predicting pathological response in non-small cell lung cancer patients receiving neoadjuvant immunotherapy

“…Our results suggest that radiomics can predict the pathological remission of NSCLC after immunotherapy-based NAT. Similar to previous studies ( 19 , 47 – 49 ), the prediction efficiency of the entire model is higher than that of the single DL model and the radiomics traditional features combined clinical features model. CT images and clinicopathological information obtained before NAT were constructed as the before_rad_cil model.…”

“…The general procedure of radiomics is as follows: 1) Acquisition of medical imaging data; 2) Region of interest (ROI) segmentation and feature extraction; 3) Feature selection, model building, and validation; 4) Statistical data analysis. Since then, the concept of radiomics has been widely studied in the differentiation of benign and malignant lesions ( 18 ), in the preoperative prediction of lymph node metastasis in lung cancer ( 19 ), and in the assessment of the mutational status of genes such as Epidermal Growth Factor Receptor (EGFR) ( 20 ) and anaplastic lymphoma kinase (ALK) ( 21 ). More radiomics studies include the prediction of treatment effects and prognosis in cancer ( 22 , 23 ) and even in non-neoplastic diseases such as the early diagnosis of Alzheimer’s disease ( 24 ) and the rapid radiological diagnosis of coronavirus disease 2019 (COVID-19) pneumonia ( 25 , 26 ).…”

Automated data preparation for in vivo tumor characterization with machine learning

“…employed LASSO logistic regression for feature selection combined with various oversampling techniques for imbalance correction to predict lymph node metastasis (LNM) in clinical stage T1 lung adenocarcinoma (LUAD). The authors reported highest performance increase of +0.05 AUC (0.70 AUC vs 0.75 AUC) utilizing the edited nearest neighbors (ENN) method ( 53 ). Du et al.…”

Impact of harmonization and oversampling methods on radiomics analysis of multi-center imbalanced datasets: Application to PET-based prediction of lung cancer subtypes

“…Since the accurate prediction of minority classes is crucial for clinical practice, re-sampling techniques that generate more minority examples, like Oversampling, have been shown to be effective in improving prediction accuracy [22]. However, only a limited number of studies have investigated the e cacy of re-sampling approaches on imbalanced PET radiomics [23,24]. In particular, Xie et al [24] conducted a multi-center study and highlighted the improvement of re-sampling techniques combined with classi ers on PET radiomicsbased prognostic performance.…”