<p>Evaluating Solid Lung Adenocarcinoma Anaplastic Lymphoma Kinase Gene Rearrangement Using Noninvasive Radiomics Biomarkers</p>

Abstract

To develop a radiogenomics classifier to assess anaplastic lymphoma kinase (ALK) gene rearrangement status in pretreated solid lung adenocarcinoma noninvasively. Materials and Methods: This study consisted of 140 consecutive pretreated solid lung adenocarcinoma patients with complete enhanced CT scans who were tested for both EGFR mutations and ALK status. Pre-contrast CT and standard post-contrast CT radiogenomics machine learning classifiers were designed as two separate classifiers. In each classifier, dataset was randomly split into training and independent testing group on a 7:3 ratio, accordingly subjected to a 5-fold cross-validation. After normalization, best feature subsets were selected by Pearson correlation coefficient (PCC) and analysis of variance (ANOVA) or recursive feature elimination (RFE), whereupon a radiomics classifier was built with support vector machine (SVM). The discriminating performance was assessed with the area under receiver-operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Results: In classifier one, 98 cases were selected as training data set, 42 cases as independent testing data set. In classifier two, 87 cases were selected as training data set, 37 cases as independent testing data set. Both classifiers extracted 851 radiomics features. The top 25 precontrast features and top 19 post-contrast features were selected to build optimal ALK+ radiogenomics classifiers accordingly. The accuracies, AUCs, sensitivity, specificity, PPV, and NPVof pre-contrast CT classifier were 78.57%, 80.10% (CI: 0.6538-0.9222), 71.43%, 82.14%, 66.67%, and 85.19%, respectively. Those results of standard post-contrast CT classifier were 81.08%, 82.85% (CI: 0.6630-0.9567), 76.92%, 83.33%, 71.43%, and 86.96%. Conclusion: Solid lung adenocarcinoma ALK+ radiogenomics classifier of standard postcontrast CT radiomics biomarkers produced superior performance compared with that of precontrast one, suggesting that post-contrast CT radiomics should be recommended in the context of solid lung adenocarcinoma radiogenomics AI. Standard post-contrast CT machine learning radiogenomics classifier could help precisely identify solid adenocarcinoma ALK rearrangement status, which may act as a pragmatic and cost-efficient substitute for traditional invasive ALK status test.

MRI‐based radiomics analysis for predicting the EGFR mutation based on thoracic spinal metastases in lung adenocarcinoma patients

“…The logistic regression (LR), 27 random forest (RF), 28 neural network (NN), and support vector machine (SVM) 29 were developed and implemented in Python v.3.6 using the “scikit‐learn” package. The linear kernel function was used as a kernel function of SVM 30 …”

“…The linear kernel function was used as a kernel function of SVM. 30 The most important clinical factor was identified using the logistic regression with AIC. The clinical-radiomics nomogram was constructed incorporating the Combined Rad score and the most important clinical predictor using the "rms" package in R v.3.6.…”

Radiogenomics prediction for MYCN amplification in Neuroblastoma: a hypothesis generating study

“…In this context, radiomics and radiogenomics, may play a significant role thanks to the quantitative noninvasive and repeatable analysis of standard clinical imaging that encompasses the whole tumor volume, taking into account the hetMNA. Radiomics features may be associated with specific molecular pathways or mutations, offering reliable predictive tools for clinicians in breast cancer [20], glioma [14], lung cancer [15,21] and even in healthy tissue [22]. Furthermore, CT represent the most common modality used to diagnose and stage NB and for this reason the images are easily accessible even for retrospective analysis.…”

“…To date radiomics is mainly used in oncology to predict diagnosis, survival, progression of disease and gene mutation (i.e. radiogenomics) [12][13][14][15]. Since radiomics has the potential for predicting molecular characteristics, it can potentially be used for determination of mutational or amplification status of specific genes, as suggested by promising preliminary experiences also in the specific field.…”

Radiogenomics prediction for MYCN amplification in neuroblastoma: A hypothesis generating study

“…Radiomics features identification and selection procedure with Boruta selection algorithm Radiomics prediction performance on total sample for MYCN amplified status prediction in breast cancer,20 glioma,14 lung cancer,15,21 and even in healthy tissue 22. Furthermore, CT represents the most common modality used to diagnose and stage NB and for this reason the images are easily accessible even for retrospective analysis.…”