Detailed identification of epidermal growth factor receptor mutations in lung adenocarcinoma: Combining radiomics with machine learning

CT-based non-invasive identification of the most common gene mutation status in patients with non-small cell lung cancer

BACKGROUND

Epidermal growth factor receptor (EGFR) and Kirsten rat sarcoma viral oncogene homolog (KRAS) are mutually exclusive, and they are two important genes that are most prone to mutation in patients with non-small cell lung cancer.

PURPOSE

This retrospective study investigated the ability of radiomics to predict the mutation status of EGFR and KRAS in patients with non-small cell lung cancer (NSCLC) and guide precision medicine.

METHODS

Computed tomography images of 1045 NSCLC patients from five different institutions were collected, and 1204 imaging features were extracted. In the training set (EGFR: 678, KRAS: 246), Max-Relevance and Min-Redundancy and least absolute shrinkage and selection operator logistic regression were used to screen radiomics features. The combination of selected radiomics features and clinical factors was used to establish the combined models in identifying EGFR and KRAS mutation status, respectively, through stepwise logistic regression. Then, on two independent external validation sets (EGFR: 203/164, KRAS: 123/95), the performance of each model was evaluated separately, and then the overall performance of predicting the two mutation states was calculated.

RESULTS

In the EGFR and KRAS groups, radiomics signatures comprised 14 and 10 radiomics features, respectively. They were mutually exclusive between the tumors with positive EGFR mutation and those with positive KRAS mutation in imaging phenotype. For the EGFR group, the area under the curve (AUC) of the combined model in the two validation sets was 0.871 (95% CI: 0.821-0.926) and 0.861 (95% CI: 0.802-0.911), respectively, whereas the AUC of the combined model in the two validation sets was 0.798 (95% CI: 0.739-0.850) and 0.778 (95% CI: 0.735-0.821), respectively, for the KRAS group. Considering both EGFR and KRAS, the overall precision, recall, and F1-score of the combined model in the two validation sets were 0.704, 0.844, and 0.768, as well as 0.754, 0.693, and 0.722, respectively.

CONCLUSIONS

Our study demonstrates the potential of radiomics in the non-invasive identification of EGFR and KRAS mutation status, which may guide patients with non-small cell lung cancer to choose the most appropriate personalized treatment. This method can be used when biopsy will bring unacceptable risk to patients with NSCLC.

AI/ML advances in non-small cell lung cancer biomarker discovery

Lung cancer is the leading cause of cancer deaths among both men and women, representing approximately 25% of cancer fatalities each year. The treatment landscape for non-small cell lung cancer (NSCLC) is rapidly evolving due to the progress made in biomarker-driven targeted therapies. While advancements in targeted treatments have improved survival rates for NSCLC patients with actionable biomarkers, long-term survival remains low, with an overall 5-year relative survival rate below 20%. Artificial intelligence/machine learning (AI/ML) algorithms have shown promise in biomarker discovery, yet NSCLC-specific studies capturing the clinical challenges targeted and emerging patterns identified using AI/ML approaches are lacking. Here, we employed a text-mining approach and identified 215 studies that reported potential biomarkers of NSCLC using AI/ML algorithms. We catalogued these studies with respect to BEST (Biomarkers, EndpointS, and other Tools) biomarker sub-types and summarized emerging patterns and trends in AI/ML-driven NSCLC biomarker discovery. We anticipate that our comprehensive review will contribute to the current understanding of AI/ML advances in NSCLC biomarker research and provide an important catalogue that may facilitate clinical adoption of AI/ML-derived biomarkers.

Three-dimensional topological radiogenomics of epidermal growth factor receptor Del19 and L858R mutation subtypes on computed tomography images of lung cancer patients

OBJECTIVES

To elucidate a novel radiogenomics approach using three-dimensional (3D) topologically invariant Betti numbers (BNs) for topological characterization of epidermal growth factor receptor (EGFR) Del19 and L858R mutation subtypes.

METHODS

In total, 154 patients (wild-type EGFR, 72 patients; Del19 mutation, 45 patients; and L858R mutation, 37 patients) were retrospectively enrolled and randomly divided into 92 training and 62 test cases. Two support vector machine (SVM) models to distinguish between wild-type and mutant EGFR (mutation [M] classification) as well as between the Del19 and L858R subtypes (subtype [S] classification) were trained using 3DBN features. These features were computed from 3DBN maps by using histogram and texture analyses. The 3DBN maps were generated using computed tomography (CT) images based on the Čech complex constructed on sets of points in the images. These points were defined by coordinates of voxels with CT values higher than several threshold values. The M classification model was built using image features and demographic parameters of sex and smoking status. The SVM models were evaluated by determining their classification accuracies. The feasibility of the 3DBN model was compared with those of conventional radiomic models based on pseudo-3D BN (p3DBN), two-dimensional BN (2DBN), and CT and wavelet-decomposition (WD) images. The validation of the model was repeated with 100 times random sampling.

RESULTS

The mean test accuracies for M classification with 3DBN, p3DBN, 2DBN, CT, and WD images were 0.810, 0.733, 0.838, 0.782, and 0.799, respectively. The mean test accuracies for S classification with 3DBN, p3DBN, 2DBN, CT, and WD images were 0.773, 0.694, 0.657, 0.581, and 0.696, respectively.

CONCLUSION

3DBN features, which showed a radiogenomic association with the characteristics of the EGFR Del19/L858R mutation subtypes, yielded higher accuracy for subtype classifications in comparison with conventional features.

Comparison of the radiomics-based predictive models using machine learning and nomogram for epidermal growth factor receptor mutation status and subtypes in lung adenocarcinoma

The purpose of this study is to develop the predictive models for epidermal growth factor receptor (EGFR) mutation status and subtypes [exon 21-point mutation (L858R) and exon 19 deletion mutation (19Del)] and evaluate their clinical usefulness. Total 172 patients with lung adenocarcinoma were retrospectively analyzed. The analysis of variance and the least absolute shrinkage were used for feature selection from plain computed tomography images. Then, radiomic score (rad-score) was calculated for the training and test cohorts. Two machine learning (ML) models with 5-fold were applied to construct the predictive models with rad-score, clinical features, and the combination of rad-score and clinical features. The nomogram was developed using rad-score and clinical features. The prediction performance was evaluated by the area under the receiver operating characteristic curve (AUC). Finally, decision curve analysis (DCA) was performed using the best ML and nomogram models. In the test cohorts, the AUC of the best ML and the nomogram model were 0.73 (95% confidence interval, 0.59-0.87) and 0.79 (0.65-0.92) in the EGFR mutation groups, 0.83 (0.67-0.99) and 0.85 (0.72-0.97) in the L858R mutation groups, as well as 0.77 (0.58-0.97) and 0.77 (0.60-0.95) in the 19Del groups. The DCA showed that the nomogram models have comparable results with ML models. We constructed two predictive models for EGFR mutation status and subtypes. The nomogram models had comparable results to the ML models. Because the superiority of the performance of ML and nomogram models varied depending on the prediction groups, appropriate model selection is necessary.

Comprehensive analysis of prediction of the EGFR mutation and subtypes based on the spinal metastasis from primary lung adenocarcinoma

Purpose

To investigate the use of multiparameter MRI-based radiomics in the in-depth prediction of epidermal growth factor receptor (EGFR) mutation and subtypes based on the spinal metastasis in patients with primary lung adenocarcinoma.

Methods

A primary cohort was conducted with 257 patients who pathologically confirmed spinal bone metastasis from the first center between Feb. 2016 and Oct. 2020. An external cohort was developed with 42 patients from the second center between Apr. 2017 and Jun. 2021. All patients underwent sagittal T1-weighted imaging (T1W) and sagittal fat-suppressed T2-weight imaging (T2FS) MRI imaging. Radiomics features were extracted and selected to build radiomics signatures (RSs). Machine learning classify with 5-fold cross-validation were used to establish radiomics models for predicting the EGFR mutation and subtypes. Clinical characteristics were analyzed with Mann-Whitney U and Chi-Square tests to identify the most important factors. Nomogram models were developed integrating the RSs and important clinical factors.

Results

The RSs derived from T1W showed better performance for predicting the EGFR mutation and subtypes compared with those from T2FS in terms of AUC, accuracy and specificity. The nomogram models integrating RSs from combination of the two MRI sequences and important clinical factors achieved the best prediction capabilities in the training (AUCs, EGFR vs. Exon 19 vs. Exon 21, 0.829 vs. 0.885 vs.0.919), internal validation (AUCs, EGFR vs. Exon 19 vs. Exon 21, 0.760 vs. 0.777 vs.0.811), external validation (AUCs, EGFR vs. Exon 19 vs. Exon 21, 0.780 vs. 0.846 vs.0.818). DCA curves indicated potential clinical values of the radiomics models.

Conclusions

This study indicated potentials of multi-parametric MRI-based radiomics to assess the EGFR mutation and subtypes. The proposed clinical-radiomics nomogram models can be considered as non-invasive tools to assist clinicians in making individual treatment plans.

The value of magnetic resonance imaging-based tumor shape features for assessing microsatellite instability status in endometrial cancer

BACKGROUND

Microsatellite instability (MSI) status can be used for the classification and risk stratification of endometrial cancer (EC). This study aimed to investigate whether magnetic resonance imaging (MRI)-based tumor shape features can help assess MSI status in EC before surgery.

METHODS

The medical records of 88 EC patients with MSI status were retrospectively reviewed. Quantitative and subjective shape features based on MRI were used to assess MSI status. Variables were compared using the Student's t-test, χ² test, or Wilcoxon rank-sum test where appropriate. Univariate and multivariate analyses were performed by the logistic regression model. The area under the curve (AUC) was used to estimate the discrimination performance of variables.

RESULTS

There were 23 patients with MSI, and 65 patients with microsatellite stability (MSS) in this study. Eccentricity and shape type showed significant differences between MSI and MSS (P=0.039 and P=0.033, respectively). The AUC values of eccentricity, shape type, and the combination of 2 features for assessing MSI were 0.662 [95% confidence interval (CI): 0.554-0.770], 0.627 (95% CI: 0.512-0.743), and 0.727 (95% CI: 0.613-0.842), respectively. Considering the International Federation of Gynecology and Obstetrics (FIGO) staging, eccentricity maintained a significant difference in stages I-II (P=0.039), while there was no statistical difference in stages III-IV (P=0.601).

CONCLUSIONS

It is possible that MRI-based tumor shape features, including eccentricity and shape type, could be promising markers for assessing MSI status. The features may aid in the preliminary screening of EC patients with MSI.

Radiomic Signatures for Predicting EGFR Mutation Status in Lung Cancer Brain Metastases

Background

Lung cancer is the most common primary tumor metastasizing to the brain. A significant proportion of lung cancer patients show epidermal growth factor receptor (EGFR) mutation status discordance between the primary cancer and the corresponding brain metastases, which can affect prognosis and therapeutic decision-making. However, it is not always feasible to obtain brain metastases samples. The aim of this study was to establish a radiomic model to predict the EGFR mutation status of lung cancer brain metastases.

Methods

Data from 162 patients with resected brain metastases originating from lung cancer (70 with mutant EGFR, 92 with wild-type EGFR) were retrospectively analyzed. Radiomic features were extracted using preoperative brain magnetic resonance (MR) images (contrast-enhanced T1-weighted imaging, T1CE; T2-weighted imaging, T2WI; T2 fluid-attenuated inversion recovery, T2 FLAIR; and combinations of these sequences), to establish machine learning-based models for predicting the EGFR status of excised brain metastases (108 metastases for training and 54 metastases for testing). The least absolute shrinkage selection operator was used to select informative features; radiomics models were built with logistic regression of the training cohort, and model performance was evaluated using an independent test set.

Results

The best-performing model was a combination of 10 features selected from multiple sequences (two from T1CE, five from T2WI, and three from T2 FLAIR) in both the training and test sets, resulting in classification area under the curve, accuracy, sensitivity, and specificity values of 0.85 and 0.81, 77.8% and 75.9%, 83.7% and 73.1%, and 73.8% and 78.6%, respectively.

Conclusions

Radiomic signatures integrating multi-sequence MR images have the potential to noninvasively predict the EGFR mutation status of lung cancer brain metastases.

Development and externally validate MRI-based nomogram to assess EGFR and T790M mutations in patients with metastatic lung adenocarcinoma

OBJECTIVES

This study aims to explore values of multi-parametric MRI-based radiomics for detecting the epidermal growth factor receptor (EGFR) mutation and resistance (T790M) mutation in lung adenocarcinoma (LA) patients with spinal metastasis.

METHODS

This study enrolled a group of 160 LA patients from our hospital (between Jan. 2017 and Feb. 2021) to build a primary cohort. An external cohort was developed with 32 patients from another hospital (between Jan. 2017 and Jan. 2021). All patients underwent spinal MRI (including T1-weighted (T1W) and T2-weighted fat-suppressed (T2FS)) scans. Radiomics features were extracted from the metastasis for each patient and selected to develop radiomics signatures (RSs) for detecting the EGFR and T790M mutations. The clinical-radiomics nomogram models were constructed with RSs and important clinical parameters. The receiver operating characteristics (ROC) curve was used to evaluate the predication capabilities of each model. Calibration and decision curve analyses (DCA) were constructed to verify the performance of the models.

RESULTS

For detecting the EGFR and T790M mutation, the developed RSs comprised 9 and 4 most important features, respectively. The constructed nomogram models incorporating RSs and smoking status showed favorite prediction efficacy, with AUCs of 0.849 (Sen = 0.685, Spe = 0.885), 0.828 (Sen = 0.964, Spe = 0.692), and 0.778 (Sen = 0.611, Spe = 0.929) in the training, internal validation, and external validation sets for detecting the EGFR mutation, respectively, and with AUCs of 0.0.842 (Sen = 0.750, Spe = 0.867), 0.823 (Sen = 0.667, Spe = 0.938), and 0.800 (Sen = 0.875, Spe = 0.800) in the training, internal validation, and external validation sets for detecting the T790M mutation, respectively.

CONCLUSIONS

Radiomics features from the spinal metastasis were predictive on both EGFR and T790M mutations. The constructed nomogram models can be potentially considered as new markers to guild treatment management in LA patients with spinal metastasis.

KEY POINTS

• To our knowledge, this study was the first approach to detect the EGFR T790M mutation based on spinal metastasis in patients with lung adenocarcinoma. • We identified 13 MRI features that were strongly associated with the EGFR T790M mutation. • The proposed nomogram models can be considered as potential new markers for detecting EGFR and T790M mutations based on spinal metastasis.

Machine Learning-Based Radiomics for Prediction of Epidermal Growth Factor Receptor Mutations in Lung Adenocarcinoma

Identifying an epidermal growth factor receptor (EGFR) mutation is important because EGFR tyrosine kinase inhibitors are the first-line treatment of choice for patients with EGFR mutation-positive lung adenocarcinomas (LUAC). This study is aimed at developing and validating a radiomics-based machine learning (ML) approach to identify EGFR mutations in patients with LUAC. We retrospectively collected data from 201 patients with positive EGFR mutation LUAC (140 in the training cohort and 61 in the validation cohort). We extracted 1316 radiomics features from preprocessed CT images and selected 14 radiomics features and 1 clinical feature which were most relevant to mutations through filter method. Subsequently, we built models using 7 ML approaches and established the receiver operating characteristic (ROC) curve to assess the discriminating performance of these models. In terms of predicting EGFR mutation, the model derived from radiomics features and combined models (radiomics features and relevant clinical factors) had an AUC of 0.79 (95% confidence interval (CI): 0.77-0.82), 0.86 (0.87-0.88), respectively. Our study offers a radiomics-based ML model using filter methods to detect the EGFR mutation in patients with LUAC. This convenient and low-cost method may be of help to noninvasively identify patients before obtaining tumor sample for molecule testing.

Predicting Ki-67 labeling index level in early-stage lung adenocarcinomas manifesting as ground-glass opacity nodules using intra-nodular and peri-nodular radiomic features

Objectives

To explore the diagnostic value of radiomics in differentiating between lung adenocarcinomas appearing as ground‐glass opacity nodules (GGO) with high‐ and low Ki‐67 expression levels.

Materials and Methods

From January 2018 to January 2021, patients with pulmonary GGO who received lung resection were evaluated for potential enrollment. The included GGOs were then randomly divided into a training cohort and a validation cohort with a ratio of 7:3. Logistic regression (LR), decision tree (DT), support vector machines (SVM), and adaboost (AB) were applied for radiomic model construction. Area under the curve (AUC) of the receiver operating characteristic (ROC) curve was used to evaluate the diagnostic efficacy of the established models.

Results

Seven hundred and sixty‐nine patients with 769 GGOs were included in this study. Two hundred and forty‐five GGOs were confirmed to be of high Ki‐67 labeling index (LI). In the training cohort, gender, age, spiculation sign, pleural indentation sign, bubble sign, and maximum 2D diameter of the nodule were found to be significantly different between high‐ and low Ki‐67 LI groups (p < 0.05), and spiculation sign and maximum 2D diameter of the nodule were further confirmed to be risk factors for Ki‐67 LI. The radiomic model established using SVM exhibited an AUC of 0.731 in the validation cohort, which was higher than that of the clinical‐radiographic model (AUC = 0.675). Moreover, radiomic model combining both intra‐ and peri‐nodular features showed better diagnostic efficacy than using intra‐nodular features alone (AUC = 0.731 and 0.720, respectively).

Conclusions

The established radiomic model exhibited good diagnostic efficacy in differentiating between lung adenocarcinoma GGOs with high and low Ki‐67 LI, which was higher than the clinical‐radiographic model. Peri‐nodular radiomic features showed added benefits to the radiomic model. As a novel noninvasive method, radiomics have the potential to be applied in the preliminary classification of Ki‐67 expression level in lung adenocarcinoma GGOs.

MRI-Based Radiomics Nomogram as a Potential Biomarker to Predict the EGFR Mutations in Exon 19 and 21 Based on Thoracic Spinal Metastases in Lung Adenocarcinoma

RATIONALE AND OBJECTIVES

Preoperative identifications of epidermal growth factor receptor (EGFR) mutation subtypes based on the MRI image of spinal metastases are needed to provide individualized therapy, but has not been previously investigated. This study aims to develop and evaluate an MRI-based radiomics nomogram for differentiating the exon 19 and 21 in EGFR mutation from spinal bone metastases in patients with primary lung adenocarcinoma.

MATERIALS AND METHODS

A total of 76 patients underwent T1-weighted and T2-weighted fat-suppressed MRI scans were enrolled in this study, 38 were positive for EGFR mutation in exon 19 and 38 were in exon 21.MRI imaging features were extracted and selected from each MRI pulse sequence, and used to form the radiomics signature. A radiomics nomogram was developed integrating the radiomics signature and important clinical factors with receiver operating characteristic, calibration and decision curve analysis to assess the nomogram. Clinical characteristics were analyzed with Mann-Whitney U and Chi-Square tests to identify the most important factors.

RESULTS

A total of 6 features were selected as the most discriminative predictors from the two MRI pulse sequences. The nomogram integrating the combined radiomics signature, age and CEA level generated good prediction performance in the training (AUCs, nomogram vs. combined radiomics signature vs. clinical model, 0.90 vs. 0.87 vs. 0.59) and validation (AUCs, nomogram vs. combined radiomics signature vs. clinical model, 0.88 vs. 0.86 vs. 0.72) cohort. DCA analysis confirmed the potential clinical utility of the nomogram.

CONCLUSION

This study demonstrated that MRI features from spinal bone metastases can be used to prognosticate EGFR mutation subtypes in exon 19 and 21. The developed pre-treatment nomogram can potentially guide treatments for lung adenocarcinoma patients.

Molecular typing of lung adenocarcinoma with computed tomography and CT image-based radiomics: a narrative review of research progress and prospects

Objective

The purpose of this paper was to perform a narrative review of current research evidence on conventional computed tomography (CT) imaging features and CT image-based radiomic features for predicting gene mutations in lung adenocarcinoma and discuss how to translate the research findings to guide future practice.

Background

Lung cancer, especially lung adenocarcinoma, is the leading cause of cancer-related deaths. With advances in the diagnosis and treatment of lung adenocarcinoma with the emergence of molecular testing, the prediction of oncogenes and even drug resistance gene mutations have become key to individualized and precise clinical treatment in order to prolong survival and improve quality of life. The progress of imageological examination includes the development of CT and radiomics are promising quantitative methods for predicting different gene mutations in lung adenocarcinoma, especially common mutations, such as epidermal growth factor receptor (EGFR) mutation, anaplastic lymphoma kinase (ALK) mutation and Kirsten rat sarcoma viral oncogene (KRAS) mutation.

Methods

The PubMed electronic database was searched along with a set of terms specific to lung adenocarcinoma, radiomics (including texture analysis), CT, computed tomography, EGFR, ALK, KRAS, rearranging transfection (RET) rearrangement and c-ros oncogene 1 (ROS-1), v-raf murine sarcoma viral oncogene homolog B1 (BRAF), and human epidermal growth factor receptor 2 (HER2) mutations et al. This review has been reported in compliance with the Narrative Review checklist guidelines. From each full-text article, information was extracted regarding a set of terms above.

Conclusions

Research on the application of conventional CT features and CT image-based radiomic features for predicting the gene mutation status of lung adenocarcinoma is still in a preliminary stage. Noninvasively determination of mutation status in lung adenocarcinoma before targeted therapy with conventional CT features and CT image-based radiomic features remains both hopes and challenges. Before radiomics could be applied in clinical practice, more work needs to be done.

OUP accepted manuscript

The recent, rapid advances in immuno-oncology have revolutionized cancer treatment and spurred further research into tumor biology. Yet, cancer patients respond variably to immunotherapy despite mounting evidence to support its efficacy. Current methods for predicting immunotherapy response are unreliable, as these tests cannot fully account for tumor heterogeneity and microenvironment. An improved method for predicting response to immunotherapy is needed. Recent studies have proposed radiomics—the process of converting medical images into quantitative data (features) that can be processed using machine learning algorithms to identify complex patterns and trends—for predicting response to immunotherapy. Because patients undergo numerous imaging procedures throughout the course of the disease, there exists a wealth of radiological imaging data available for training radiomics models. And because radiomic features reflect cancer biology, such as tumor heterogeneity and microenvironment, these models have enormous potential to predict immunotherapy response more accurately than current methods. Models trained on preexisting biomarkers and/or clinical outcomes have demonstrated potential to improve patient stratification and treatment outcomes. In this review, we discuss current applications of radiomics in oncology, followed by a discussion on recent studies that use radiomics to predict immunotherapy response and toxicity.

AIR-Net: A novel multi-task learning method with auxiliary image reconstruction for predicting EGFR mutation status on CT images of NSCLC patients

Automated and accurate EGFR mutation status prediction using computed tomography (CT) imagery is of great value for tailoring optimal treatments to non-small cell lung cancer (NSCLC) patients. However, existing deep learning based methods usually adopt a single task learning strategy to design and train EGFR mutation status prediction models with limited training data, which may be insufficient to learn distinguishable representations for promoting prediction performance. In this paper, a novel multi-task learning method named AIR-Net is proposed to precisely predict EGFR mutation status on CT images. First, an auxiliary image reconstruction task is effectively integrated with EGFR mutation status prediction, aiming at providing extra supervision at the training phase. Particularly, we adequately employ multi-level information in a shared encoder to generate more comprehensive representations of tumors. Second, a powerful feature consistency loss is further introduced to constrain semantic consistency of original and reconstructed images, which contributes to enhanced image reconstruction and offers more effective regularization to AIR-Net during training. Performance analysis of AIR-Net indicates that auxiliary image reconstruction plays an essential role in identifying EGFR mutation status. Furthermore, extensive experimental results demonstrate that our method achieves favorable performance against other competitive prediction methods. All the results executed in this study suggest that the effectiveness and superiority of AIR-Net in precisely predicting EGFR mutation status of NSCLC.

Deep learning predicts epidermal growth factor receptor mutation subtypes in lung adenocarcinoma

PURPOSE

This study aimed to explore the predictive ability of deep learning (DL) for the common epidermal growth factor receptor (EGFR) mutation subtypes in patients with lung adenocarcinoma.

METHODS

A total of 665 patients with lung adenocarcinoma (528/137) were recruited from two different institutions. In the training set, an 18-layer convolutional neural network (CNN) and fivefold cross-validation strategy were used to establish a CNN model. Subsequently, an independent external validation cohort from the other institution was used to evaluate the predictive efficacy of the CNN model. Grad-weighted class activation mapping (Grad-CAM) technology was used for the visual interpretation of the CNN model. In addition, this study also compared the prediction abilities of the radiomics and CNN models. Receiver operating characteristic (ROC) curves, accuracy and precision values, and recall and F1-score were used to evaluate the effectiveness of the CNN model and compare its performance with that of the radiomics model.

RESULTS

In the validation set, the micro- and macroaverage values of the area under the ROC curve of the CNN model to identify the three EGFR subtypes were 0.78 and 0.79, respectively. All evaluation indicators of the CNN model were better than those of the radiomics model.

CONCLUSIONS

Our study confirmed the potential of DL for predicting the EGFR mutation status in lung adenocarcinoma. The imaging phenotypes of the three mutation subtypes were found to be different, which can provide a basis for choosing more accurate and personalized treatment in patients with lung adenocarcinoma.

Molecular subtyping of diffuse gliomas using magnetic resonance imaging: comparison and correlation between radiomics and deep learning

OBJECTIVES

The molecular subtyping of diffuse gliomas is important. The aim of this study was to establish predictive models based on preoperative multiparametric MRI.

METHODS

A total of 1016 diffuse glioma patients were retrospectively collected from Beijing Tiantan Hospital. Patients were randomly divided into the training (n = 780) and validation (n = 236) sets. According to the 2016 WHO classification, diffuse gliomas can be classified into four binary classification tasks (tasks I-IV). Predictive models based on radiomics and deep convolutional neural network (DCNN) were developed respectively, and their performances were compared with receiver operating characteristic (ROC) curves. Additionally, the radiomics and DCNN features were visualized and compared with the t-distributed stochastic neighbor embedding technique and Spearman's correlation test.

RESULTS

In the training set, areas under the curves (AUCs) of the DCNN models (ranging from 0.99 to 1.00) outperformed the radiomics models in all tasks, and the accuracies of the DCNN models (ranging from 0.90 to 0.94) outperformed the radiomics models in tasks I, II, and III. In the independent validation set, the accuracies of the DCNN models outperformed the radiomics models in all tasks (0.74-0.83), and the AUCs of the DCNN models (0.85-0.89) outperformed the radiomics models in tasks I, II, and III. DCNN features demonstrated more superior discriminative capability than the radiomics features in feature visualization analysis, and their general correlations were weak.

CONCLUSIONS

Both the radiomics and DCNN models could preoperatively predict the molecular subtypes of diffuse gliomas, and the latter performed better in most circumstances.

KEY POINTS

• The molecular subtypes of diffuse gliomas could be predicted with MRI. • Deep learning features tend to outperform radiomics features in large cohorts. • The correlation between the radiomics features and DCNN features was low.

Deep CNN Model Using CT Radiomics Feature Mapping Recognizes EGFR Gene Mutation Status of Lung Adenocarcinoma

To recognize the epidermal growth factor receptor (EGFR) gene mutation status in lung adenocarcinoma (LADC) has become a prerequisite of deciding whether EGFR-tyrosine kinase inhibitor (EGFR-TKI) medicine can be used. Polymerase chain reaction assay or gene sequencing is for measuring EGFR status, however, the tissue samples by surgery or biopsy are required. We propose to develop deep learning models to recognize EGFR status by using radiomics features extracted from non-invasive CT images. Preoperative CT images, EGFR mutation status and clinical data have been collected in a cohort of 709 patients (the primary cohort) and an independent cohort of 205 patients. After 1,037 CT-based radiomics features are extracted from each lesion region, 784 discriminative features are selected for analysis and construct a feature mapping. One Squeeze-and-Excitation (SE) Convolutional Neural Network (SE-CNN) has been designed and trained to recognize EGFR status from the radiomics feature mapping. SE-CNN model is trained and validated by using 638 patients from the primary cohort, tested by using the rest 71 patients (the internal test cohort), and further tested by using the independent 205 patients (the external test cohort). Furthermore, SE-CNN model is compared with machine learning (ML) models using radiomics features, clinical features, and both features. EGFR(-) patients show the smaller age, higher odds of female, larger lesion volumes, and lower odds of subtype of acinar predominant adenocarcinoma (APA), compared with EGFR(+). The most discriminative features are for texture (614, 78.3%) and the features of first order of intensity (158, 20.1%) and the shape features (12, 1.5%) follow. SE-CNN model can recognize EGFR mutation status with an AUC of 0.910 and 0.841 for the internal and external test cohorts, respectively. It outperforms the CNN model without SE, the fine-tuned VGG16 and VGG19, three ML models, and the state-of-art models. Utilizing radiomics feature mapping extracted from non-invasive CT images, SE-CNN can precisely recognize EGFR mutation status of LADC patients. The proposed method combining radiomics features and deep leaning is superior to ML methods and can be expanded to other medical applications. The proposed SE-CNN model may help make decision on usage of EGFR-TKI medicine.