Radiomics for Classification of Lung Cancer Histological Subtypes Based on Nonenhanced Computed Tomography

Enhanced lung cancer subtype classification using attention-integrated DeepCNN and radiomic features from CT images: a focus on feature reproducibility

OBJECTIVE

This study aims to assess a hybrid framework that combines radiomic features with deep learning and attention mechanisms to improve the accuracy of classifying lung cancer subtypes using CT images.

MATERIALS AND METHODS

A dataset of 2725 lung cancer images was used, covering various subtypes: adenocarcinoma (552 images), SCC (380 images), small cell lung cancer (SCLC) (307 images), large cell carcinoma (215 images), and pulmonary carcinoid tumors (180 images). The images were extracted as 2D slices from 3D CT scans, with tumor-containing slices selected from scans obtained across five healthcare centers. The number of slices per patient varied between 7 and 30, depending on tumor visibility. CT images were preprocessed using standardization, cropping, and Gaussian smoothing to ensure consistency across scans from different imaging instruments used at the centers. Radiomic features, including first-order statistics (FOS), shape-based, and texture-based features, were extracted using the PyRadiomics library. A DeepCNN architecture, integrated with attention mechanisms in the second convolutional block, was used for deep feature extraction, focusing on diagnostically important regions. The dataset was split into training (60%), validation (20%), and testing (20%) sets. Various feature selection techniques, such as Non-negative Matrix Factorization (NMF) and Recursive Feature Elimination (RFE), were used, and multiple machines learning models, including XGBoost and Stacking, were evaluated using accuracy, sensitivity, and AUC metrics. The model's reproducibility was validated using ICC analysis across different imaging conditions.

RESULTS

The hybrid model, which integrates DeepCNN with attention mechanisms, outperformed traditional methods. It achieved a testing accuracy of 92.47%, an AUC of 93.99%, and a sensitivity of 92.11%. XGBoost with NMF showed the best performance across all models, and the combination of radiomic and deep features improved classification further. Attention mechanisms played a key role in enhancing model performance by focusing on relevant tumor areas, reducing misclassification from irrelevant features. This also improved the performance of the 3D Autoencoder, boosting the AUC to 93.89% and accuracy to 93.24%.

CONCLUSIONS

This study shows that combining radiomic features with deep learning-especially when enhanced by attention mechanisms-creates a powerful and accurate framework for classifying lung cancer subtypes. Clinical trial number Not applicable.

Whole lung radiomic features are associated with overall survival in patients with locally advanced non-small cell lung cancer treated with definitive radiotherapy

BACKGROUND

Several studies have suggested that lung tissue heterogeneity is associated with overall survival (OS) in lung cancer. However, the quantitative relationship between the two remains unknown. The purpose of this study is to investigate the prognostic value of whole lung-based and tumor-based radiomics for OS in LA-NSCLC treated with definitive radiotherapy.

METHODS

A total of 661 patients with LA-NSCLC treated with definitive radiotherapy in combination with chemotherapy were enrolled in this study, with 292 patients in the training set, 57 patients from the same hospital from January to December 2017 as an independent test set (test-set-1), 83 patients from a multi-institutional prospective clinical trial data set (RTOG0617) as test-set-2, and 229 patients from a Dutch radiotherapy center as test-set-3. Tumor-based radiomic features and whole lung-based radiomic features were extracted from primary tumor and whole lungs (excluding the primary tumor) delineations in planning CT images. Feature selection of radiomic features was done by the least absolute shrinkage (LASSO) method embedded with a Cox proportional hazards (CPH) model with 5-fold cross-internal validation, with 1000 bootstrap samples. Radiomics prognostic scores (RS) were calculated by CPH regression based on selected features. Three models based on a tumor RS, and a lung RS separately and their combinations were constructed. The Harrell concordance index (C-index) and calibration curves were used to evaluate the discrimination and calibration performance. Patients were stratified into high and low risk groups based on median RS, and a log-rank test was performed.

RESULTS

The discrimination ability of lung- and tumor-based radiomics model was similar in terms of C-index, 0.69 vs. 0.68 in training set, 0.68 vs. 0.66 in test-set-1, 0.61 vs. 0.62 in test-set-2, 0.65 vs. 0.64 in test-set-3. The combination of tumor- and lung-based radiomics model performed best, with C-index of 0.71 in training set, 0.70 in test-set-1, 0.69 in test-set-2, and 0.68 in test-set-3. The calibration curve showed good agreement between predicted values and actual values. Patients were well stratified in training set, test-set-1 and test-set-3. In test-set-2, it was only whole lung-based RS that could stratify patients well and tumor-based RS performed bad.

CONCLUSION

Lung- and tumor-based radiomic features have the power to predict OS in LA-NSCLC. The combination of tumor- and lung-based radiomic features can achieve optimal performance.

Enhanced NSCLC subtyping and staging through attention-augmented multi-task deep learning: A novel diagnostic tool

OBJECTIVES

The objective of this study is to develop a novel multi-task learning approach with attention encoders for classifying histologic subtypes and clinical stages of non-small cell lung cancer (NSCLC), with superior performance compared to currently popular deep-learning models.

MATERIAL AND METHODS

Data were collected from six publicly available datasets in The Cancer Imaging Archive (TCIA). Following the inclusion and exclusion criteria, a total of 4548 CT slices from 758 cases were allocated. We evaluated multiple multi-task learning models that integrate attention mechanisms to resolve challenges in NSCLC subtype classification and clinical staging. These models utilized convolution-based modules in their shared layers for feature extraction, while the task layers were dedicated to histological subtype classification and staging. Each branch sequentially processed features through convolution-based and attention-based modules prior to classification.

RESULTS

Our study evaluated 758 NSCLC patients (mean age, 66.2 years ± 10.3; 473 men), spanning ADC and SCC cases. In the classification of histological subtypes and clinical staging of NSCLC, the MobileNet-based multi-task learning model enhanced with attention mechanisms (MN-MTL-A) demonstrated superior performance, achieving Area Under the Curve (AUC) scores of 0.963 (95 % CI: 0.943, 0.981) and 0.966 (95 % CI: 0.945, 0.982) for each task, respectively. The model significantly surpassed its counterparts lacking attention mechanisms and those configured for single-task learning, as evidenced by P-values of 0.01 or less for both tasks, according to DeLong's test.

CONCLUSIONS

The integration of attention encoder blocks into our multi-task learning network significantly enhanced the accuracy of NSCLC histological subtyping and clinical staging. Given the reduced reliance on precise radiologist annotation, our proposed model shows promising potential for clinical application.

Hybrid transformer-CNN and LSTM model for lung disease segmentation and classification

According to the World Health Organization (WHO) report, lung disorders are the third leading cause of mortality worldwide. Approximately three million individuals are affected with various types of lung disorders annually. This issue alarms us to take control measures related to early diagnostics, accurate treatment procedures, etc. The precise identification through the assessment of medical images is crucial for pulmonary disease diagnosis. Also, it remains a formidable challenge due to the diverse and unpredictable nature of pathological lung appearances and shapes. Therefore, the efficient lung disease segmentation and classification model is essential. By taking this initiative, a novel lung disease segmentation with a hybrid LinkNet-Modified LSTM (L-MLSTM) model is proposed in this research article. The proposed model utilizes four essential and fundamental steps for its implementation. The first step is pre-processing, where the input lung images are pre-processed using median filtering. Consequently, an improved Transformer-based convolutional neural network (CNN) model (ITCNN) is proposed to segment the affected region in the segmentation process. After segmentation, essential features such as texture, shape, color, and deep features are retrieved. Specifically, texture features are extracted using modified Local Gradient Increasing Pattern (LGIP) and Multi-texton analysis. Then, the classification step utilizes a hybrid model, the L-MLSTM model. This work leverages two datasets such as the COVID-19 normal pneumonia-CT images dataset (Dataset 1) and the Chest CT scan images dataset (Dataset 2). The dataset is crucial for training and evaluating the model, providing a comprehensive basis for robust and generalizable results. The L-MLSTM model outperforms several existing models, including HDE-NN, DBN, LSTM, LINKNET, SVM, Bi-GRU, RNN, CNN, and VGG19 + CNN, with accuracies of 89% and 95% at learning percentages of 70 and 90, respectively, for datasets 1 and 2. The improved accuracy achieved by the L-MLSTM model highlights its capability to better handle the complexity and variability in lung images. This hybrid approach enhances the model's ability to distinguish between different types of lung diseases and reduces diagnostic errors compared to existing methods.

Radiomics in distinguishing between lung adenocarcinoma and lung squamous cell carcinoma: a systematic review and meta-analysis

Objectives

The aim of this study was to systematically review the studies on radiomics models in distinguishing between lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) and evaluate the classification performance of radiomics models using images from various imaging techniques.

Materials and methods

PubMed, Embase and Web of Science Core Collection were utilized to search for radiomics studies that differentiate between LUAD and LUSC. The assessment of the quality of studies included utilized the improved Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) and Radiomics Quality Score (RQS). Meta-analysis was conducted to assess the classification performance of radiomics models using various imaging techniques.

Results

The qualitative analysis included 40 studies, while the quantitative synthesis included 21 studies. Median RQS for 40 studies was 12 (range -5~19). Sixteen studies were deemed to have a low risk of bias and low concerns regarding applicability. The radiomics model based on CT images had a pooled sensitivity of 0.78 (95%CI: 0.71~0.83), specificity of 0.85 (95%CI:0.73~0.92), and the area under summary receiver operating characteristic curve (SROC-AUC) of 0.86 (95%CI:0.82~0.89). As for PET images, the pooled sensitivity was 0.80 (95%CI: 0.61~0.91), specificity was 0.77 (95%CI: 0.60~0.88), and the SROC-AUC was 0.85 (95%CI: 0.82~0.88). PET/CT images had a pooled sensitivity of 0.87 (95%CI: 0.72~0.94), specificity of 0.88 (95%CI: 0.80~0.93), and an SROC-AUC of 0.93 (95%CI: 0.91~0.95). MRI images had a pooled sensitivity of 0.73 (95%CI: 0.61~0.82), specificity of 0.80 (95%CI: 0.65~0.90), and an SROC-AUC of 0.79 (95%CI: 0.75~0.82).

Conclusion

Radiomics models demonstrate potential in distinguishing between LUAD and LUSC. Nevertheless, it is crucial to conduct a well-designed and powered prospective radiomics studies to establish their credibility in clinical application.

Systematic review registration

https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=412851, identifier CRD42023412851.

Impact of Preprocessing Parameters in Medical Imaging-Based Radiomic Studies: A Systematic Review

BACKGROUND

Lately, radiomic studies featuring the development of a signature to use in prediction models in diagnosis or prognosis outcomes have been increasingly published. While the results are shown to be promising, these studies still have many pitfalls and limitations. One of the main issues of these studies is that radiomic features depend on how the images are preprocessed before their computation. Since, in widely known and used software for radiomic features calculation, it is possible to set these preprocessing parameters before the calculation of the radiomic feature, there are ongoing studies assessing the stability and repeatability of radiomic features to find the most suitable preprocessing parameters for every used imaging modality.

MATERIALS AND METHODS

We performed a comprehensive literature search using four electronic databases: PubMed, Cochrane Library, Embase, and Scopus. Mesh terms and free text were modeled in search strategies for databases. The inclusion criteria were studies where preprocessing parameters' influence on feature values and model predictions was addressed. Records lacking information on image acquisition parameters were excluded, and any eligible studies with full-text versions were included in the review process, while conference proceedings and monographs were disregarded. We used the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2) tool to investigate the risk of bias. We synthesized our data in a table divided by the imaging modalities subgroups.

RESULTS

After applying the inclusion and exclusion criteria, we selected 43 works. This review examines the impact of preprocessing parameters on the reproducibility and reliability of radiomic features extracted from multimodality imaging (CT, MRI, CBCT, and PET/CT). Standardized preprocessing is crucial for consistent radiomic feature extraction. Key preprocessing steps include voxel resampling, normalization, and discretization, which influence feature robustness and reproducibility. In total, 44% of the included works studied the effects of an isotropic voxel resampling, and most studies opted to employ a discretization strategy. From 2021, several studies started selecting the best set of preprocessing parameters based on models' best performance. As for comparison metrics, ICC was the most used in MRI studies in 58% of the screened works.

CONCLUSIONS

From our work, we highlighted the need to harmonize the use of preprocessing parameters and their values, especially in light of future studies of prospective studies, which are still lacking in the current literature.

Lung Cancer Subtyping: A Short Review

As of 2022, lung cancer is the most commonly diagnosed cancer worldwide, with the highest mortality rate. There are three main histological types of lung cancer, and it is more important than ever to accurately identify the subtypes since the development of personalized, type-specific targeted therapies that have improved mortality rates. Traditionally, the gold standard for the confirmation of histological subtyping is tissue biopsy and histopathology. This, however, comes with its own challenges, which call for newer sampling techniques and adjunctive tools to assist in and improve upon the existing diagnostic workflow. This review aims to list and describe studies from the last decade (n = 47) that investigate three such potential omics techniques-namely (1) transcriptomics, (2) proteomics, and (3) metabolomics, as well as immunohistochemistry, a tool that has already been adopted as a diagnostic adjunct. The novelty of this review compared to similar comprehensive studies lies with its detailed description of each adjunctive technique exclusively in the context of lung cancer subtyping. Similarities between studies evaluating individual techniques and markers are drawn, and any discrepancies are addressed. The findings of this study indicate that there is promising evidence that supports the successful use of omics methods as adjuncts to the subtyping of lung cancer, thereby directing clinician practice in an economical and less invasive manner.

Contrast Enhanced CT Radiogenomics in a Retrospective NSCLC Cohort: Models, Attempted Validation of a Published Model and the Relevance of the Clinical Context

RATIONALE AND OBJECTIVE

To develop a radiogenomic predictive model for non-small cell lung cancer (NSCLC) patients studied through contrast enhanced chest computed tomography (CE-CT) targeting the most frequent gene alterations. M&M: A retrospective study of patients with NSCLC imaged with CE-CT before treatment and had their tumor genomics sequenced at our institution was performed. Data was gathered from their imaging studies, their electronic medical records and a web-based database search (cBioPortal.ca). All of the patient data was tabulated for analysis. Two predictive models (M1 & M2) were created using different approaches and a third model was extracted from the literature to also be tested in our population.

RESULTS

Out of 157 patients, eighty were male (51%) and 124 (79%) had a history of smoking. The three most prevalent genes were KRAS, TP53 and EGFR. The M1 radiomics-only model median AUC were 0.61 (TP53), 0.53 (KRAS) and 0.64 (EGFR) and for M1 radiomics + clinical were 0.61 (TP53), 0.61 (KRAS) and 0.80 (EGFR). The M2 radiomics-only model median AUC were 0.63 (TP53), 0.60 (KRAS) and 0.65 (EGFR) and for M2 radiomics + clinical were 0.64 (TP53), 0.62 (KRAS) and 0.81 (EGFR). The external EGFR radiomic model showed an AUC of 0.69 and 0.86 for the radiomics-only and combined radiomics + clinical respectively.

CONCLUSION

Our study was able to provide robust predictive radiomics model evaluation for the detection of TP53, KRAS and EGFR. We also compared our performance with an already published model and observed how impactful clinical variables can be on models' performance.

CLINICAL RELEVANCE STATEMENT

Identifying tumor mutations in patients that can't undergo biopsy is critical for their outcomes.

KEYPOINTS

• Tumor genomic profiling is critical for treatment selection • CE-CT radiomics produce robust predictive models comparable to those already published • Clinical variables should be considered/included in predictive models.

Advancing NSCLC pathological subtype prediction with interpretable machine learning: a comprehensive radiomics-based approach

Objective

This research aims to develop and assess the performance of interpretable machine learning models for diagnosing three histological subtypes of non-small cell lung cancer (NSCLC) utilizing CT imaging data.

Methods

A retrospective cohort of 317 patients diagnosed with NSCLC was included in the study. These individuals were randomly segregated into two groups: a training set comprising 222 patients and a validation set with 95 patients, adhering to a 7:3 ratio. A comprehensive extraction yielded 1,834 radiomic features. For feature selection, statistical methodologies such as the Mann-Whitney U test, Spearman's rank correlation, and one-way logistic regression were employed. To address data imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was utilized. The study designed three distinct models to predict adenocarcinoma (ADC), squamous cell carcinoma (SCC), and large cell carcinoma (LCC). Six different classifiers, namely Logistic Regression, Support Vector Machine, Decision Tree, Random Forest, eXtreme Gradient Boosting (XGB), and LightGBM, were deployed for model training. Model performance was gauged through accuracy metrics and the area under the receiver operating characteristic (ROC) curves (AUC). To interpret the diagnostic process, the Shapley Additive Explanations (SHAP) approach was applied.

Results

For the ADC, SCC, and LCC groups, 9, 12, and 8 key radiomic features were selected, respectively. In terms of model performance, the XGB model demonstrated superior performance in predicting SCC and LCC, with AUC values of 0.789 and 0.848, respectively. For ADC prediction, the Random Forest model excelled, showcasing an AUC of 0.748.

Conclusion

The constructed machine learning models, leveraging CT imaging, exhibited robust predictive capabilities for SCC, LCC, and ADC subtypes of NSCLC. These interpretable models serve as substantial support for clinical decision-making processes.

Effectiveness of CT radiomic features combined with clinical factors in predicting prognosis in patients with limited-stage small cell lung cancer

BACKGROUND

The prognosis of SCLC is poor and difficult to predict. The aim of this study was to explore whether a model based on radiomics and clinical features could predict the prognosis of patients with limited-stage small cell lung cancer (LS-SCLC).

METHODS

Simulated positioning CT images and clinical features were retrospectively collected from 200 patients with histological diagnosis of LS-SCLC admitted between 2013 and 2021, which were randomly divided into the training (n = 140) and testing (n = 60) groups. Radiomics features were extracted from simulated positioning CT images, and the t-test and the least absolute shrinkage and selection operator (LASSO) were used to screen radiomics features. We then constructed radiomic score (RadScore) based on the filtered radiomics features. Clinical factors were analyzed using the Kaplan-Meier method. The Cox proportional hazards model was used for further analyses of possible prognostic features and clinical factors to build three models including a radiomic model, a clinical model, and a combined model including clinical factors and RadScore. When a model has prognostic predictive value (AUC > 0.7) in both train and test groups, a nomogram will be created. The performance of three models was evaluated using area under the receiver operating characteristic curve (AUC) and Kaplan-Meier analysis.

RESULTS

A total of 1037 features were extracted from simulated positioning CT images which were contrast enhanced CT of the chest. The combined model showed the best prediction, with very poor AUC for the radiomic model and the clinical model. The combined model of OS included 4 clinical features and RadScore, with AUCs of 0.71 and 0.70 in the training and test groups. The combined model of PFS included 4 clinical features and RadScore, with AUCs of 0.72 and 0.71 in the training and test groups. T stages, ProGRP and smoke status were the independent variables for OS in the combined model, whereas T stages, ProGRP and prophylactic cranial irradiation (PCI) were the independent factors for PFS. There was a statistically significant difference between the low- and high-risk groups in the combined model of OS (training group, p < 0.0001; testing group, p = 0.0269) and PFS (training group, p < 0.0001; testing group, p < 0.0001).

CONCLUSION

Combined models involved RadScore and clinical factors can predict prognosis in LS-SCLC and show better performance than individual radiomics and clinical models.

A combination of radiomic features, clinic characteristics, and serum tumor biomarkers to predict the possibility of the micropapillary/solid component of lung adenocarcinoma

BACKGROUND

Invasive lung adenocarcinoma with MPP/SOL components has a poor prognosis and often shows a tendency to recurrence and metastasis. This poor prognosis may require adjustment of treatment strategies. Preoperative identification is essential for decision-making for subsequent treatment.

OBJECTIVE

This study aimed to preoperatively predict the probability of MPP/SOL components in lung adenocarcinomas by a comprehensive model that includes radiomics features, clinical characteristics, and serum tumor biomarkers.

DESIGN

A retrospective case control, diagnostic accuracy study.

METHODS

This study retrospectively recruited 273 patients (males: females, 130: 143; mean age ± standard deviation, 63.29 ± 10.03 years; range 21-83 years) who underwent resection of invasive lung adenocarcinoma. Sixty-one patients (22.3%) were diagnosed with lung adenocarcinoma with MPP/SOL components. Radiomic features were extracted from CT before surgery. Clinical, radiomic, and combined models were developed using the logistic regression algorithm. The clinical and radiomic signatures were integrated into a nomogram. The diagnostic performance of the models was evaluated using the area under the curve (AUC). Studies were scored according to the Radiomics Quality Score and Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis guidelines.

RESULTS

The radiomics model achieved the best AUC values of 0.858 and 0.822 in the training and test cohort, respectively. Tumor size (T_size), solid tumor size (ST_size), consolidation-to-tumor ratio (CTR), years of smoking, CYFRA 21-1, and squamous cell carcinoma antigen were used to construct the clinical model. The clinical model achieved AUC values of 0.741 and 0.705 in the training and test cohort, respectively. The nomogram showed higher AUCs of 0.894 and 0.843 in the training and test cohort, respectively.

CONCLUSION

This study has developed and validated a combined nomogram, a visual tool that integrates CT radiomics features with clinical indicators and serum tumor biomarkers. This innovative model facilitates the differentiation of micropapillary or solid components within lung adenocarcinoma and achieves a higher AUC, indicating superior predictive accuracy.

Prediction of Interval Growth of Lung Adenocarcinomas Manifesting as Persistent Subsolid Nodules ≤3 cm Based on Radiomic Features

RATIONALES AND OBJECTIVES

To investigate the prognostic value of the radiomic-based prediction model in predicting the interval growth rate of persistent subsolid nodules (SSNs) with an initial size of ≤ 3 cm manifesting as lung adenocarcinomas.

MATERIALS AND METHODS

A total of 133 patients (mean age, 59.02 years; male, 37.6%) with 133 SSNs who underwent a series of CT examinations at our hospital between 2012 and 2022 were included in this study. Forty-one radiomic features were extracted from each volumetric region of interest. Radiomic features combined with conventional clinical and semantic parameters were then selected for radiomic-based model building. To investigate the model performance in terms of substantial SSN growth and stage shift growth, the model performance was compared by the area under the curve (AUC) obtained by receiver operating characteristic analysis.

RESULTS

The mean follow-up period was 3.62 years. For substantial SSN growth, a radiomic-based model (Model 2) based on clinical characteristics, CT semantic features, and radiomic features yielded an AUCs of 0.869 (95% CI: 0.799-0.922). In comparison with Model 1 (clinical characteristics and CT semantic features), Model 2 performed better than Model 1 for substantial SSN growth (AUC model 1:0.793 versus AUC model 2:0.869, p = 0.028). A radiomic-based nomogram combining sex, follow-up period, and three radiomic features was built for substantial SSN growth prediction. For the stage shift growth, a radiomic-based model (Model 4) based on clinical characteristics, CT semantic features, and radiomic features yielded an AUCs of 0.883 (95% CI: 0.815-0.933). Compared with Model 3 (clinical characteristics and CT semantic features), Model 4 performed better than the model 3 for stage shift growth (AUC model 1: 0.769 versus AUC model 2: 0.883, p = 0.006). A radiomic-based nomogram combining the initial nodule size, SSN classification, follow-up period, and three radiomic features was built to predict the stage shift growth.

CONCLUSION

Radiomic-based models have superior utility in estimating the prognostic interval growth of patients with early lung adenocarcinomas (≤ 3 cm) than conventional clinical-semantic models in terms of substantial interval growth and stage shift growth, potentially guiding clinical decision-making with follow-up strategies of SSNs in personalized precision medicine.

Computed Tomography Radiomics-based Prediction Model for Gender-Age-Physiology Staging of Connective Tissue Disease-associated Interstitial Lung Disease

PURPOSE

To analyze the feasibility of predicting gender-age-physiology (GAP) staging in patients with connective tissue disease-associated interstitial lung disease (CTD-ILD) by radiomics based on computed tomography (CT) of the chest.

MATERIALS AND METHODS

Chest CT images of 184 patients with CTD-ILD were retrospectively analyzed. GAP staging was performed on the basis of gender, age, and pulmonary function test results. GAP I, II, and III have 137, 36, and 11 cases, respectively. The cases in GAP Ⅱ and Ⅲ were then combined into one group, and the two groups of patients were randomly divided into the training and testing groups with a 7:3 ratio. The radiomics features were extracted using AK software. Multivariate logistic regression analysis was then conducted to establish a radiomics model. A nomogram model was established on the basis of Rad-score and clinical factors (age and gender).

RESULTS

For the radiomics model, four significant radiomics features were selected to construct the model and showed excellent ability to differentiate GAP I from GAP Ⅱ and Ⅲ in both the training group (the area under the curve [AUC] = 0.803, 95% confidence interval [CI]: 0.724-0.874) and testing group (AUC = 0.801, 95% CI:0.663-0.912). The nomogram model that combined clinical factors and radiomics features improved higher accuracy of both training (88.4% vs. 82.1%) and testing (83.3% vs. 79.2%).

CONCLUSION

The disease severity assessment of patients with CTD-ILD can be evaluated by applying the radiomics method based on CT images. The nomogram model demonstrates better performance for predicting the GAP staging.

Deep learning-enhanced radiomics for histologic classification and grade stratification of stage IA lung adenocarcinoma: a multicenter study

Background

Preoperative prediction models for histologic subtype and grade of stage IA lung adenocarcinoma (LUAD) according to the update of the WHO Classification of Tumors of the Lung in 2021 and the 2020 new grade system are yet to be explored. We aim to develop the noninvasive pathology and grade evaluation approach for patients with stage IA LUAD via CT-based radiomics approach and evaluate their performance in clinical practice.

Methods

Chest CT scans were retrospectively collected from patients who were diagnosed with stage IA LUAD and underwent complete resection at two hospitals. A deep learning segmentation algorithm was first applied to assist lesion delineation. Expansion strategies such as bounding-box annotations were further applied. Radiomics features were then extracted and selected followed by radiomics modeling based on four classic machine learning algorithms for histologic subtype classification and grade stratification. The area under the receiver operating characteristic curve (AUC) was used to evaluate model performance.

Results

The study included 294 and 145 patients with stage IA LUAD from two hospitals for radiomics analysis, respectively. For classification of four histological subtypes, multilayer perceptron (MLP) algorithm presented no annotation strategy preference and achieved the average AUC of 0.855, 0.922, and 0.720 on internal, independent, and external test sets with 1-pixel expansion annotation. Bounding-box annotation strategy also enabled MLP an acceptable and stable accuracy among test sets. Meanwhile, logistic regression was selected for grade stratification and achieved the average AUC of 0.928, 0.837, and 0.748 on internal, independent, and external test sets with optimal annotation strategies.

Conclusions

DL-enhanced radiomics models had great potential to predict the fine histological subtypes and grades of early-stage LUADs based on CT images, which might serve as a promising noninvasive approach for the diagnosis and management of early LUADs.

Performance of alternative manual and automated deep learning segmentation techniques for the prediction of benign and malignant lung nodules

Purpose

We aim to evaluate the performance of radiomic biopsy (RB), best-fit bounding box (BB), and a deep-learning-based segmentation method called no-new-U-Net (nnU-Net), compared to the standard full manual (FM) segmentation method for predicting benign and malignant lung nodules using a computed tomography (CT) radiomic machine learning model.

Materials and Methods

A total of 188 CT scans of lung nodules from 2 institutions were used for our study. One radiologist identified and delineated all 188 lung nodules, whereas a second radiologist segmented a subset (n = 20 ) of these nodules. Both radiologists employed FM and RB segmentation methods. BB segmentations were generated computationally from the FM segmentations. The nnU-Net, a deep-learning-based segmentation method, performed automatic nodule detection and segmentation. The time radiologists took to perform segmentations was recorded. Radiomic features were extracted from each segmentation method, and models to predict benign and malignant lung nodules were developed. The Kruskal-Wallis and DeLong tests were used to compare segmentation times and areas under the curve (AUC), respectively.

Results

For the delineation of the FM, RB, and BB segmentations, the two radiologists required a median time (IQR) of 113 (54 to 251.5), 21 (9.25 to 38), and 16 (12 to 64.25) s, respectively (p = 0.04 ). In dataset 1, the mean AUC (95% CI) of the FM, RB, BB, and nnU-Net model were 0.964 (0.96 to 0.968), 0.985 (0.983 to 0.987), 0.961 (0.956 to 0.965), and 0.878 (0.869 to 0.888). In dataset 2, the mean AUC (95% CI) of the FM, RB, BB, and nnU-Net model were 0.717 (0.705 to 0.729), 0.919 (0.913 to 0.924), 0.699 (0.687 to 0.711), and 0.644 (0.632 to 0.657).

Conclusion

Radiomic biopsy-based models outperformed FM and BB models in prediction of benign and malignant lung nodules in two independent datasets while deep-learning segmentation-based models performed similarly to FM and BB. RB could be a more efficient segmentation method, but further validation is needed.

Classification of Histological Types and Stages in Non-small Cell Lung Cancer Using Radiomic Features Based on CT Images

Non-invasive diagnostic method based on radiomic features in patients with non-small cell lung cancer (NSCLC) has attracted attention. This study aimed to develop a CT image-based model for both histological typing and clinical staging of patients with NSCLC. A total of 309 NSCLC patients with 537 CT series from The Cancer Imaging Archive (TCIA) database were included in this study. All patients were randomly divided into the training set (247 patients, 425 CT series) and testing set (62 patients, 112 CT series). A total of 107 radiomic features were extracted. Four classifiers including random forest, XGBoost, support vector machine, and logistic regression were used to construct the classification model. The classification model had two output layers: histological type (adenocarcinoma, squamous cell carcinoma, and large cell) and clinical stage (I, II, and III) of NSCLC patients. The area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) with 95% confidence interval (CI) were utilized to evaluate the performance of the model. Seven features were selected for inclusion in the classification model. The random forest model had the best classification ability compared with other classifiers. The AUC of the RF model for histological typing and clinical staging of NSCLC patients in the testing set was 0.700 (95% CI, 0.641-0.759) and 0.881 (95% CI, 0.842-0.920), respectively. The CT image-based radiomic feature model had good classification ability for both histological typing and clinical staging of patients with NSCLC.

Oligodendrocyte Transcription Factor 2 as a Potential Prognostic Biomarker of Glioblastoma: Kaplan-Meier Analysis and the Development of a Binary Predictive Model Based on Visually Accessible Rembrandt Image and Magnetic Resonance Imaging Radiomic Features

OBJECTIVE

Oligodendrocyte transcription factor 2 (OLIG2) is universally expressed in human glioblastoma (GB). Our study explores whether OLIG2 expression impacts GB patients' overall survival and establishes a machine learning model for OLIG2 level prediction in patients with GB based on clinical, semantic, and magnetic resonance imaging radiomic features.

METHODS

Kaplan-Meier analysis was used to determine the optimal cutoff value of the OLIG2 in 168 GB patients. Three hundred thirteen patients enrolled in the OLIG2 prediction model were randomly divided into training and testing sets in a ratio of 7:3. The radiomic, semantic, and clinical features were collected for each patient. Recursive feature elimination (RFE) was used for feature selection. The random forest (RF) model was built and fine-tuned, and the area under the curve was calculated to evaluate the performance. Finally, a new testing set excluding IDH-mutant patients was built and tested in a predictive model using the fifth edition of the central nervous system tumor classification criteria.

RESULTS

One hundred nineteen patients were included in the survival analysis. Oligodendrocyte transcription factor 2 was positively associated with GB survival, with an optimal cutoff of 10% (P = 0.00093). One hundred thirty-four patients were eligible for the OLIG2 prediction model. An RFE-RF model based on 2 semantic and 21 radiomic signatures achieved areas under the curve of 0.854 in the training set, 0.819 in the testing set, and 0.825 in the new testing set.

CONCLUSIONS

Glioblastoma patients with ≤10% OLIG2 expression tended to have worse overall survival. An RFE-RF model integrating 23 features can predict the OLIG2 level of GB patients preoperatively, irrespective of the central nervous system classification criteria, further guiding individualized treatment.

Research on automatic classification technology of kidney tumor and normal kidney tissue based on computed tomography radiomics

Purpose

By using a radiomics-based approach, multiple radiomics features can be extracted from regions of interest in computed tomography (CT) images, which may be applied to automatically classify kidney tumors and normal kidney tissues. The study proposes a method based on CT radiomics and aims to use extracted radiomics features to automatically classify of kidney tumors and normal kidney tissues and to establish an automatic classification model.

Methods

CT data were retrieved from the 2019 Kidney and Kidney Tumor Segmentation Challenge (KiTS19) in The Cancer Imaging Archive (TCIA) open access database. Arterial phase-enhanced CT images from 210 cases were used to establish an automatic classification model. These CT images of patients were randomly divided into training (168 cases) and test (42 cases) sets. Furthermore, the radiomics features of gross tumor volume (GTV) and normal kidney tissues in the training set were extracted and screened, and a binary logistic regression model was established. For the test set, the radiomic features and cutoff value of P were consistent with the training set.

Results

Three radiomics features were selected to establish the binary logistic regression model. The accuracy (ACC), sensitivity (SENS), specificity (SPEC), area under the curve (AUC), and Youden index of the training and test sets based on the CT radiomics classification model were all higher than 0.85.

Conclusion

The automatic classification model of kidney tumors and normal kidney tissues based on CT radiomics exhibited good classification ability. Kidney tumors could be distinguished from normal kidney tissues. This study may complement automated tumor delineation techniques and warrants further research.

Molecular Characterization and Therapeutic Approaches to Small Cell Lung Cancer: Imaging Implications

Small cell lung cancer (SCLC) is a highly aggressive malignancy with exceptionally poor prognosis, comprising approximately 15% of lung cancers. Emerging knowledge of the molecular and genomic landscape of SCLC and recent successful clinical applications of new systemic agents have allowed for precision oncology treatment approaches. Imaging is essential for the diagnosis, staging, and treatment monitoring of patients with SCLC. The role of imaging is increasing with the approval of new treatment agents, including immune checkpoint inhibitors, which lead to novel imaging manifestations of response and toxicities. The purpose of this state-of-the-art review is to provide the reader with the latest information about SCLC, focusing on the subtyping of this malignancy (molecular characterization) and the emerging systemic therapeutic approaches and their implications for imaging. The review will also discuss the future directions of SCLC imaging, radiomics and machine learning.

CT radiomics-based long-term survival prediction for locally advanced non-small cell lung cancer patients treated with concurrent chemoradiotherapy using features from tumor and tumor organismal environment

Background

Definitive concurrent chemoradiotherapy (CCRT) is the standard treatment for locally advanced non-small cell lung cancer (LANSCLC) patients, but the treatment response and survival outcomes varied among these patients. We aimed to identify pretreatment computed tomography-based radiomics features extracted from tumor and tumor organismal environment (TOE) for long-term survival prediction in these patients treated with CCRT.

Methods

A total of 298 eligible patients were randomly assigned into the training cohort and validation cohort with a ratio 2:1. An integrated feature selection and model training approach using support vector machine combined with genetic algorithm was performed to predict 3-year overall survival (OS). Patients were stratified into the high-risk and low-risk group based on the predicted survival status. Pulmonary function test and blood gas analysis indicators were associated with radiomic features. Dynamic changes of peripheral blood lymphocytes counts before and after CCRT had been documented.

Results

Nine features including 5 tumor-related features and 4 pulmonary features were selected in the predictive model. The areas under the receiver operating characteristic curve for the training and validation cohort were 0.965 and 0.869, and were reduced by 0.179 and 0.223 when all pulmonary features were excluded. Based on radiomics-derived stratification, the low-risk group yielded better 3-year OS (68.4% vs. 3.3%, p < 0.001) than the high-risk group. Patients in the low-risk group had better baseline FEV1/FVC% (96.3% vs. 85.9%, p = 0.046), less Grade ≥ 3 lymphopenia during CCRT (63.2% vs. 83.3%, p = 0.031), better recovery of lymphopenia from CCRT (71.4% vs. 27.8%, p < 0.001), lower incidence of Grade ≥ 2 radiation-induced pneumonitis (31.6% vs. 53.3%, p = 0.040), superior tumor remission (84.2% vs. 66.7%, p = 0.003).

Conclusion

Pretreatment radiomics features from tumor and TOE could boost the long-term survival forecast accuracy in LANSCLC patients, and the predictive results could be utilized as an effective indicator for survival risk stratification. Low-risk patients might benefit more from radical CCRT and further adjuvant immunotherapy.

Trial registration:

retrospectively registered.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13014-022-02136-w.

Circ-EIF3I facilitates proliferation, migration, and invasion of lung cancer via regulating the activity of Wnt/β-catenin pathway through the miR-1253/NOVA2 axis

Many studies have shown that circular RNA (circRNA) is an important regulator mediating the malignant progression of cancer. However, the role and mechanism of circ-EIF3I in lung cancer (LC) development are still unclear. A total 36 paired LC tumor tissues and adjacent normal tissues were enrolled. The expression of circ-EIF3I, microRNA (miR)-1253, and neuro-oncological ventral antigen 2 (NOVA2) was measured by quantitative real-time PCR. The proliferation, apoptosis, migration, and invasion of LC cells were determined by MTT assay, colony formation assay, flow cytometry, and transwell assay. Dual-luciferase reporter assay was performed to verify the interaction between miR-1253 and circ-EIF3I or NOVA2. The protein levels of NOVA2 and Wnt/β-catenin pathway-related markers were detected by western blot analysis. Xenograft tumor was constructed to explore the function of circ-EIF3I on LC tumor growth. Circ-EIF3I was upregulated in LC tumor tissues and cells. Silenced circ-EIF3I could suppress the proliferation, migration, invasion, and enhance the apoptosis of LC cells in vitro, as well as reduce LC tumor growth in vivo. Circ-EIF3I could sponge miR-1253, and miR-1253 inhibitor overturned the regulation of circ-EIF3I knockdown on LC cell progression. NOVA2 was confirmed to be a target of miR-1253, which could reverse the inhibitory effects of miR-1253 on LC cell progression. Further experiments showed that circ-EIF3I regulated NOVA2 expression by sponging miR-1253. In addition, circ-EIF3I silencing could inhibit the activity of Wnt/β-catenin pathway via regulating the miR-1253/NOVA2 axis. Circ-EIF3I might function as an oncogene in LC, which promoted LC progression by the miR-1253/NOVA2/Wnt/β-catenin network.

The Impact of Image Acquisition Parameters and ComBat Harmonization on the Predictive Performance of Radiomics: A Renal Cell Carcinoma Model

Radiomics, one of the potential methods for developing clinical biomarker, is one of the exponentially growing research fields. In addition to its potential, several limitations have been identified in this field, and most importantly the effects of variations in imaging parameters on radiomic features (RFs). In this study, we investigate the potential of RFs to predict overall survival in patients with clear cell renal cell carcinoma, as well as the impact of ComBat harmonization on the performance of RF models. We assessed the robustness of the results by performing the analyses a thousand times. Publicly available CT scans of 179 patients were retrospectively collected and analyzed. The scans were acquired using different imaging vendors and parameters in different medical centers. The performance was calculated by averaging the metrics over all runs. On average, the clinical model significantly outperformed the radiomic models. The use of ComBat harmonization, on average, did not significantly improve the performance of radiomic models. Hence, the variability in image acquisition and reconstruction parameters significantly affect the performance of radiomic models. The development of radiomic specific harmonization techniques remain a necessity for the advancement of the field.

Malignancy risk of gastrointestinal stromal tumors evaluated with noninvasive radiomics: A multi-center study

Purpose

This study was to investigate the diagnostic efficacy of radiomics models based on the enhanced CT images in differentiating the malignant risk of gastrointestinal stromal tumors (GIST) in comparison with the clinical indicators model and traditional CT diagnostic criteria.

Materials and methods

A total of 342 patients with GISTs confirmed histopathologically were enrolled from five medical centers. Data of patients wrom two centers comprised the training group (n=196), and data from the remaining three centers constituted the validation group (n=146). After CT image segmentation and feature extraction and selection, the arterial phase model and venous phase model were established. The maximum diameter of the tumor and internal necrosis were used to establish a clinical indicators model. The traditional CT diagnostic criteria were established for the classification of malignant potential of tumor. The performance of the four models was assessed using the receiver operating characteristics curve.

Reuslts

In the training group, the area under the curves(AUCs) of the arterial phase model, venous phase model, clinical indicators model, and traditional CT diagnostic criteria were 0.930 [95% confidence interval (CI): 0.895-0.965), 0.933 (95%CI 0.898-0.967), 0.917 (95%CI 0.872-0.961) and 0.782 (95%CI 0.717-0.848), respectively. In the validation group, the AUCs of the models were 0.960 (95%CI 0.930-0.990), 0.961 (95% CI 0.930-0.992), 0.922 (95%CI 0.884-0.960) and 0.768 (95%CI 0.692-0.844), respectively. No significant difference was detected in the AUC between the arterial phase model, venous phase model, and clinical indicators model by the DeLong test, whereas a significant difference was observed between the traditional CT diagnostic criteria and the other three models.

Conclusion

The radiomics model using the morphological features of GISTs play a significant role in tumor risk stratification and can provide a reference for clinical diagnosis and treatment plan.

A Genetic Folding Strategy Based Support Vector Machine to Optimize Lung Cancer Classification

Cancer is defined as an abnormal growth of human cells classified into benign and malignant. The site makes further classification of cancers of initiation and genomic underpinnings. Lung cancer displays extreme heterogeneity, making genomic classification vital for future targeted therapies. Especially considering lung cancers account for 1.76 million deaths worldwide annually. However, tumors do not always correlate to cancer as they can be benign, severely dysplastic (pre-cancerous), or malignant (cancerous). Lung cancer presents with ambiguous symptoms, thus is difficult to diagnose and is detected later compared to other cancers. Diagnosis relies heavily on radiology and invasive procedures. Different models developed employing Artificial Intelligence (AI), and Machine Learning (ML) have been used to classify various cancers. In this study, the authors propose a Genetic Folding Strategy (GFS) based model to predict lung cancer from a lung cancer dataset. We developed and implemented GF to improve Support Vector Machines (SVM) classification kernel functions and used it to classify lung cancer. We developed and implemented GF to improve SVM classification kernel functions and used it to classify lung cancer. Classification performance evaluations and comparisons between the authors' GFS model and three SVM kernels, linear, polynomial and radial basis function, were conducted thoroughly on real lung cancer datasets. While using GFS in classifying lung cancer, the authors obtained an accuracy of 96.2%. This is the highest current accuracy compared to other kernels.

Machine learning-based radiomics for histological classification of parotid tumors using morphological MRI: a comparative study

OBJECTIVES

To evaluate the effectiveness of machine learning models based on morphological magnetic resonance imaging (MRI) radiomics in the classification of parotid tumors.

METHODS

In total, 298 patients with parotid tumors were randomly assigned to a training and test set at a ratio of 7:3. Radiomics features were extracted from the morphological MRI images and screened using the Select K Best and LASSO algorithm. Three-step machine learning models with XGBoost, SVM, and DT algorithms were developed to classify the parotid neoplasms into four subtypes. The ROC curve was used to measure the performance in each step. Diagnostic confusion matrices of these models were calculated for the test cohort and compared with those of the radiologists.

RESULTS

Six, twelve, and eight optimal features were selected in each step of the three-step process, respectively. XGBoost produced the highest area under the curve (AUC) for all three steps in the training cohort (0.857, 0.882, and 0.908, respectively), and for the first step in the test cohort (0.826), but produced slightly lower AUCs than SVM in the latter two steps in the test cohort (0.817 vs. 0.833, and 0.789 vs. 0.821, respectively). The total accuracies of XGBoost and SVM in the confusion matrices (70.8% and 59.6%) outperformed those of DT and the radiologist (46.1% and 49.2%).

CONCLUSION

This study demonstrated that machine learning models based on morphological MRI radiomics might be an assistive tool for parotid tumor classification, especially for preliminary screening in absence of more advanced scanning sequences, such as DWI.

KEY POINTS

• Machine learning algorithms combined with morphological MRI radiomics could be useful in the preliminary classification of parotid tumors. • XGBoost algorithm performed better than SVM and DT in subtype differentiation of parotid tumors, while DT seemed to have a poor validation performance. • Using morphological MRI only, the XGBoost and SVM algorithms outperformed radiologists in the four-type classification task for parotid tumors, thus making these models a useful assistant diagnostic tool in clinical practice.

Role of Artificial Intelligence in Radiogenomics for Cancers in the Era of Precision Medicine

Simple Summary

Recently, radiogenomics has played a significant role and offered a new understanding of cancer’s biology and behavior in response to standard therapy. It also provides a more precise prognosis, investigation, and analysis of the patient’s cancer. Over the years, Artificial Intelligence (AI) has provided a significant strength in radiogenomics. In this paper, we offer computational and oncological prospects of the role of AI in radiogenomics, as well as its offers, achievements, opportunities, and limitations in the current clinical practices.

Abstract

Radiogenomics, a combination of “Radiomics” and “Genomics,” using Artificial Intelligence (AI) has recently emerged as the state-of-the-art science in precision medicine, especially in oncology care. Radiogenomics syndicates large-scale quantifiable data extracted from radiological medical images enveloped with personalized genomic phenotypes. It fabricates a prediction model through various AI methods to stratify the risk of patients, monitor therapeutic approaches, and assess clinical outcomes. It has recently shown tremendous achievements in prognosis, treatment planning, survival prediction, heterogeneity analysis, reoccurrence, and progression-free survival for human cancer study. Although AI has shown immense performance in oncology care in various clinical aspects, it has several challenges and limitations. The proposed review provides an overview of radiogenomics with the viewpoints on the role of AI in terms of its promises for computational as well as oncological aspects and offers achievements and opportunities in the era of precision medicine. The review also presents various recommendations to diminish these obstacles.

The value of radiomics based on dual-energy CT for differentiating benign from malignant solitary pulmonary nodules

OBJECTIVE

To investigate the value of monochromatic dual-energy CT (DECT) images based on radiomics in differentiating benign from malignant solitary pulmonary nodules.

MATERIALS AND METHODS

This retrospective study was approved by the institutional review board, and informed consent was waived. Pathologically confirmed lung nodules smaller than 3 cm with integrated arterial phase and venous phase (AP and VP) gemstone spectral imaging were retrospectively identified. After extracting the radiomic features of each case, principal component analysis (PCA) was used for feature selection, and after training with the logistic regression method, three classification models (Model_AP, Model_VP and Model_Combination) were constructed. The performance was assessed by the area under the receiver operating curve (AUC), and the efficacy of the models was validated using an independent cohort.

RESULTS

A total of 153 patients were included and divided into a training cohort (n = 107) and a validation cohort (n = 46). A total of 1130 radiomic features were extracted from each case. The PCA method selected 22, 25 and 35 principal components to construct the three models. The diagnostic accuracy of Model_AP, Model_VP and Model_Combination was 0.8043, 0.6739, and 0.7826 in the validation set, with AUCs of 0.8148 (95% CI 0.682-0.948), 0.7485 (95% CI 0.602-0.895), and 0.8772 (95% CI 0.780-0.974), respectively. The DeLong test showed that there were significant differences in the AUCs between Model_AP and Model_Combination (P = 0.0396) and between Model_VP and Model_Combination (P = 0.0465). However, the difference in AUCs between Model_AP and Model_VP was not significant (P = 0.5061). These results demonstrate that Model_Combination shows a better performance than the other models. Decision curve analysis proved the clinical utility of this model.

CONCLUSIONS

We developed a radiomics model based on monochromatic DECT images to identify solitary pulmonary nodules. This model could serve as an effective tool for discriminating benign from malignant pulmonary nodules in patients. The combination of arterial phase and venous phase imaging could significantly improve the model performance.

Radiomics in Early Lung Cancer Diagnosis: From Diagnosis to Clinical Decision Support and Education

Lung cancer is the most frequent cause of cancer-related death around the world. With the recent introduction of low-dose lung computed tomography for lung cancer screening, there has been an increasing number of smoking- and non-smoking-related lung cancer cases worldwide that are manifesting with subsolid nodules, especially in Asian populations. However, the pros and cons of lung cancer screening also follow the implementation of lung cancer screening programs. Here, we review the literature related to radiomics for early lung cancer diagnosis. There are four main radiomics applications: the classification of lung nodules as being malignant/benign; determining the degree of invasiveness of the lung adenocarcinoma; histopathologic subtyping; and prognostication in lung cancer prediction models. In conclusion, radiomics offers great potential to improve diagnosis and personalized risk stratification in early lung cancer diagnosis through patient–doctor cooperation and shared decision making.

CT Reconstruction Kernels and the Effect of Pre- and Post-Processing on the Reproducibility of Handcrafted Radiomic Features

Handcrafted radiomics features (HRFs) are quantitative features extracted from medical images to decode biological information to improve clinical decision making. Despite the potential of the field, limitations have been identified. The most important identified limitation, currently, is the sensitivity of HRF to variations in image acquisition and reconstruction parameters. In this study, we investigated the use of Reconstruction Kernel Normalization (RKN) and ComBat harmonization to improve the reproducibility of HRFs across scans acquired with different reconstruction kernels. A set of phantom scans (n = 28) acquired on five different scanner models was analyzed. HRFs were extracted from the original scans, and scans were harmonized using the RKN method. ComBat harmonization was applied on both sets of HRFs. The reproducibility of HRFs was assessed using the concordance correlation coefficient. The difference in the number of reproducible HRFs in each scenario was assessed using McNemar’s test. The majority of HRFs were found to be sensitive to variations in the reconstruction kernels, and only six HRFs were found to be robust with respect to variations in reconstruction kernels. The use of RKN resulted in a significant increment in the number of reproducible HRFs in 19 out of the 67 investigated scenarios (28.4%), while the ComBat technique resulted in a significant increment in 36 (53.7%) scenarios. The combination of methods resulted in a significant increment in 53 (79.1%) scenarios compared to the HRFs extracted from original images. Since the benefit of applying the harmonization methods depended on the data being harmonized, reproducibility analysis is recommended before performing radiomics analysis. For future radiomics studies incorporating images acquired with similar image acquisition and reconstruction parameters, except for the reconstruction kernels, we recommend the systematic use of the pre- and post-processing approaches (respectively, RKN and ComBat).

Multitask Learning Radiomics on Longitudinal Imaging to Predict Survival Outcomes following Risk-Adaptive Chemoradiation for Non-Small Cell Lung Cancer

Medical imaging provides quantitative and spatial information to evaluate treatment response in the management of patients with non-small cell lung cancer (NSCLC). High throughput extraction of radiomic features on these images can potentially phenotype tumors non-invasively and support risk stratification based on survival outcome prediction. The prognostic value of radiomics from different imaging modalities and time points prior to and during chemoradiation therapy of NSCLC, relative to conventional imaging biomarker or delta radiomics models, remains uncharacterized. We investigated the utility of multitask learning of multi-time point radiomic features, as opposed to single-task learning, for improving survival outcome prediction relative to conventional clinical imaging feature model benchmarks. Survival outcomes were prospectively collected for 45 patients with unresectable NSCLC enrolled on the FLARE-RT phase II trial of risk-adaptive chemoradiation and optional consolidation PD-L1 checkpoint blockade (NCT02773238). FDG-PET, CT, and perfusion SPECT imaging pretreatment and week 3 mid-treatment was performed and 110 IBSI-compliant pyradiomics shape-/intensity-/texture-based features from the metabolic tumor volume were extracted. Outcome modeling consisted of a fused Laplacian sparse group LASSO with component-wise gradient boosting survival regression in a multitask learning framework. Testing performance under stratified 10-fold cross-validation was evaluated for multitask learning radiomics of different imaging modalities and time points. Multitask learning models were benchmarked against conventional clinical imaging and delta radiomics models and evaluated with the concordance index (c-index) and index of prediction accuracy (IPA). FDG-PET radiomics had higher prognostic value for overall survival in test folds (c-index 0.71 [0.67, 0.75]) than CT radiomics (c-index 0.64 [0.60, 0.71]) or perfusion SPECT radiomics (c-index 0.60 [0.57, 0.63]). Multitask learning of pre-/mid-treatment FDG-PET radiomics (c-index 0.71 [0.67, 0.75]) outperformed benchmark clinical imaging (c-index 0.65 [0.59, 0.71]) and FDG-PET delta radiomics (c-index 0.52 [0.48, 0.58]) models. Similarly, the IPA for multitask learning FDG-PET radiomics (30%) was higher than clinical imaging (26%) and delta radiomics (15%) models. Radiomics models performed consistently under different voxel resampling conditions. Multitask learning radiomics for outcome modeling provides a clinical decision support platform that leverages longitudinal imaging information. This framework can reveal the relative importance of different imaging modalities and time points when designing risk-adaptive cancer treatment strategies.

Comparison of Comprehensive Morphological and Radiomics Features of Subsolid Pulmonary Nodules to Distinguish Minimally Invasive Adenocarcinomas and Invasive Adenocarcinomas in CT Scan

Objective

To investigative the diagnostic performance of the morphological model, radiomics model, and combined model in differentiating invasive adenocarcinomas (IACs) from minimally invasive adenocarcinomas (MIAs).

Methods

This study retrospectively involved 307 patients who underwent chest computed tomography (CT) examination and presented as subsolid pulmonary nodules whose pathological findings were MIAs or IACs from January 2010 to May 2018. These patients were randomly assigned to training and validation groups in a ratio of 4:1 for 10 times. Eighteen categories of morphological features of pulmonary nodules including internal and surrounding structure were labeled. The following radiomics features are extracted: first-order features, shape-based features, gray-level co-occurrence matrix (GLCM) features, gray-level size zone matrix (GLSZM) features, gray-level run length matrix (GLRLM) features, and gray-level dependence matrix (GLDM) features. The chi-square test and F1 test selected morphology features, and LASSO selected radiomics features. Logistic regression was used to establish models. Receiver operating characteristic (ROC) curves evaluated the effectiveness, and Delong analysis compared ROC statistic difference among three models.

Results

In validation cohorts, areas under the curve (AUC) of the morphological model, radiomics model, and combined model of distinguishing MIAs from IACs were 0.88, 0.87, and 0.89; the sensitivity (SE) was 0.68, 0.81, and 0.83; and the specificity (SP) was 0.93, 0.79, and 0.87. There was no statistically significant difference in AUC between three models (p > 0.05).

Conclusion

The morphological model, radiomics model, and combined model all have a high efficiency in the differentiation between MIAs and IACs and have potential to provide non-invasive assistant information for clinical decision-making.

Effect of CT image acquisition parameters on diagnostic performance of radiomics in predicting malignancy of pulmonary nodules of different sizes

OBJECTIVES

To investigate the effect of CT image acquisition parameters on the performance of radiomics in classifying benign and malignant pulmonary nodules (PNs) with respect to nodule size.

METHODS

We retrospectively collected CT images of 696 patients with PNs from March 2015 to March 2018. PNs were grouped by nodule diameter: T1a (diameter ≤ 1.0 cm), T1b (1.0 cm < diameter ≤ 2.0 cm), and T1c (2.0 cm < diameter ≤ 3.0 cm). CT images were divided into four settings according to slice-thickness-convolution-kernels: setting 1 (slice thickness/reconstruction type: 1.25 mm sharp), setting 2 (5 mm sharp), setting 3 (5 mm smooth), and random setting. We created twelve groups from two interacting conditions. Each PN was segmented and had 1160 radiomics features extracted. Non-redundant features with high predictive ability in training were selected to build a distinct model under each of the twelve subsets.

RESULTS

The performance (AUCs) on predicting PN malignancy were as follows: T1a group: 0.84, 0.64, 0.68, and 0.68; T1b group: 0.68, 0.74, 0.76, and 0.70; T1c group: 0.66, 0.64, 0.63, and 0.70, for the setting 1, setting 2, setting 3, and random setting, respectively. In the T1a group, the AUC of radiomics model in setting 1 was statistically significantly higher than all others; In the T1b group, AUCs of radiomics models in setting 3 were statistically significantly higher than some; and in the T1c group, there were no statistically significant differences among models.

CONCLUSIONS

For PNs less than 1 cm, CT image acquisition parameters have a significant influence on diagnostic performance of radiomics in predicting malignancy, and a model created using images reconstructed with thin section and a sharp kernel algorithm achieved the best performance. For PNs larger than 1 cm, CT reconstruction parameters did not affect diagnostic performance substantially.

KEY POINTS

• CT image acquisition parameters have a significant influence on the diagnostic performance of radiomics in pulmonary nodules less than 1 cm. • In pulmonary nodules less than 1 cm, a radiomics model created by using images reconstructed with thin section and a sharp kernel algorithm achieved the best diagnostic performance. • For PNs larger than 1 cm, CT image acquisition parameters do not affect diagnostic performance substantially.

Non-small cell lung carcinoma histopathological subtype phenotyping using high-dimensional multinomial multiclass CT radiomics signature

OBJECTIVE

The aim of this study was to identify the most important features and assess their discriminative power in the classification of the subtypes of NSCLC.

METHODS

This study involved 354 pathologically proven NSCLC patients including 134 squamous cell carcinoma (SCC), 110 large cell carcinoma (LCC), 62 not other specified (NOS), and 48 adenocarcinoma (ADC). In total, 1433 radiomics features were extracted from 3D volumes of interest drawn on the malignant lesion identified on CT images. Wrapper algorithm and multivariate adaptive regression splines were implemented to identify the most relevant/discriminative features. A multivariable multinomial logistic regression was employed with 1000 bootstrapping samples based on the selected features to classify four main subtypes of NSCLC.

RESULTS

The results revealed that the texture features, specifically gray level size zone matrix features (GLSZM), were the significant indicators of NSCLC subtypes. The optimized classifier achieved an average precision, recall, F1-score, and accuracy of 0.710, 0.703, 0.706, and 0.865, respectively, based on the selected features by the wrapper algorithm.

CONCLUSIONS

Our CT radiomics approach demonstrated impressive potential for the classification of the four main histological subtypes of NSCLC, It is anticipated that CT radiomics could be useful in treatment planning and precision medicine.

Radiomics Prediction of EGFR Status in Lung Cancer-Our Experience in Using Multiple Feature Extractors and The Cancer Imaging Archive Data

We investigated the performance of multiple radiomics feature extractors/software on predicting epidermal growth factor receptor mutation status in 228 patients with non–small cell lung cancer from publicly available data sets in The Cancer Imaging Archive. The imaging and clinical data were split into training (n = 105) and validation cohorts (n = 123). Two of the most cited open-source feature extractors, IBEX (1563 features) and Pyradiomics (1319 features), and our in-house software, Columbia Image Feature Extractor (CIFE) (1160 features), were used to extract radiomics features. Univariate and multivariate analyses were performed sequentially to predict EGFR mutation status using each individual feature extractor. Our univariate analysis integrated an unsupervised clustering method to identify nonredundant and informative candidate features for the creation of prediction models by multivariate analyses. In training, unsupervised clustering-based univariate analysis identified 5, 6, and 4 features from IBEX, Pyradiomics, and CIFE as candidate features, respectively. Multivariate prediction models using these features from IBEX, Pyradiomics, and CIFE yielded similar areas under the receiver operating characteristic curve of 0.68, 0.67, and 0.69. However, in validation, areas under the receiver operating characteristic curve of multivariate prediction models from IBEX, Pyradiomics, and CIFE decreased to 0.54, 0.56 and 0.64, respectively. Different feature extractors select different radiomics features, which leads to prediction models with varying performance. However, correlation between those selected features from different extractors may indicate these features measure similar imaging phenotypes associated with similar biological characteristics. Overall, attention should be paid to the generalizability of individual radiomics features and radiomics prediction models.

Machine Learning Radiomics Model for Early Identification of Small-Cell Lung Cancer on Computed Tomography Scans

PURPOSE

Small-cell lung cancer (SCLC) is the deadliest form of lung cancer, partly because of its short doubling time. Delays in imaging identification and diagnosis of nodules create a risk for stage migration. The purpose of our study was to determine if a machine learning radiomics model can detect SCLC on computed tomography (CT) among all nodules at least 1 cm in size.

MATERIALS AND METHODS

Computed tomography scans from a single institution were selected and resampled to 1 × 1 × 1 mm. Studies were divided into SCLC and other scans comprising benign, adenocarcinoma, and squamous cell carcinoma that were segregated into group A (noncontrast scans) and group B (contrast-enhanced scans). Four machine learning classification models, support vector classifier, random forest (RF), XGBoost, and logistic regression, were used to generate radiomic models using 59 quantitative first-order and texture Imaging Biomarker Standardization Initiative compliant PyRadiomics features, which were found to be robust between two segmenters with minimum Redundancy Maximum Relevance feature selection within each leave-one-out-cross-validation to avoid overfitting. The performance was evaluated using a receiver operating characteristic curve. A final model was created using the RF classifier and aggregate minimum Redundancy Maximum Relevance to determine feature importance.

RESULTS

A total of 103 studies were included in the analysis. The area under the receiver operating characteristic curve for RF, support vector classifier, XGBoost, and logistic regression was 0.81, 0.77, 0.84, and 0.84 in group A, and 0.88, 0.87, 0.85, and 0.81 in group B, respectively. Nine radiomic features in group A and 14 radiomic features in group B were predictive of SCLC. Six radiomic features overlapped between groups A and B.

CONCLUSION

A machine learning radiomics model may help differentiate SCLC from other lung lesions.

Improving the Subtype Classification of Non-small Cell Lung Cancer by Elastic Deformation Based Machine Learning

Non-invasive image-based machine learning models have been used to classify subtypes of non-small cell lung cancer (NSCLC). However, the classification performance is limited by the dataset size, because insufficient data cannot fully represent the characteristics of the tumor lesions. In this work, a data augmentation method named elastic deformation is proposed to artificially enlarge the image dataset of NSCLC patients with two subtypes (squamous cell carcinoma and large cell carcinoma) of 3158 images. Elastic deformation effectively expanded the dataset by generating new images, in which tumor lesions go through elastic shape transformation. To evaluate the proposed method, two classification models were trained on the original and augmented dataset, respectively. Using augmented dataset for training significantly increased classification metrics including area under the curve (AUC) values of receiver operating characteristics (ROC) curves, accuracy, sensitivity, specificity, and f1-score, thus improved the NSCLC subtype classification performance. These results suggest that elastic deformation could be an effective data augmentation method for NSCLC tumor lesion images, and building classification models with the help of elastic deformation has the potential to serve for clinical lung cancer diagnosis and treatment design.

Radiomics in Oncology, Part 2: Thoracic, Genito-Urinary, Breast, Neurological, Hematologic and Musculoskeletal Applications

Simple Summary

This Part II is an overview of the main applications of Radiomics in oncologic imaging with a focus on diagnosis, prognosis prediction and assessment of response to therapy in thoracic, genito-urinary, breast, neurologic, hematologic and musculoskeletal oncology. In this part II we describe the radiomic applications, limitations and future perspectives for each pre-eminent tumor. In the future, Radiomics could have a pivotal role in management of cancer patients as an imaging tool to support clinicians in decision making process. However, further investigations need to obtain some stable results and to standardize radiomic analysis (i.e., image acquisitions, segmentation and model building) in clinical routine.

Abstract

Radiomics has the potential to play a pivotal role in oncological translational imaging, particularly in cancer detection, prognosis prediction and response to therapy evaluation. To date, several studies established Radiomics as a useful tool in oncologic imaging, able to support clinicians in practicing evidence-based medicine, uniquely tailored to each patient and tumor. Mineable data, extracted from medical images could be combined with clinical and survival parameters to develop models useful for the clinicians in cancer patients’ assessment. As such, adding Radiomics to traditional subjective imaging may provide a quantitative and extensive cancer evaluation reflecting histologic architecture. In this Part II, we present an overview of radiomic applications in thoracic, genito-urinary, breast, neurological, hematologic and musculoskeletal oncologic applications.

Identifying Robust Radiomics Features for Lung Cancer by Using In-Vivo and Phantom Lung Lesions

We propose a novel framework for determining radiomics feature robustness by considering the effects of both biological and noise signals. This framework is preliminarily tested in a study predicting the epidermal growth factor receptor (EGFR) mutation status in non-small cell lung cancer (NSCLC) patients. Pairs of CT images (baseline, 3-week post therapy) of 46 NSCLC patients with known EGFR mutation status were collected and a FDA-customized anthropomorphic thoracic phantom was scanned on two vendors' scanners at four different tube currents. Delta radiomics features were extracted from the NSCLC patient CTs and reproducible, non-redundant, and informative features were identified. The feature value differences between EGFR mutant and EGFR wildtype patients were quantitatively measured as the biological signal. Similarly, radiomics features were extracted from the phantom CTs. A pairwise comparison between settings resulted in a feature value difference that was quantitatively measured as the noise signal. Biological signals were compared to noise signals at each setting to determine if the distributions were significantly different by two-sample t-test, and thus robust. Four optimal features were selected to predict EGFR mutation status, Tumor-Mass, Sigmoid-Offset-Mean, Gabor-Energy and DWT-Energy, which quantified tumor mass, tumor-parenchyma density transition at boundary, line-like pattern inside tumor and intratumoral heterogeneity, respectively. The first three variables showed robustness across the majority of studied CT acquisition parameters. The textual feature DWT-Energy was less robust. The proposed framework was able to determine robustness of radiomics features at specific settings by comparing biological signal to noise signal. Identification of robust radiomics features may improve the generalizability of radiomics models in future studies.

Role of SOX Protein Groups F and H in Lung Cancer Progression

Simple Summary

The expression of SOX proteins has been demonstrated in many tissues at various stages of embryogenesis, where they play the role of transcription factors. The SOX18 protein (along with SOX7 and SOX17) belongs to the SOXF group and is mainly involved in the development of the cardiovascular system, where its expression was found in the endothelium. SOX18 expression was also demonstrated in neoplastic lines of gastric, pancreatic and colon adenocarcinomas. The prognostic role of SOX30 expression has only been studied in lung adenocarcinomas, where a low expression of this factor in the stromal tumor was associated with a worse prognosis for patients. Because of the complexity of non-small-cell lung cancer (NSCLC) development, the role of the SOX proteins in this malignancy is still not fully understood. Many recently published papers show that SOX family protein members play a crucial role in the progression of NSCLC.

Abstract

The SOX family proteins are proved to play a crucial role in the development of the lymphatic ducts and the cardiovascular system. Moreover, an increased expression level of the SOX18 protein has been found in many malignances, such as melanoma, stomach, pancreatic breast and lung cancers. Another SOX family protein, the SOX30 transcription factor, is responsible for the development of male germ cells. Additionally, recent studies have shown its proapoptotic character in non-small cell lung cancer cells. Our preliminary studies showed a disparity in the amount of mRNA of the SOX18 gene relative to the amount of protein. This is why our attention has been focused on microRNA (miRNA) molecules, which could regulate the SOX18 gene transcript level. Recent data point to the fact that, in practically all types of cancer, hundreds of genes exhibit an abnormal methylation, covering around 5–10% of the thousands of CpG islands present in the promoter sequences, which in normal cells should not be methylated from the moment the embryo finishes its development. It has been demonstrated that in non-small-cell lung cancer (NSCLC) cases there is a large heterogeneity of the methylation process. The role of the SOX18 and SOX30 expression in non-small-cell lung cancers (NSCLCs) is not yet fully understood. However, if we take into account previous reports, these proteins may be important factors in the development and progression of these malignancies.

Exploratory Study of a CT Radiomics Model for the Classification of Small Cell Lung Cancer and Non-small-Cell Lung Cancer

Background: Radiomics can quantify tumor phenotypic characteristics non-invasively by applying feature algorithms to medical imaging data. In this study, we investigated the association between radiomics features and the tumor histological subtypes, and we aimed to establish a nomogram for the classification of small cell lung cancer (SCLC) and non-small-cell lung cancer (NSCLC). Methods: This was a retrospective single center study. In total, 468 cases including 202 patients with SCLC and 266 patients with NSCLC were enrolled in our study, and were randomly divided into a training set (n = 327) and a validation set (n = 141) in a 7:3 ratio. The clinical data of the patients, including age, sex, smoking history, tumor maximum diameter, clinical stage, and serum tumor markers, were collected. All patients underwent enhanced computed tomography (CT) scans, and all lesions were pathologically confirmed. A radiomics signature was generated from the training set using the least absolute shrinkage and selection operator algorithm. Independent risk factors were identified by multivariate logistic regression analysis, and a radiomics nomogram based on the radiomics signature and clinical features was constructed. The capability of the nomogram was evaluated in the training set and validated in the validation set. Results: Fourteen of 396 radiomics parameters were screened as important factors for establishing the radiomics model. The radiomics signature performed well in differentiating SCLC and NSCLC, with an area under the curve (AUC) of 0.86 (95% CI: 0.82-0.90) in the training set and 0.82 (95% CI: 0.75-0.89) in the validation set. The radiomics nomogram had better predictive performance [AUC = 0.94 (95% CI: 0.90-0.98) in the validation set] than the clinical model [AUC = 0.86 (95% CI: 0.80-0.93)] and the radiomics signature [AUC = 0.82 (95% CI: 0.75-0.89)], and the accuracy was 86.2% (95% CI: 0.79-0.92) in the validation set. Conclusion: The enhanced CT radiomics signature performed well in the classification of SCLC and NSCLC. The nomogram based on the radiomics signature and clinical factors has better diagnostic performance for the classification of SCLC and NSCLC than the simple application of the radiomics signature.

The application of artificial intelligence and radiomics in lung cancer

Lung cancer is one of the most leading causes of death throughout the world, and there is an urgent requirement for the precision medical management of it. Artificial intelligence (AI) consisting of numerous advanced techniques has been widely applied in the field of medical care. Meanwhile, radiomics based on traditional machine learning also does a great job in mining information through medical images. With the integration of AI and radiomics, great progress has been made in the early diagnosis, specific characterization, and prognosis of lung cancer, which has aroused attention all over the world. In this study, we give a brief review of the current application of AI and radiomics for precision medical management in lung cancer.

Differentiation of Focal-Type Autoimmune Pancreatitis From Pancreatic Ductal Adenocarcinoma Using Radiomics Based on Multiphasic Computed Tomography

OBJECTIVES

The aim of this study was to develop a radiomics model for a differential diagnosis of focal-type autoimmune pancreatitis (AIP) from pancreatic ductal adenocarcinoma.

METHODS

A total of 96 patients, 45 with AIP and 51 with pancreatic ductal adenocarcinoma, were retrospectively evaluated. All patients underwent pretreatment abdominal computed tomography imaging acquired at noncontrast, arterial, and venous phases. Furthermore, 1160 radiomics features were extracted from each phasic image to build radiomics models. The performance of radiomics model was evaluated by sensitivity, specificity, and accuracy. The results of radiomics model were also compared with those of radiologists' visual assessments.

RESULTS

The sensitivity, specificity, and accuracy of the optimal radiomics model were 93.3%, 96.1%, and 94.8%, respectively. They were higher than those of the radiologists' assessments with sensitivity of 57.78% and 73.33%, specificity of 88.24% and 90.20%, and accuracy of 75.00% and 81.25%, respectively.

CONCLUSION

Radiomics is helpful for a differential diagnosis of AIP in clinical practice as a noninvasive and quantitative method.

Preoperative prediction for pathological grade of hepatocellular carcinoma via machine learning-based radiomics

OBJECTIVE

To investigate the efficacy of contrast-enhanced computed tomography (CECT)-based radiomics signatures for preoperative prediction of pathological grades of hepatocellular carcinoma (HCC) via machine learning.

METHODS

In this single-center retrospective study, data collected from 297 consecutive subjects with HCC were allocated to training dataset (n = 237) and test dataset (n = 60). Manual segmentation of lesion sites was performed with ITK-SNAP, the radiomics features were extracted by the Pyradiomics, and radiomics signatures were synthesized using recursive feature elimination (RFE) method. The prediction models for pathological grading of HCC were established by using eXtreme Gradient Boosting (XGBoost). The performance of the models was evaluated using the AUC along with 95% confidence intervals (CIs) and standard deviation, sensitivity, specificity, and accuracy.

RESULTS

The radiomics signatures were found highly efficient for machine learning to differentiate high-grade HCC from low-grade HCC. For the clinical factors, when they were merely applied to train a machine learning model, the model achieved an AUC of 0.6698, along with 95% CI and standard deviation of 0.5307-0.8089 and 0.0710, respectively (sensitivity, 0.6522; specificity, 0.4595; accuracy, 0.5333). Meanwhile, when the radiomics signatures were applied in association with clinical factors to train a machine learning model, the performance of the model remarkably increased with AUC of 0.8014, along with 95% CI and standard deviation of 0.6899-0.9129 and 0.0569, respectively (sensitivity, 0.6522; specificity, 0.7297; accuracy, 0.7000).

CONCLUSIONS

The radiomics signatures could non-invasively explore the underlying association between CECT images and pathological grades of HCC.

KEY POINTS

• The radiomics signatures may non-invasively explore the underlying association between CECT images and pathological grades of HCC via machine learning. • The radiomics signatures of CECT images may enhance the prediction performance of pathological grading of HCC, and further validation is required. • The features extracted from arterial phase CECT images may be more reliable than venous phase CECT images for predicting pathological grades of HCC.

Differentiating Peripherally-Located Small Cell Lung Cancer From Non-small Cell Lung Cancer Using a CT Radiomic Approach

Lung cancer can be classified into two main categories: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC), which are different in treatment strategy and survival probability. The lung CT images of SCLC and NSCLC are similar such that their subtle differences are hardly visually discernible by the human eye through conventional imaging evaluation. We hypothesize that SCLC/NSCLC differentiation could be achieved via computerized image feature analysis and classification in feature space, as termed a radiomic model. The purpose of this study was to use CT radiomics to differentiate SCLC from NSCLC adenocarcinoma. Patients with primary lung cancer, either SCLC or NSCLC adenocarcinoma, were retrospectively identified. The post-diagnosis pre-treatment lung CT images were used to segment the lung cancers. Radiomic features were extracted from histogram-based statistics, textural analysis of tumor images and their wavelet transforms. A minimal-redundancy-maximal-relevance method was used for feature selection. The predictive model was constructed with a multilayer artificial neural network. The performance of the SCLC/NSCLC adenocarcinoma classifier was evaluated by the area under the receiver operating characteristic curve (AUC). Our study cohort consisted of 69 primary lung cancer patients with SCLC (n = 35; age mean ± SD = 66.91± 9.75 years), and NSCLC adenocarcinoma (n = 34; age mean ± SD = 58.55 ± 11.94 years). The SCLC group had more male patients and smokers than the NSCLC group (P < 0.05). Our SCLC/NSCLC classifier achieved an overall performance of AUC of 0.93 (95% confidence interval = [0.85, 0.97]), sensitivity = 0.85, and specificity = 0.85). Adding clinical data such as smoking history could improve the performance slightly. The top ranking radiomic features were mostly textural features. Our results showed that CT radiomics could quantitatively represent tumor heterogeneity and therefore could be used to differentiate primary lung cancer subtypes with satisfying results. CT image processing with the wavelet transformation technique enhanced the radiomic features for SCLC/NSCLC classification. Our pilot study should motivate further investigation of radiomics as a non-invasive approach for early diagnosis and treatment of lung cancer.

Siamese neural networks for the classification of high-dimensional radiomic features

This study demonstrates that a variant of a Siamese neural network architecture is more effective at classifying high-dimensional radiomic features (extracted from T2 MRI images) than traditional models, such as a Support Vector Machine or Discriminant Analysis. Ninety-nine female patients, between the ages of 20 and 48, were imaged with T2 MRI. Using biopsy pathology, the patients were separated into two groups: those with breast cancer (N=55) and those with GLM (N=44). Lesions were segmented by a trained radiologist and the ROIs were used for radiomic feature extraction. The radiomic features include 536 published features from Aerts et al., along with 20 features recurrent quantification analysis features. A Student T-Test was used to select features found to be statistically significant between the two patient groups. These features were then used to train a Siamese neural network. The label given to test features was the label of whichever class the test features with the highest percentile similarity within the training group. Within the two highest-dimensional feature sets, the Siamese network produced an AUC of 0.853 and 0.894, respectively. This is compared to best non-Siamese model, Discriminant Analysis, which produced an AUC of 0.823 and 0.836 for the two respective feature sets. However, when it came to the lower-dimensional recurrent features and the top-20 most significant features from Aerts et al., the Siamese network performed on-par or worse than the competing models. The proposed Siamese neural network architecture can outperform competing other models in high-dimensional, low-sample size spaces with regards to tabular data.

A quantitative imaging biomarker for predicting disease-free-survival-associated histologic subgroups in lung adenocarcinoma

OBJECTIVES

Classification of histologic subgroups has significant prognostic value for lung adenocarcinoma patients who undergo surgical resection. However, clinical histopathology assessment is generally performed on only a small portion of the overall tumor from biopsy or surgery. Our objective is to identify a noninvasive quantitative imaging biomarker (QIB) for the classification of histologic subgroups in lung adenocarcinoma patients.

METHODS

We retrospectively collected and reviewed 1313 CT scans of patients with resected lung adenocarcinomas from two geographically distant institutions who were seen between January 2014 and October 2017. Three study cohorts, the training, internal validation, and external validation cohorts, were created, within which lung adenocarcinomas were divided into two disease-free-survival (DFS)-associated histologic subgroups, the mid/poor and good DFS groups. A comprehensive machine learning- and deep learning-based analytical system was adopted to identify reproducible QIBs and help to understand QIBs' significance.

RESULTS

Intensity-Skewness, a QIB quantifying tumor density distribution, was identified as the optimal biomarker for predicting histologic subgroups. Intensity-Skewness achieved high AUCs (95% CI) of 0.849(0.813,0.881), 0.820(0.781,0.856) and 0.863(0.827,0.895) on the training, internal validation, and external validation cohorts, respectively. A criterion of Intensity-Skewness ≤ 1.5, which indicated high tumor density, showed high specificity of 96% (sensitivity 46%) and 99% (sensitivity 53%) on predicting the mid/poor DFS group in the training and external validation cohorts, respectively.

CONCLUSIONS

A QIB derived from routinely acquired CT was able to predict lung adenocarcinoma histologic subgroups, providing a noninvasive method that could potentially benefit personalized treatment decision-making for lung cancer patients.

KEY POINTS

• A noninvasive imaging biomarker, Intensity-Skewness, which described the distortion of pixel-intensity distribution within lesions on CT images, was identified as a biomarker to predict disease-free-survival-associated histologic subgroups in lung adenocarcinoma. • An Intensity-Skewness of ≤ 1.5 has high specificity in predicting the mid/poor disease-free survival histologic patient group in both the training cohort and the external validation cohort. • The Intensity-Skewness is a feature that can be automatically computed with high reproducibility and robustness.

Tumor volume and sphericity as predictors of local control after stereotactic radiosurgery for limited number (1-4) brain metastases from nonsmall cell lung cancer

AIM

This study aims to evaluate the usage of brain metastases (BM) tumor volume and sphericity as prognostic factors in local control (LC) after stereotactic radiosurgery (SRS) for limited number (1-4) BM from nonsmall cell lung cancer (NSCLC).

METHODS

We retrospectively reviewed 80 patients, with 141 BM, who were treated with SRS from 2012 to 2017. Local failure was defined as an increase in lesion size after SRS. LC and overall survival (OS) were estimated using Kaplan-Meier method. The Cox proportional hazards model was used for univariate and multivariate analysis.

RESULTS

The median clinical and radiographic follow-up was 11.2 and 9.0 months, respectively. The median BM tumor volume was 0.31 cm³ (0.01-21.64 cm³ ) and the median tumor sphericity was 0.76 (0.39-0.95). The median LC of the entire cohort was 28.8 months. LC rate at last follow-up was achieved in 84.4% of patients (35.5% CR, 35.5% PR, and 13.5% SD). LC was 83.8% at 1 year and 56.3% at 2 years. On multivariate analysis, only sphericity (P < .001) and volume (P = .004) were found to be a strong predictor for LC. The median OS of the entire cohort was 24.1 months. On multivariate analysis, only GPA score was found to be a predictor for OS.

CONCLUSION

BM tumor sphericity and volume were found to be strong predictors for LC. Tumor sphericity and volume should be taken into consideration when treating patients with BM and when designing future prospective studies and developing prognostic indices.

Independent prognostic value of HIF-1α expression in radiofrequency ablation of lung cancer

Radiofrequency ablation (RFA) is widely used in the treatment of lung cancer. Hypoxia-inducible factor-1α (HIF-1α) is a crucial transcription factor regulating oxygen homeostasis that is involved in tumor cell metastasis. The present study investigated the impact of HIF-1α expression and other factors, such as postoperative blood CD4⁺/CD8⁺ ratio, on the prognosis of patients with lung cancer who had received RFA treatment. A total of 80 patients with lung cancer were recruited between January 2011 and October 2016 at The Shenzhen People's Hospital. Lung cancer was confirmed following pathological or histological examination. All patients underwent RFA treatment. Patients were followed up for 6–66 months. HIF-1α expression in lung cancer tissues was assessed by immunohistochemistry. Multivariate survival analysis was performed using Cox proportional hazards model. The results demonstrated that HIF-1α level was low in 36 patients and overexpressed in 44 patients with lung cancer. Kaplan-Meier (KM) curve analysis demonstrated that the overall survival time of patients with high HIF-1α expression was significantly shorter compared with patients with low HIF-1α expression (P<0.05). Furthermore, the results from the KM model and log-rank test revealed that age, Union for International Cancer Control stage, primary or metastatic cancer, chemotherapy, postoperative blood CD4⁺/CD8⁺ ratio, Eastern Cooperative Oncology Group performance status and HIF-1α expression had significant effects on overall survival of patients with lung cancer. The results from Cox analysis demonstrated that high HIF-1α expression, advanced age, clinical staging and chemotherapy were independent risk factors for the prognosis of lung cancer following RFA treatment, and that high HIF-1α expression was associated with the increased risk (5.91-fold) of mortality. In conclusion, the present study demonstrated that HIF-1α expression was increased in lung cancer tissues and was associated with the prognosis of patients with lung cancer who were treated with RFA. These findings suggest that HIF-1α expression may be considered as a marker for evaluating the prognosis of these patients.