Impact of feature selection methods and subgroup factors on prognostic analysis with CT-based radiomics in non-small cell lung cancer patients

A distributed feature selection pipeline for survival analysis using radiomics in non-small cell lung cancer patients

Predictive modelling of cancer outcomes using radiomics faces dimensionality problems and data limitations, as radiomics features often number in the hundreds, and multi-institutional data sharing is ()often unfeasible. Federated learning (FL) and feature selection (FS) techniques combined can help overcome these issues, as one provides the means of training models without exchanging sensitive data, while the other identifies the most informative features, reduces overfitting, and improves model interpretability. Our proposed FS pipeline based on FL principles targets data-driven radiomics FS in a multivariate survival study of non-small cell lung cancer patients. The pipeline was run across datasets from three institutions without patient-level data exchange. It includes two FS techniques, Correlation-based Feature Selection and LASSO regularization, and Cox Proportional-Hazard regression with Overall Survival as endpoint. Trained and validated on 828 patients overall, our pipeline yielded a radiomic signature comprising "intensity-based energy" and "mean discretised intensity". Validation resulted in a mean Harrell C-index of 0.59, showcasing fair efficacy in risk stratification. In conclusion, we suggest a distributed radiomics approach that incorporates preliminary feature selection to systematically decrease the feature set based on data-driven considerations. This aims to address dimensionality challenges beyond those associated with data constraints and interpretability concerns.

Early acquired resistance to EGFR-TKIs in lung adenocarcinomas before radiographic advanced identified by CT radiomic delta model based on two central studies

Early acquired resistance (EAR) to epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) in lung adenocarcinomas before radiographic advance cannot be perceived by the naked eye. This study aimed to discover and validate a CT radiomic model to precisely identify the EAR. Training cohort (n = 67) and internal test cohort (n = 29) were from the First Affiliated Hospital of Fujian Medical University, and external test cohort (n = 29) was from the Second Affiliated Hospital of Xiamen Medical College. Follow-up CT images at three different times of each patient were collected: (1) baseline images before EGFR-TKIs therapy; (2) first follow-up images after EGFR-TKIs therapy (FFT); (3) EAR images, which were the last follow-up images before radiographic advance. The features extracted from FFT and EAR were used to construct the classic radiomic model. The delta features which were calculated by subtracting the baseline from either FFT or EAR were used to construct the delta radiomic model. The classic radiomic model achieved AUC 0.682 and 0.641 in training and internal test cohorts, respectively. The delta radiomic model achieved AUC 0.730 and 0.704 in training and internal test cohorts, respectively. Over the external test cohort, the delta radiomic model achieved AUC 0.661. The decision curve analysis showed that when threshold of the probability of the EAR to the EGFR-TKIs was between 0.3 and 0.82, the proposed model was more benefit than treating all patients. Based on two central studies, the delta radiomic model derived from the follow-up non-enhanced CT images can help clinicians to identify the EAR to EGFR-TKIs in lung adenocarcinomas before radiographic advance and optimize clinical outcomes.

Interpretability of radiomics models is improved when using feature group selection strategies for predicting molecular and clinical targets in clear-cell renal cell carcinoma: insights from the TRACERx Renal study

BACKGROUND

The aim of this work is to evaluate the performance of radiomics predictions for a range of molecular, genomic and clinical targets in patients with clear cell renal cell carcinoma (ccRCC) and demonstrate the impact of novel feature selection strategies and sub-segmentations on model interpretability.

METHODS

Contrast-enhanced CT scans from the first 101 patients recruited to the TRACERx Renal Cancer study (NCT03226886) were used to derive radiomics classification models to predict 20 molecular, histopathology and clinical target variables. Manual 3D segmentation was used in conjunction with automatic sub-segmentation to generate radiomics features from the core, rim, high and low enhancing sub-regions, and the whole tumour. Comparisons were made between two classification model pipelines: a Conventional pipeline reflecting common radiomics practice, and a Proposed pipeline including two novel feature selection steps designed to improve model interpretability. For both pipelines nested cross-validation was used to estimate prediction performance and tune model hyper-parameters, and permutation testing was used to evaluate the statistical significance of the estimated performance measures. Further model robustness assessments were conducted by evaluating model variability across the cross-validation folds.

RESULTS

Classification performance was significant (p < 0.05, H0:AUROC = 0.5) for 11 of 20 targets using either pipeline and for these targets the AUROCs were within ± 0.05 for the two pipelines, except for one target where the Proposed pipeline performance increased by > 0.1. Five of these targets (necrosis on histology, presence of renal vein invasion, overall histological stage, linear evolutionary subtype and loss of 9p21.3 somatic alteration marker) had AUROC > 0.8. Models derived using the Proposed pipeline contained fewer feature groups than the Conventional pipeline, leading to more straightforward model interpretations without loss of performance. Sub-segmentations lead to improved performance and/or improved interpretability when predicting the presence of sarcomatoid differentiation and tumour stage.

CONCLUSIONS

Use of the Proposed pipeline, which includes the novel feature selection methods, leads to more interpretable models without compromising prediction performance.

TRIAL REGISTRATION

NCT03226886 (TRACERx Renal).

Identification of radiomic biomarkers in a set of four skeletal muscle groups on Dixon MRI of the NAKO MR study

In this work, we propose a processing pipeline for the extraction and identification of meaningful radiomics biomarkers in skeletal muscle tissue as displayed using Dixon-weighted MRI. Diverse and robust radiomics features can be identified that may be of aid in the accurate quantification e.g. varying degrees of sarcopenia in respective muscles of large cohorts. As such, the approach comprises the texture feature extraction from raw data based on well established approaches, such as a nnU-Net neural network and the Pyradiomics toolbox, a subsequent selection according to adequate conditions for the muscle tissue of the general population, and an importance-based ranking to further narrow the amount of meaningful features with respect to auxiliary targets. The performance was investigated with respect to the included auxiliary targets, namely age, body mass index (BMI), and fat fraction (FF). Four skeletal muscles with different fiber architecture were included: the mm. glutaei, m. psoas, as well as the extensors and adductors of the thigh. The selection allowed for a reduction from 1015 available texture features to 65 for age, 53 for BMI, and 36 for FF from the available fat/water contrast images considering all muscles jointly. Further, the dependence of the importance rankings calculated for the auxiliary targets on validation sets (in a cross-validation scheme) was investigated by boxplots. In addition, significant differences between subgroups of respective auxiliary targets as well as between both sexes were shown to be present within the ten lowest ranked features by means of Kruskal-Wallis H-tests and Mann-Whitney U-tests. The prediction performance for the selected features and the ranking scheme were verified on validation sets by a random forest based multi-class classification, with strong area under the curve (AUC) values of the receiver operator characteristic (ROC) of 73.03 ± 0.70 % and 73.63 ± 0.70 % for the water and fat images in age, 80.68 ± 0.30 % and 88.03 ± 0.89 % in BMI, as well as 98.36 ± 0.03 % and 98.52 ± 0.09 % in FF.

Development and Validation of CT-Based Radiomics Signature for Overall Survival Prediction in Multi-organ Cancer

The malignant tumors in nature share some common morphological characteristics. Radiomics is not only images but also data; we think that a probability exists in a set of radiomics signatures extracted from CT scan images of one cancer tumor in one specific organ also be utilized for overall survival prediction in different types of cancers in different organs. The retrospective study enrolled four data sets of cancer patients in three different organs (420, 157, 137, and 191 patients for lung 1 training, lung 2 testing, and two external validation set: kidney and head and neck, respectively). In the training set, radiomics features were obtained from CT scan images, and essential features were chosen by LASSO algorithm. Univariable and multivariable analyses were then conducted to find a radiomics signature via Cox proportional hazard regression. The Kaplan-Meier curve was performed based on the risk score. The integrated time-dependent area under the ROC curve (iAUC) was calculated for each predictive model. In the training set, Kaplan-Meier curve classified patients as high or low-risk groups (p-value < 0.001; log-rank test). The risk score of radiomics signature was locked and independently evaluated in the testing set, and two external validation sets showed significant differences (p-value < 0.05; log-rank test). A combined model (radiomics + clinical) showed improved iAUC in lung 1, lung 2, head and neck, and kidney data set are 0.621 (95% CI 0.588, 0.654), 0.736 (95% CI 0.654, 0.819), 0.732 (95% CI 0.655, 0.809), and 0.834 (95% CI 0.722, 0.946), respectively. We believe that CT-based radiomics signatures for predicting overall survival in various cancer sites may exist.

Feature selection methods and predictive models in CT lung cancer radiomics

Radiomics is a technique that extracts quantitative features from medical images using data-characterization algorithms. Radiomic features can be used to identify tissue characteristics and radiologic phenotyping that is not observable by clinicians. A typical workflow for a radiomics study includes cohort selection, radiomic feature extraction, feature and predictive model selection, and model training and validation. While there has been increasing attention given to radiomic feature extraction, standardization, and reproducibility, currently, there is a lack of rigorous evaluation of feature selection methods and predictive models. Herein, we review the published radiomics investigations in CT lung cancer and provide an overview of the commonly used radiomic feature selection methods and predictive models. We also compare limitations of various methods in clinical applications and present sources of uncertainty associated with those methods. This review is expected to help raise awareness of the impact of radiomic feature and model selection methods on the integrity of radiomics studies.

Risk analysis of grade ≥ 2 radiation pneumonitis based on radiotherapy timeline in stage III/IV non-small cell lung cancer treated with volumetric modulated arc therapy: a retrospective study

Background

Radiotherapy is an important treatment for patients with stage III/IV non-small cell lung cancer (NSCLC), and due to its high incidence of radiation pneumonitis, it is essential to identify high-risk people as early as possible. The present work investigates the value of the application of different phase data throughout the radiotherapy process in analyzing risk of grade ≥ 2 radiation pneumonitis in stage III/IV NSCLC. Furthermore, the phase data fusion was gradually performed with the radiotherapy timeline to develop a risk assessment model.

Methods

This study retrospectively collected data from 91 stage III/IV NSCLC cases treated with Volumetric modulated arc therapy (VMAT). Patient data were collected according to the radiotherapy timeline for four phases: clinical characteristics, radiomics features, radiation dosimetry parameters, and hematological indexes during treatment. Risk assessment models for single-phase and stepwise fusion phases were established according to logistic regression. In addition, a nomogram of the final fusion phase model and risk classification system was generated. Receiver operating characteristic (ROC), decision curve, and calibration curve analysis were conducted to internally validate the nomogram to analyze its discrimination.

Results

Smoking status, PTV and lung radiomics feature, lung and esophageal dosimetry parameters, and platelets at the third week of radiotherapy were independent risk factors for the four single-phase models. The ROC result analysis of the risk assessment models created by stepwise phase fusion were: (area under curve [AUC]: 0.67,95% confidence interval [CI]: 0.52–0.81), (AUC: 0.82,95%CI: 0.70–0.94), (AUC: 0.90,95%CI: 0.80–1.00), and (AUC:0.90,95%CI: 0.80–1.00), respectively. The nomogram based on the final fusion phase model was validated using calibration curve analysis and decision curve analysis, demonstrating good consistency and clinical utility. The nomogram-based risk classification system could correctly classify cases into three diverse risk groups: low-(ratio:3.6%; 0 < score < 135), intermediate-(ratio:30.7%, 135 < score < 160) and high-risk group (ratio:80.0%, score > 160).

Conclusions

In our study, the risk assessment model makes it easy for physicians to assess the risk of grade ≥ 2 radiation pneumonitis at various phases in the radiotherapy process, and the risk classification system and nomogram identify the patient’s risk level after completion of radiation therapy.

Machine learning classification of mediastinal lymph node metastasis in NSCLC: a multicentre study in a Western European patient population

BACKGROUND

[¹⁸F] FDG PET-CT has an important role in the initial staging of lung cancer; however, accurate differentiation between activity in malignant and benign intrathoracic lymph nodes on PET-CT scans can be challenging. The purpose of the current study was to investigate the effect of incorporating primary tumour data and clinical features to differentiate between [¹⁸F] FDG-avid malignant and benign intrathoracic lymph nodes.

METHODS

We retrospectively selected lung cancer patients who underwent PET-CT for initial staging in two centres in the Netherlands. The primary tumour and suspected lymph node metastases were annotated and cross-referenced with pathology results. Lymph nodes were classified as malignant or benign. From the image data, we extracted radiomic features and trained the classifier model using the extreme gradient boost (XGB) algorithm. Various scenarios were defined by selecting different combinations of data input and clinical features. Data from centre 1 were used for training and validation of the models using the XGB algorithm. To determine the performance of the model in a different hospital, the XGB model was tested using data from centre 2.

RESULTS

Adding primary tumour data resulted in a significant gain in the performance of the trained classifier model. Adding the clinical information about distant metastases did not lead to significant improvement. The performance of the model in the test set (centre 2) was slightly but statistically significantly lower than in the validation set (centre 1).

CONCLUSIONS

Using the XGB algorithm potentially leads to an improved model for the classification of intrathoracic lymph nodes. The inclusion of primary tumour data improved the performance of the model, while additional knowledge of distant metastases did not. In patients in whom metastases are limited to lymph nodes in the thorax, this may reduce costly and invasive procedures such as endobronchial ultrasound or mediastinoscopy procedures.

An Investigation on Radiomics Feature Handling for HNSCC Staging Classification

The incidence of Head and Neck Squamous Cell Carcinoma (HNSCC) has been growing in the last few decades. Its diagnosis is usually performed through clinical evaluation and analyzing radiological images, then confirmed by histopathological examination, an invasive and time-consuming operation. The recent advances in the artificial intelligence field are leading to interesting results in the early diagnosis, personalized treatment and monitoring of HNSCC only by analyzing radiological images, without performing a tissue biopsy. The large amount of radiological images and the increasing interest in radiomics approaches can help to develop machine learning (ML) methods to support diagnosis. In this work, we propose an ML method based on the use of radiomics features, extracted from CT and PET images, to classify the disease in terms of pN-Stage, pT-Stage and Overall Stage. After the extraction of radiomics features, a selection step is performed to remove dataset redundancy. Finally, ML methods are employed to complete the classification task. Our pipeline is applied on the “Head-Neck-PET-CT” TCIA open-source dataset, considering a cohort of 201 patients from four different institutions. An AUC of 97%, 83% and 93% in terms of pN-Stage, pT-Stage and Overall Stage classification, respectively, is achieved. The obtained results are promising, showing the potential efficiency of the use of radiomics approaches in staging classification.

Assessment of a computed tomography-based radiomics approach for assessing lung function in lung cancer patients

PURPOSE

We aimed to assess radiomics approaches for estimating three pulmonary function test (PFT) results (forced expiratory volume in one second [FEV1], forced vital capacity [FVC], and the ratio of FEV1 to FVC [FEV₁/FVC]) using data extracted from chest computed tomography (CT) images.

METHODS

This retrospective study included 85 lung cancer patients (mean age, 75 years ±8; 69 men) who underwent stereotactic body radiotherapy between 2012 and 2020. Their pretreatment chest breath-hold CT and PFT data before radiotherapy were obtained. A total of 107 radiomics features (Shape: 14, Intensity: 18, Texture: 75) were extracted using two methods: extraction of the lung tissue (<-250 HU) (APPROACH 1), and extraction of small blood vessels and lung tissue (APPROACH 2). The PFT results were estimated using the least absolute shrinkage and selection operator regression. Pearson's correlation coefficients (r) were determined for all PFT results, and the area under the curve (AUC) was calculated for FEV₁/FVC (<70 %). Finally, we compared our approaches with the conventional formula (Conventional).

RESULTS

For the estimated FEV₁/FVC, the Pearson's r were 0.21 (P =.06), 0.69 (P <.01), and 0.73 (P <.01) for Conventional, APPROACH 1, and APPROACH 2, respectively; the AUCs for FEV₁/FVC (<70 %) were 0.67 (95 % confidence interval [CI]: 0.55, 0.79), 0.82 (CI: 0.72, 0.91; P =.047) and 0.86 (CI: 0.78, 0.94; P =.01), respectively.

CONCLUSIONS

The radiomics approach performed better than the conventional equation and may be useful for assessing lung function based on CT images.

A deep learning-based radiomics approach to predict head and neck tumor regression for adaptive radiotherapy

Early regression—the regression in tumor volume during the initial phase of radiotherapy (approximately 2 weeks after treatment initiation)—is a common occurrence during radiotherapy. This rapid radiation-induced tumor regression may alter target coordinates, necessitating adaptive radiotherapy (ART). We developed a deep learning-based radiomics (DLR) approach to predict early head and neck tumor regression and thereby facilitate ART. Primary gross tumor volume (GTVp) was monitored in 96 patients and nodal GTV (GTVn) in 79 patients during treatment. All patients underwent two computed tomography (CT) scans: one before the start of radiotherapy for initial planning and one during radiotherapy for boost planning. Patients were assigned to regression and nonregression groups according to their median tumor regression rate (ΔGTV/treatment day from initial to boost CT scan). We input a GTV image into the convolutional neural network model, which was pretrained using natural image datasets, via transfer learning. The deep features were extracted from the last fully connected layer. To clarify the prognostic power of the deep features, machine learning models were trained. The models then predicted the regression and nonregression of GTVp and GTVn and evaluated the predictive performance by 0.632 + bootstrap area under the curve (AUC). Predictive performance for GTVp regression was highest using the InceptionResNetv2 model (mean AUC = 0.75) and that for GTVn was highest using NASNetLarge (mean AUC = 0.73). Both models outperformed the handcrafted radiomics features (mean AUC = 0.63 for GTVp and 0.61 for GTVn) or clinical factors (0.64 and 0.67, respectively). DLR may facilitate ART for improved radiation side-effects and target coverage.

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

(1) Background: 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. (2) Methods: 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). (3) Results: 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. (4) Conclusion: 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.

Risk Score Generated from CT-Based Radiomics Signatures for Overall Survival Prediction in Non-Small Cell Lung Cancer

Simple Summary

Despite recent advancements in lung cancer treatment, individuals with lung cancer have a dismal 5-year survival rate of only 15%. In patients with non-small cell lung cancer (NSCLC), medical images have lately been employed as a valuable marker for predicting overall survival. The primary goal of this study was to develop a risk score based on computed tomography (CT) based radiomics feature signatures that may be used to predict survival in NSCLC patients. After analyzing 577 NSCLC patients from two data sets, we discovered that the risk score model’s prediction ability as a prognostic indicator was superior to other clinical indicators (age, stage, and gender), and the possibility of patient risk stratification with survival was evaluated using a risk score representation of 10 radiomics signatures. According to this study, the risk score generated using CT-based radiomics signatures promises to predict overall survival in NSCLC patients.

Abstract

This study aimed to create a risk score generated from CT-based radiomics signatures that could be used to predict overall survival in patients with non-small cell lung cancer (NSCLC). We retrospectively enrolled three sets of NSCLC patients (including 336, 84, and 157 patients for training, testing, and validation set, respectively). A total of 851 radiomics features for each patient from CT images were extracted for further analyses. The most important features (strongly linked with overall survival) were chosen by pairwise correlation analysis, Least Absolute Shrinkage and Selection Operator (LASSO) regression model, and univariate Cox proportional hazard regression. Multivariate Cox proportional hazard model survival analysis was used to create risk scores for each patient, and Kaplan–Meier was used to separate patients into two groups: high-risk and low-risk, respectively. ROC curve assessed the prediction ability of the risk score model for overall survival compared to clinical parameters. The risk score, which developed from ten radiomics signatures model, was found to be independent of age, gender, and stage for predicting overall survival in NSCLC patients (HR, 2.99; 95% CI, 2.27–3.93; p < 0.001) and overall survival prediction ability was 0.696 (95% CI, 0.635–0.758), 0.705 (95% CI, 0.649–0.762), 0.657 (95% CI, 0.589–0.726) (AUC) for 1, 3, and 5 years, respectively, in the training set. The risk score is more likely to have a better accuracy in predicting survival at 1, 3, and 5 years than clinical parameters, such as age 0.57 (95% CI, 0.499–0.64), 0.552 (95% CI, 0.489–0.616), 0.621 (95% CI, 0.544–0.689) (AUC); gender 0.554, 0.546, 0.566 (AUC); stage 0.527, 0.501, 0.459 (AUC), respectively, in 1, 3 and 5 years in the training set. In the training set, the Kaplan–Meier curve revealed that NSCLC patients in the high-risk group had a lower overall survival time than the low-risk group (p < 0.001). We also had similar results that were statistically significant in the testing and validation set. In conclusion, risk scores developed from ten radiomics signatures models have great potential to predict overall survival in NSCLC patients compared to the clinical parameters. This model was able to stratify NSCLC patients into high-risk and low-risk groups regarding the overall survival prediction.