Radiomics and Dosiomics Signature From Whole Lung Predicts Radiation Pneumonitis: A Model Development Study With Prospective External Validation and Decision-curve Analysis

Multicenter development of a deep learning radiomics and dosiomics nomogram to predict radiation pneumonia risk in non-small cell lung cancer

Radiation pneumonia (RP) is the most common side effect of chest radiotherapy, and can affect patients' quality of life. This study aimed to establish a combined model of radiomics, dosiomics, deep learning (DL) based on simulated location CT and dosimetry images combining with clinical parameters to improve the predictive ability of ≥ 2 grade RP (RP2) in patients with non-small cell lung cancer (NSCLC). This study retrospectively collected 245 patients with NSCLC who received radiotherapy from three hospitals. 162 patients from Hospital I were randomly divided into training cohort and internal validation cohort according to 7:3. 83 patients from two other hospitals served as an external validation cohort. Multivariate analysis was used to screen independent clinical predictors and establish clinical model (CM). The radiomic and dosiomics (RD) features and DL features were extracted from simulated location CT and dosimetry images based on the region of interest (ROI) of total lung-PTV (TL-PTV). The features screened by the t-test and least absolute shrinkage and selection operator (LASSO) were used to construct the RD and DL model, and RD-score and DL-score were calculated. RD-score, DL-score and independent clinical features were combined to establish deep learning radiomics and dosiomics nomogram (DLRDN). The model performance was evaluated by area under the curve (AUC). Three clinical factors, including V20, V30, and mean lung dose (MLD), were used to establish the CM. 7 RD features including 4 radiomics features and 3 dosiomics features were selected to establish RD model. 10 DL features were selected to establish DL model. Among the different models, DLRDN showed the best predictions, with the AUCs of 0.891 (0.826-0.957), 0.825 (0.693-0.957), and 0.801 (0.698-0.904) in the training cohort, internal validation cohort and external validation cohort, respectively. DCA showed that DLRDN had a higher overall net benefit than other models. The calibration curve showed that the predicted value of DLRDN was in good agreement with the actual value. Overall, radiomics, dosiomics, and DL features based on simulated location CT and dosimetry images have the potential to help predict RP2. The combination of multi-dimensional data produced the optimal predictive model, which could provide guidance for clinicians.

A systematic review and meta-analysis of the utility of quantitative, imaging-based approaches to predict radiation-induced toxicity in lung cancer patients

BACKGROUND AND PURPOSE

To conduct a systematic review and meta-analysis of the performance of radiomics, dosiomics and machine learning in generating toxicity prediction in thoracic radiotherapy.

MATERIALS AND METHODS

An electronic database search was conducted and dual-screened by independent authors to identify eligible studies for systematic review and meta-analysis. Data was extracted and study quality was assessed using TRIPOD for machine learning studies, RQS for Radiomics and RoB for dosiomics.

RESULTS

10,703 studies were identified, and 5,252 entered screening. 104 studies including 23,373 patients were eligible for systematic review. Primary toxicity predicted was radiation pneumonitis (81), followed by esophagitis (12) and lymphopenia (4). Fourty-two studies studying radiation pneumonitis were eligible for meta-analysis, with pooled area-under-curve (AUC) of 0.82 (95% CI 0.79-0.85). Studies with machine learning had the best performance, with classical and deep learning models having similar performance. There is a trend towards an improvement of the performance of models with the year of publication. There is variability in study quality among the three study categories and dosiomic studies scored the highest among these. Publication bias was not observed.

CONCLUSION

The majority of existing literature using radiomics, dosiomics and machine learning has focused on radiation pneumonitis prediction. Future research should focus on toxicity prediction of other organs at risk and the adoption of these models into clinical practice.

The impact of pre-existing interstitial lung disease on radiation and checkpoint inhibitor pneumonitis in lung cancer patients: a systematic review and meta-analysis

Background

Current research presents conflicting evidence on whether pre-existing interstitial lung disease (ILD) serves as a risk factor for radiation pneumonitis (RP) and checkpoint inhibitor pneumonitis (CIP) in patients with lung cancer.

Objectives

This study aims to systematically evaluate the impact of pre-existing ILD on the risk of developing RP and CIP in lung cancer patients.

Design

A systematic review and meta-analysis was conducted using a random-effects model.

Data sources and methods

PubMed, Embase, and Web of Science were searched to identify relevant studies. A random-effects model was applied to estimate the risk and incidence of RP and CIP in lung cancer patients with pre-existing ILD compared to those without ILD. Sensitivity analyses were performed to assess the robustness of the pooled findings, and potential publication bias was evaluated using Begg's and Egger's tests.

Results

A total of 12 studies involving 2576 patients were included in the RP risk assessment, while 29 studies with 8037 patients were analyzed for CIP risk. The pooled results indicated that pre-existing ILD significantly increased the risk of developing any-grade RP (odds ratio (OR): 3.63, 95% confidence interval (CI): 2.26-5.83) and severe RP (OR: 6.10, 95% CI: 2.68-13.86) in lung cancer patients. Subgroup analyses identified stereotactic body radiation therapy as the modality associated with the lowest risk of any-grade RP in these patients. Similarly, pre-existing ILD was associated with a significantly higher risk of any-grade CIP (OR: 3.86, 95% CI: 2.65-5.61) and severe CIP (OR: 3.24, 95% CI: 2.07-5.07), with anti-programmed cell death 1 therapies showing the highest CIP risk.

Conclusion

Pre-existing ILD markedly increases the risk of both RP and CIP in lung cancer patients. These findings underscore the critical importance of thorough ILD evaluation and the development of personalized treatment strategies to mitigate these risks prior to initiating cancer therapy.

Plan complexity and dosiomics signatures for gamma passing rate classification in volumetric modulated arc therapy: External validation across different LINACs

PURPOSE

This study aims to enhance gamma passing rate (GPR) classification by integrating plan complexity signature, dosiomics signature, and comprehensive plan parameters, and to validate this method using data from different linear accelerators (LINACs).

METHODS

This study included 235 volumetric modulated arc therapy (VMAT) treatment plans delivered using the TrueBeam LINAC as the primary dataset, along with 47 plans from the VitalBeam LINAC for external validation. The primary dataset was split into training (N = 166) and test (N = 69) subsets. Extracted features included 47 plan complexity metrics, 851 dosiomics features, and 20 plan parameters. Plan complexity score (PCscore) and dosiomics score (Doscore) were derived using the least absolute shrinkage and selection operator (LASSO). Four classification models were developed by combining PCscore, Doscore, and plan parameters according to a gamma criterion of 2 %/2 mm (γ2%/2 mm). A nomogram was constructed to combine these signatures with plan parameters. Model performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).

RESULTS

The combined model incorporating PCscore, Doscore, and plan parameters exhibited high discriminative power, with areas under the curve (AUC) of 0.894, 0.899, and 0.904 for the training, test, and external datasets, respectively. At γ3%/2 mm, the model maintained robust performance with AUCs of 0.842 and 0.833 in the test and external datasets. Calibration curves and DCA validated the model's effectiveness.

CONCLUSIONS

Integrating plan complexity and dosiomics signatures with key plan parameters significantly improves GPR classification for VMAT treatment plans, offering a robust approach for patient-specific quality assurance (PSQA).

Potential of Radiomics, Dosiomics, and Dose Volume Histograms for Tumor Response Prediction in Hepatocellular Carcinoma following 90Y-SIRT

PURPOSE

We evaluate the role of radiomics, dosiomics, and dose-volume constraints (DVCs) in predicting the response of hepatocellular carcinoma to selective internal radiation therapy with 90Y with glass microspheres.

METHODS

99mTc-macroagregated albumin (99mTc-MAA) and 90Y SPECT/CT images of 17 patients were included. Tumor responses at three months were evaluated using modified response evaluation criteria in solid tumors criteria and patients were categorized as responders or non-responders. Dosimetry was conducted using the local deposition method (Dose) and biologically effective dosimetry. A total of 264 DVCs, 321 radiomic features, and 321 dosiomic features were extracted from the tumor, normal perfused liver (NPL), and whole normal liver (WNL). Five different feature selection methods in combination with eight machine learning algorithms were employed. Model performance was evaluated using area under the AUC, accuracy, sensitivity, and specificity.

RESULTS

No statistically significant differences were observed between neither the dose metrics nor radiomicas or dosiomics features of responders and non-responder groups. 90Y-dosiomics models with any given set of inputs outperformed other models. This was also true for 90Y-radiomics from SPECT and SPECT-clinical features, achieving an AUC, accuracy, sensitivity, and specificity of 1. Among MAA-dosiomic and radiomic models, two models showed AUC ≥ 0.91. While the performance of MAA-dose volume histogram (DVH)-based models were less promising, the 90Y-DVH-based models showed strong performance (AUC ≥ 0.91) when considered independently of clinical features.

CONCLUSION

This study demonstrated the potential of 99mTc-MAA and 90Y SPECT-derived radiomics, dosiomics, and dosimetry metrics in establishing predictive models for tumor response.

Radiomics and prognostic nutritional index for predicting postoperative survival in esophageal carcinoma

BACKGROUND

Surgery offers the potential for a radical cure and prolonged survival in individuals diagnosed with esophageal squamous cell carcinoma (ESCC). However, survival rates exhibit significant variability among patients. Accurately assessing surgical outcomes remains a critical challenge. This study aimed to evaluate the predictive value of preoperative radiomics and the prognostic nutritional index for individuals with ESCC and to develop a comprehensive model for estimating postoperative overall survival (OS) in these patients.

METHODS

A retrospective analysis was conducted on 466 patients with ESCC from two medical centers. The dataset was randomly divided into a training cohort (TC, hospital 1, 246 cases), an internal validation cohort (IVC, hospital 1, 106 cases), and an external validation cohort (EVC, hospital 2, 114 cases). Radiological features were extracted after delineating the region of interest, followed by the application of the least absolute shrinkage and selection operator (LASSO) regression to identify optimal radiomics features and compute the radiomics score (RS). Independent prognostic factors identified via Cox regression analysis were incorporated with the RS to construct a combined nomogram. The predictive performance of the model was evaluated using the concordance index, time-dependent receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis.

RESULTS

The predictive model, which integrated preoperative radiomics, the prognostic nutritional index, and tumor-node-metastasis (TNM) staging to estimate the 3 year OS rate, achieved area under the ROC curve (AUC) values of 0.812, 0.786, and 0.810 in the TC, IVC, and EVC, respectively, demonstrating excellent prognostic accuracy. These values surpassed the AUCs of the TNM staging model in the TC, IVC, and EVC, which were 0.717, 0.612, and 0.699, respectively. The combined model's concordance indexes in the TC, IVC, and EVC were 0.780, 0.760, and 0.764, respectively. Calibration and decision curves highlighted the nomogram's superior calibration and clinical utility.

CONCLUSION

This study developed a predictive model by combining radiomics with the prognostic nutritional index, enabling the estimation of OS in postoperative patients with ESCC. This model holds promise as a tool for preoperative risk stratification.

Voxel-level radiomics and deep learning for predicting pathologic complete response in esophageal squamous cell carcinoma after neoadjuvant immunotherapy and chemotherapy

BACKGROUND

Accurate prediction of pathologic complete response (pCR) following neoadjuvant immunotherapy combined with chemotherapy (nICT) is crucial for tailoring patient care in esophageal squamous cell carcinoma (ESCC). This study aimed to develop and validate a deep learning model using a novel voxel-level radiomics approach to predict pCR based on preoperative CT images.

METHODS

In this multicenter, retrospective study, 741 patients with ESCC who underwent nICT followed by radical esophagectomy were enrolled from three institutions. Patients from one center were divided into a training set (469 patients) and an internal validation set (118 patients) while the data from the other two centers was used as external validation sets (120 and 34 patients, respectively). The deep learning model, Vision-Mamba, integrated voxel-level radiomics feature maps and CT images for pCR prediction. Additionally, other commonly used deep learning models, including 3D-ResNet and Vision Transformer, as well as traditional radiomics methods, were developed for comparison. Model performance was evaluated using accuracy, area under the curve (AUC), sensitivity, specificity, and prognostic stratification capabilities. The SHapley Additive exPlanations analysis was employed to interpret the model's predictions.

RESULTS

The Vision-Mamba model demonstrated robust predictive performance in the training set (accuracy: 0.89, AUC: 0.91, sensitivity: 0.82, specificity: 0.92) and validation sets (accuracy: 0.83-0.91, AUC: 0.83-0.92, sensitivity: 0.73-0.94, specificity: 0.84-1.0). The model outperformed other deep learning models and traditional radiomics methods. The model's ability to stratify patients into high and low-risk groups was validated, showing superior prognostic stratification compared with traditional methods. SHAP provided quantitative and visual model interpretation.

CONCLUSIONS

We present a voxel-level radiomics-based deep learning model to predict pCR to neoadjuvant immunotherapy combined with chemotherapy based on pretreatment diagnostic CT images with high accuracy and robustness. This model could provide a promising tool for individualized management of patients with ESCC.

Radiation-induced lung injury: from mechanism to prognosis and drug therapy

Radiation induced lung injury, known as the main complication of thoracic radiation, remains to be a major resistance to tumor treatment. Based on the recent studies on radiation-induced lung injury, this review collated the possible mechanisms at the level of target cells and key pathways, corresponding prognostic models including predictors, patient size, number of centers, radiotherapy technology, construction methods and accuracy, and pharmacotherapy including drugs, targets, therapeutic effects, impact on anti-tumor treatment and research types. The research priorities and limitations are summarized to provide a reference for the research and management of radiation-induced lung injury.

Clinical, Dosimetric and Radiomic Features Predictive of Lung Toxicity After (Chemo)Radiotherapy

BACKGROUND

Treatment of locally advanced non small cell lung cancer (LA-NSCLC) is based on (chemo)radiotherapy, which may cause acute lung toxicity: radiation pneumonitis (RP). Its frequency seems to increase by the use of adjuvant durvalumab therapy.

AIMS

To identify clinical, dosimetric, and radiomic factors associated with grade (G)≥2 RP and build a prediction model based on selected parameters.

PATIENTS AND METHODS

This is a retrospective multicenter cohort study including patients receiving radiation therapy between 2015 and 2019 for LA-NSCLC. Baseline computed tomography scanners were segmented to extract radiomic features from the "Lung - Tumor" volume. Variables associated with the risk of G≥2 RP in the descriptive analysis were then selected for explanatory analysis, followed by predictive analysis.

RESULTS

153 patients were included in 4 centers (51 with G≥2 RP). Factors associated with G≥2 RP included a high initial hemoglobin level, dosimetric factors (mean dose to healthy lungs, lung V20Gy and V13Gy), the addition of maintenance durvalumab, and 7 radiomic features (intensity distribution and texture). Other factors were associated with an increased risk of G≥2 RP in our explanatory model, such as older age, low Tiffeneau ratio, and a decreased initial platelet count. The best-performing predictive model was a random forest-based learning model using clinical, dosimetric, and radiomic variables, with an area under the ROC curve of 0.72 (95%CI [0.63; 0.80]) versus 0.64 for models using one type of data.

CONCLUSION

The addition of radiomic features to clinical and dosimetric ones improves prediction of the occurrence of G≥2 RP in patients receiving radiotherapy for lung cancer.

Normal tissue complication probability model for acute oral mucositis in patients with head and neck cancer undergoing carbon ion radiation therapy based on dosimetry, radiomics, and dosiomics

BACKGROUND AND PURPOSE

To develop a normal tissue complication probability (NTCP) model for predicting grade ≥ 2 acute oral mucositis (AOM) in head and neck cancer patients undergoing carbon-ion radiation therapy (CIRT).

METHODS AND MATERIALS

We retrospectively included 178 patients, collecting clinical, dose-volume histogram (DVH), radiomics, and dosiomics data. Patients were randomly divided into training (70%) and test sets (30%). Feature selection involved univariable logistic regression, least absolute shrinkage and selection operator regression, stepwise backward regression, and Spearman's correlation test, with the bootstrap method ensuring reliability. Multivariable models were built on the training set and evaluated using the test set.

RESULTS

The optimal NTCP model incorporated a DVH parameter (V37Gy [relative biological effectiveness, RBE]), radiomics, and dosiomics features, achieving an area under the curve (AUC) of 0.932 in the training set and 0.959 in the test set. This hybrid model outperformed those based on single DVH, radiomics, dosiomics, or clinical data (Bonferroni-adjusted p < 0.001 and ΔAUC > 0 for all comparisons in 1,000 bootstrap validations). Calibration curves showed strong agreement between predictions and outcomes. A 44.0 % AOM risk threshold was proposed, yielding accuracies of 87.1 % in the training set and 90.7 % in the test set.

CONCLUSIONS

We developed the first NTCP model for estimating AOM risk in head and neck cancer patients undergoing CIRT and proposed a risk stratification. This model may assist in clinical decision-making and improve treatment planning for AOM prevention and management by identifying high-risk patients.

Radiomics and Deep Learning Prediction of Immunotherapy-Induced Pneumonitis From Computed Tomography

PURPOSE

Primary barriers to application of immune checkpoint inhibitor (ICI) therapy for cancer include severe side effects (such as potentially life threatening pneumonitis [PN]), which can cause the discontinuation of treatment. Predicting which patients may develop PN while on ICI would improve both safety and potential efficacy because treatments could be safely administered for longer or discontinued before severe toxicity.

METHODS

Starting from a cohort of 3,351 patients with cancer who received previous ICI therapy at the Vanderbilt University Medical Center, we curated 2,700 contrast chest computed tomography (CT) volumes for 671 patients. Three different pure imaging models predicted the potential for PN using only a single time point before the first ICI dose.

RESULTS

The first model used 109 radiomics features only and achieved an AUC of 0.747 (CI, 0.705 to 0.789) with a positive predictive value (PPV) of 0.244 (CI, 0.211 to 0.276) at a sensitivity of 0.553 (CI, 0.485 to 0.621) using mainly features describing the global lung properties. The second model used a convolutional neural network (CNN) on the raw CTs to improve to an AUC of 0.819 (CI, 0.781 to 0.857) with a PPV of 0.244 (CI, 0.203 to 0.284) at a sensitivity of 0.743 (CI, 0.681 to 0.806). The third model combined both radiomics and deep learning but, with an AUC of 0.829 (CI, 0.797 to 0.862) and a PPV of 0.254 (CI, 0.228 to 0.281) at a sensitivity of 0.780 (CI, 0.721 to 0.840), did not show a significant improvement on the CNN-only model.

CONCLUSION

This new model suggests the utility of deep learning in PN prediction over traditional pure radiomics and promises better management for patients receiving ICI and the ability to better stratify patients in immunotherapy drug trials.

CT-based whole lung radiomics nomogram to identify middle-aged and elderly COVID-19 patients at high risk of progressing to critical disease

BACKGROUND

COVID-19 remains widespread and poses a threat to people's physical and mental health, especially middle-aged and elderly individuals. Early identification of COVID-19 patients at high risk of progressing to critical disease helps improve overall patient outcomes and healthcare efficiency.

PURPOSE

To develop a radiomics nomogram to predict the risk of newly admitted middle-aged and elderly COVID-19 patients progressing to critical disease.

METHODS

A total of 794 patients (aged 40 years or above) were retrospectively included in the study from two institutions, all of them were with non-critical COVID-19 on admission. At follow-up, patients were divided into non-critical group and critical group. About 443 patients (384 non-critical and 59 critical) from the first hospital were randomly assigned to the training (n = 311) and internal validation (n = 132) set in a 7:3 ratio. Additionally, an independent external cohort of 351 patients (292 non-critical and 59 critical) from another hospital was evaluated. Radiomics signatures and clinical indicators were used to build a radiomics model and a clinical model after computed tomography (CT) image processing, CT whole-lung segmentation, feature extraction, and feature selection. The radiomics nomogram model integrated radiomics model and clinical model. The receiver operating characteristic curve (AUC) was used to assess the performance of the proposed models. Calibration curves and decision curve analysis were used to assess the performance of the radiomics nomogram.

RESULTS

For the training, internal validation, and external validation sets, the AUC values of the radiomic nomogram for the prediction of COVID-19 progression were 0.916, 0.917, and 0.890, respectively. Calibration curves indicated that there was no significant departure between prediction and observation in three sets. The decision curve image demonstrated the clinical utility of the nomogram model.

CONCLUSIONS

Our nomogram model incorporates radiomics features and clinical indicators, it provides a new pathway to increase predictive accuracy or clinical utility, further helping to provide personalized management for middle-aged and elderly patients with COVID-19.

Multi-Institutional MR-Derived Radiomics to Predict Post-Exenteration Disease Recurrence in Patients With T4 Rectal Cancer

INTRODUCTION

Local recurrence and distant metastasis remain a concern in advanced rectal cancer, with up to 10% and 20%-30% of patients suffering local and distal progression, respectively. Radiomics refers to a novel technology that extracts and analyses quantitative imaging features from images, which can be subsequently used to develop and test clinical models predictive of outcomes. We aim to develop and test an MRI-based radiomics nomogram predictive of disease recurrence in patients with T4 rectal cancer.

METHODS

We conducted a multi-institutional retrospective analysis of 55 patients with T4 rectal cancer treated with neoadjuvant chemoradiotherapy followed by exenterative surgery. Radiomic features were extracted from pre-treatment T2-weighted MRI scans and used to construct predictive models. The top-performing radiomic signatures were identified, and internal validation with 1000 bootstrap samples was performed to calculate optimism-corrected performance measures.

RESULTS

Two radiomic signatures were identified as strong predictors of post-operative disease recurrence. The best-performing model achieved an optimism-corrected AUC of 0.75, demonstrating good discriminative ability. Calibration plots showed a satisfactory fit of the predictions to the actual rates, and decision curve analyses confirmed the positive net benefit of the models.

CONCLUSION

The MRI-based radiomics nomogram provides a promising tool for predicting disease recurrence in T4 rectal cancer patients post-exenteration. This model could improve risk stratification and guide more personalized treatment strategies. Further studies with larger cohorts and external validation are needed to confirm these findings and enhance the model's generalizability.

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.

Machine Learning Model Integrating CT Radiomics of the Lung to Predict Checkpoint Inhibitor Pneumonitis in Patients with Advanced Cancer

ObjectiveCheckpoint inhibitor pneumonitis (CIP) is a potentially life-threatening immune-related adverse event. Efficient strategies to select patients at risk are still required. The aim of our study was to assess the utility of a machine learning model, integrating pre-treatment CT lung radiomics features with clinical data, to predict patients at risk of developing CIP.MethodsIn this retrospective study, 116 patients with varied malignancies treated with immune checkpoint inhibitors (ICIs) were included. In this cohort, 35 patients presented with CIP and 81 patients did not. Each lung and its lobes were segmented on pre-treatment CT scans to perform a handcrafted radiomic analysis. Radiomic features were associated with clinical parameters to build generalized linear (GLM) and random forest (RF) models, to predict occurrence of CIP. The models were fine-tuned, validated and tested using a nested 5-fold cross-validation method.ResultsThe RF models combining radiomic and clinical features showed the best performances with an area under the ROC curve (AUC) of 0.75 (95%CI:0.62-0.88) on the test set. The most accurate clinical model was a RF model and achieved an AUC of 0.72 (95%CI:0.51-0.92). The best radiomic model was a GLM model and achieved an AUC of 0.71 (95%CI:0.58-0.84).ConclusionsOur CT-based lung radiomic models showed moderate to good performance at predicting CIP. We demonstrated the potential role of machine learning models associating clinical parameters and lung CT radiomic features to better identify patients treated with ICIs at risk of developing CIP.Advances in knowledge: Radiomics analysis of the lung parenchyma could be used as a non-invasive tool to select patients at risk of developing immune-checkpoint pneumonitis.

Radiomic and dosimetric parameter-based nomogram predicts radiation esophagitis in patients with non-small cell lung cancer undergoing combined immunotherapy and radiotherapy

Background

The combination of immune checkpoint inhibitors (ICIs) and radiotherapy (RT) may increase the risk of radiation esophagitis (RE). This study aimed to establish and validate a new nomogram to predict RE in patients with non-small cell lung cancer (NSCLC) undergoing immunochemotherapy followed by RT (ICI-RT).

Methods

The 102 eligible patients with NSCLC treated with ICI-RT were divided into training (n = 71) and validation (n = 31) cohorts. Clinicopathologic features, dosimetric parameters, inflammatory markers, and radiomic score (Rad-score) were included in the univariate logistic regression analysis, and factors with p < 0.05 in the univariate analysis were included in the multivariate logistic regression analysis. Factors with significant predictive values were obtained and used for developing the nomogram. The area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve were used to validate the model.

Results

A total of 38 (37.3%) patients developed RE. Univariate and multivariate analyses identified the following independent predictors of RE: a maximum dose delivered to the esophagus >58.4 Gy, a mean esophagus dose >13.3 Gy, and the Rad-score. The AUCs of the nomogram in the training and validation cohorts were 0.918 (95% confidence interval [CI]: 0.824-1.000) and 0.833 (95% CI: 0.697-0.969), respectively, indicating good discrimination. The calibration curves showed good agreement between the predicted occurrence of RE and the actual observations. The decision curve showed a satisfactory positive net benefit at most threshold probabilities, suggesting a good clinical effect.

Conclusions

We developed and validated a nomogram based on imaging histological features and RT dosimetric parameters. This model can effectively predict the occurrence of RE in patients with NSCLC treated using ICI-RT.

A CT-Based Lung Radiomics Nomogram for Classifying the Severity of Chronic Obstructive Pulmonary Disease

Background

Chronic obstructive pulmonary disease (COPD) is a major global health concern, and while traditional pulmonary function tests are effective, recent radiomics advancements offer enhanced evaluation by providing detailed insights into the heterogeneous lung changes.

Purpose

To develop and validate a radiomics nomogram based on clinical and whole-lung computed tomography (CT) radiomics features to stratify COPD severity.

Patients and Methods

One thousand ninety-nine patients with COPD (including 308, 132, and 659 in the training, internal and external validation sets, respectively), confirmed by pulmonary function test, were enrolled from two institutions. The whole-lung radiomics features were obtained after a fully automated segmentation. Thereafter, a clinical model, radiomics signature, and radiomics nomogram incorporating radiomics signature as well as independent clinical factors were constructed and validated. Additionally, receiver-operating characteristic (ROC) curve, area under the ROC curve (AUC), decision curve analysis (DCA), and the DeLong test were used for performance assessment and comparison.

Results

In comparison with clinical model, both radiomics signature and radiomics nomogram outperformed better on COPD severity (GOLD I-II and GOLD III-IV) in three sets. The AUC of radiomics nomogram integrating age, height and Radscore, was 0.865 (95% CI, 0.818-0.913), 0.851 (95% CI, 0.778-0.923), and 0.781 (95% CI, 0.740-0.823) in three sets, which was the highest among three models (0.857; 0.850; 0.774, respectively) but not significantly different (P > 0.05). Decision curve analysis demonstrated the superiority of the radiomics nomogram in terms of clinical usefulness.

Conclusion

The present work constructed and verified the novel, diagnostic radiomics nomogram for identifying the severity of COPD, showing the added value of chest CT to evaluate not only the pulmonary structure but also the lung function status.

Constructing and identifying an eighteen-gene tumor microenvironment prognostic model for non-small cell lung cancer

BACKGROUND

The tumor microenvironment (TME) plays a crucial role in tumorigenesis and tumor progression. This study aimed to identify novel TME-related biomarkers and develop a prognostic model for patients with non-small-cell lung cancer (NSCLC).

METHODS

After downloading and preprocessing data from The Cancer Genome Atlas (TCGA) data portal and Gene Expression Omnibus (GEO) datasets, we classified the molecular subtypes using the "NMF" R package. We performed survival analysis and quantified immune scores between clusters. A Cox proportional hazards model was then constructed, and its formula was produced. We assessed model performance and clinical utility. A prediction nomogram was also constructed and validated. Additionally, we explored the potential regulatory mechanisms of our TME gene signature using Gene Set Enrichment Analysis (GSEA).

RESULTS

From data processing and univariate Cox regression analysis, 57 TME-related prognostic genes were identified, and two significantly distinct clusters were established. Using Cox regression and Lasso regression, an 18-gene TME-related prognostic model was developed. Patients were stratified into high- and low-risk groups based on the risk score, with survival analysis showing that the low-risk group had significantly better outcomes than the high-risk group (P < 0.01). ROC curve analysis demonstrated strong predictive performance, with 1-year, 3-year, and 5-year AUC values ranging from 0.654 to 0.702 across different cohorts. The model accurately predicted survival outcomes across subgroups with varying clinical features, and its predictive accuracy was validated through a nomogram.

CONCLUSIONS

We developed a prognostic model based on TME-related genes in NSCLC. Our 18-gene TME signature can effectively predict the prognosis of NSCLC with high accuracy.

Predicting grade II-IV bone marrow suppression in patients with cervical cancer based on radiomics and dosiomics

Objective

The objective of this study is to develop a machine learning model integrating clinical characteristics with radiomics and dosiomics data, aiming to assess their predictive utility in anticipating grade 2 or higher BMS occurrences in cervical cancer patients undergoing radiotherapy.

Methods

A retrospective analysis was conducted on the clinical data, planning CT images, and radiotherapy planning documents of 106 cervical cancer patients who underwent radiotherapy at our hospital. The patients were randomly divided into training set and test set in an 8:2 ratio. The radiomic features and dosiomic features were extracted from the pelvic bone marrow (PBM) of planning CT images and radiotherapy planning documents, and the least absolute shrinkage and selection operator (LASSO) algorithm was employed to identify the best predictive characteristics. Subsequently, the dosiomic score (D-score) and the radiomic score (R-score) was calculated. Clinical predictors were identified through both univariate and multivariate logistic regression analysis. Predictive models were constructed by intergrating clinical predictors with DVH parameters, combining DVH parameters and R-score with clinical predictors, and amalgamating clinical predictors with both D-score and R-score. The predictive model's efficacy was assessed by plotting the receiver operating characteristic (ROC) curve and evaluating its performance through the area under the ROC curve (AUC), the calibration curve, and decision curve analysis (DCA).

Results

Seven radiomic features and eight dosiomic features exhibited a strong correlation with the occurrence of BMS. Through univariate and multivariate logistic regression analyses, age, planning target volume (PTV) size and chemotherapy were identified as clinical predictors. The AUC values for the training and test sets were 0.751 and 0.743, respectively, surpassing those of clinical DVH R-score model (AUC=0.707 and 0.679) and clinical DVH model (AUC=0.650 and 0.638). Furthermore, the analysis of both the calibration and the DCA suggested that the combined model provided superior calibration and demonstrated a higher net clinical benefit.

Conclusion

The combined model is of high diagnostic value in predicting the occurrence of BMS in patients with cervical cancer during radiotherapy.

Predicting radiation pneumonitis in lung cancer using machine learning and multimodal features: a systematic review and meta-analysis of diagnostic accuracy

OBJECTIVES

To evaluate the diagnostic accuracy of machine learning models incorporating multimodal features for predicting radiation pneumonitis in lung cancer through a systematic review and meta-analysis.

METHODS

Relevant studies were identified through a systematic search of PubMed, Web of Science, Embase, and the Cochrane Library from October 2003 to December 2023. Additional studies were located by reviewing bibliographies and relevant websites. Two independent researchers screened titles, abstracts, and full-text articles according to predefined inclusion and exclusion criteria. Data extraction was performed using standardized forms, and study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. The primary outcomes, including combined sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and area under the curve (AUC), were calculated using STATA MP-64 software(Stata Corporation LLC, College Station, USA) with a random-effects model. Meta-analysis was conducted to synthesize diagnostic accuracy measures, and analyses of heterogeneity and publication bias were performed.

RESULTS

A total of 1,406 patients with primary lung cancer were included in this systematic review, drawing data from 9 studies. The pooled analysis revealed a sensitivity of 0.74 [0.58-0.85] and a specificity of 0.91 [0.87-0.95] for machine learning models in diagnosing radiation pneumonitis. The positive likelihood ratio (PLR) was 8.69 [5.21-14.50], the negative likelihood ratio (NLR) was 0.28 [0.16-0.49], and the diagnostic odds ratio (DOR) was 30.73 [11.96-78.97]. The area under the curve (AUC) was 0.93 [0.90-0.95], indicating excellent diagnostic performance. Meta-regression analysis identified that the number of machine learning models, year of publication, and study design contributed to heterogeneity among studies. No evidence of publication bias was found. Overall, machine learning models incorporating multimodal characteristics demonstrated 75% accuracy in predicting moderate to severe radiation pneumonitis.

CONCLUSION

In conclusion, by integrating the current machine learning (ML) algorithm's ability in big data mining, a predictive model can be constructed by combining multi-modal features such as genetics, imaging, and cell factors. By selecting multiple machine learning algorithm frameworks and competing for the best combination model based on research goals, the reliability and accuracy of the radiation pneumonitis prediction model can be greatly improved.

TRIAL REGISTRATION

PROSPERO (CRD42024497599).

Xerostomia prediction in patients with nasopharyngeal carcinoma during radiotherapy using segmental dose distribution in dosiomics and radiomics models

OBJECTIVES

This study aimed to integrate radiomics and dosiomics features to develop a predictive model for xerostomia (XM) in nasopharyngeal carcinoma after radiotherapy. It explores the influence of distinct feature extraction methods and dose ranges on the performance.

MATERIALS AND METHODS

Data from 363 patients with nasopharyngeal carcinoma were retrospectively analyzed. We pioneered a dose-segmentation strategy, where the overall dose distribution (OD) was divided into four segmental dose distributions (SDs) at intervals of 15 Gy. Features were extracted using manual definition and deep learning, applying OD or SD and integrating radiomics and dosiomics, yielding corresponding feature scores (manually defined radiomics, MDR; manually defined dosiomics, MDD; deep learning-based radiomics, DLR; deep learning-based dosiomics, DLD). Subsequently, 18 models were developed by combining features and model types (random forest and support vector machine).

RESULTS AND CONCLUSION

Under OD, O(DLR_DLD) demonstrated exceptional performance, with an optimal area under the curve (AUC) of 0.81 and an average AUC of 0.71. Within SD, S(DLR_DLD) surpassed the other models, achieving an optimal AUC of 0.90 and an average AUC of 0.85. Therefore, the integration of dosiomics into radiomics can augment predictive efficacy. The dose-segmentation strategy can facilitate the extraction of more profound information. This indicates that ScoreDLR and ScoreMDR were negatively associated with XM, whereas ScoreDLD, derived from SD exceeding 15 Gy, displayed a positive association with XM. For feature extraction, deep learning was superior to manual definition.

Data Science Opportunities To Improve Radiotherapy Planning and Clinical Decision Making

Radiotherapy aims to achieve a high tumor control probability while minimizing damage to normal tissues. Personalizing radiotherapy treatments for individual patients, therefore, depends on integrating physical treatment planning with predictive models of tumor control and normal tissue complications. Predictive models could be improved using a wide range of rich data sources, including tumor and normal tissue genomics, radiomics, and dosiomics. Deep learning will drive improvements in classifying normal tissue tolerance, predicting intra-treatment tumor changes, tracking accumulated dose distributions, and quantifying the tumor response to radiotherapy based on imaging. Mechanistic patient-specific computer simulations ('digital twins') could also be used to guide adaptive radiotherapy. Overall, we are entering an era where improved modeling methods will allow the use of newly available data sources to better guide radiotherapy treatments.

Multi-omics deep learning for radiation pneumonitis prediction in lung cancer patients underwent volumetric modulated arc therapy

BACKGROUND AND OBJECTIVE

To evaluate the feasibility and accuracy of radiomics, dosiomics, and deep learning (DL) in predicting Radiation Pneumonitis (RP) in lung cancer patients underwent volumetric modulated arc therapy (VMAT) to improve radiotherapy safety and management.

METHODS

Total of 318 and 31 lung cancer patients underwent VMAT from First Affiliated Hospital of Wenzhou Medical University (WMU) and Quzhou Affiliated Hospital of WMU were enrolled for training and external validation, respectively. Models based on radiomics (R), dosiomics (D), and combined radiomics and dosiomics features (R+D) were constructed and validated using three machine learning (ML) methods. DL models trained with CT (DLR), dose distribution (DLD), and combined CT and dose distribution (DL(R+D)) images were constructed. DL features were then extracted from the fully connected layers of the best-performing DL model to combine with features of the ML model with the best performance to construct models of R+DLR, D+DLD, R+D+DL(R+D)) for RP prediction.

RESULTS

The R+D model achieved a best area under curve (AUC) of 0.84, 0.73, and 0.73 in the internal validation cohorts with Support Vector Machine (SVM), XGBoost, and Logistic Regression (LR), respectively. The DL(R+D) model achieved a best AUC of 0.89 and 0.86 using ResNet-34 in training and internal validation cohorts, respectively. The R+D+DL(R+D) model achieved a best performance in the external validation cohorts with an AUC, accuracy, sensitivity, and specificity of 0.81(0.62-0.99), 0.81, 0.84, and 0.67, respectively.

CONCLUSIONS

The integration of radiomics, dosiomics, and DL features is feasible and accurate for the RP prediction to improve the management of lung cancer patients underwent VMAT.

CT-based different regions of interest radiomics analysis for acute radiation pneumonitis in patients with locally advanced NSCLC after chemoradiotherapy

Purpose

To establish a radiomics model using radiomics features from different region of interests (ROI) based on dosimetry-related regions in enhanced computed tomography (CT) simulated images to predict radiation pneumonitis (RP) in patients with non-small cell lung cancer (NSCLC).

Methods

Our retrospective study was conducted based on a cohort of 236 NSCLC patients (59 of them with RP≥2) who were treated in 2 institutions and divided into the primary cohort (n = 182,46 of them with RP≥2) and external validation cohort (n = 54,13 of them with RP≥2). Radiomic features extracted from three ROIs were defined as the whole lung (WL), the dose volume histogram (DVH) of the lung V20 (V20_Lung) and the DVH of the V30 of lung minus the planning target volume (PTV) (V30 Lung-PTV). A total of 107 radiomics features were extracted from each ROIs. The U test, correlation coefficient and least absolute shrinkage and selection operator (LASSO) were performed for features selection. Six models based on different classification algorithms were developed to select the best radiomics model (R model).In addition, we built a dosimetry model then combined it with the best R model to create a mixed model (R+D model) The receiver operating characteristic (ROC) curve was delineated to assess the predictive efficacy of the models. Decision curve analysis could benefit from the model proposals through the assessment of clinical utility.

Results

Among the three ROIs, the best R model constructed from the LightGBM algorithm demonstrated the strongest discriminative ability in the ROI of V30 Lung-PTV. The corresponding area under the curve (AUC) value was 0.930 (95 % confidence interval (CI): 0.829-0.941). The D model, R model and R+D model achieved AUC values of 0.798 (95 %CI: 0.732-0.865), 0.930 (95 %CI: 0.829-0.941) and 0.940 (95 %CI: 0.906-0.974) in primary cohort, and in external validation cohort, the AUC values were 0.793 (95 %CI:0.637-0.949), 0.887 (95 %CI:0.810-0.993), 0.951 (95CI%:0.891-1.000). Decision curve demonstrate that R+D model could benefit for patients through the assessment of clinical utility.

Conclusion

The radiomics model was able to predict the acute RP more effectively in comparison with the traditional dosimetry model. Especially the radiomics model based on the V30 Lung-PTV region was able to achieve a higher accuracy when compared to the other regions.

A novel predictor for dosimetry data of lung and the radiation pneumonitis incidence prior to SBRT in lung cancer patients

Normal tissue complication probability (NTCP) models for radiation pneumonitis (RP) in lung cancer patients with stereotactic body radiation therapy (SBRT), which based on dosimetric data from treatment planning, are limited to patients who have already received radiation therapy (RT). This study aims to identify a novel predictive factor for lung dose distribution and RP probability before devising actionable SBRT plans for lung cancer patients. A comprehensive correlation analysis was performed on the clinical and dose parameters of lung cancer patients who underwent SBRT. Linear regression models were utilized to analyze the dosimetric data of lungs. The performance of the regression models was evaluated using mean squared error (MSE) and the coefficient of determination (R2). Correlational analysis revealed that most clinical data exhibited weak correlations with dosimetric data. However, nearly all dosimetric variables showed "strong" or "very strong" correlations with each other, particularly concerning the mean dose of the ipsilateral lung (MI) and the other dosimetric parameters. Further study verified that the lung tumor ratio (LTR) was a significant predictor for MI, which could predict the incidence of RP. As a result, LTR can predict the probability of RP without the need to design an elaborate treatment plan. This study, as the first to offer a comprehensive correlation analysis of dose parameters, explored the specific relationships among them. Significantly, it identified LTR as a novel predictor for both dose parameters and the incidence of RP, without the need to design an elaborate treatment plan.

Radiomics for clinical decision support in radiation oncology

Radiomics is a promising tool for the development of quantitative biomarkers to support clinical decision-making. It has been shown to improve the prediction of response to treatment and outcome in different settings, particularly in the field of radiation oncology by optimising the dose delivery solutions and reducing the rate of radiation-induced side effects, leading to a fully personalised approach. Despite the promising results offered by radiomics at each of these stages, standardised methodologies, reproducibility and interpretability of results are still lacking, limiting the potential clinical impact of these tools. In this review, we briefly describe the principles of radiomics and the most relevant applications of radiomics at each stage of cancer management in the framework of radiation oncology. Furthermore, the integration of radiomics into clinical decision support systems is analysed, defining the challenges and offering possible solutions for translating radiomics into a clinically applicable tool.

Radiomics-based hybrid model for predicting radiation pneumonitis: A systematic review and meta-analysis

PURPOSE

This study reviewed and meta-analyzed evidence on radiomics-based hybrid models for predicting radiation pneumonitis (RP). These models are crucial for improving thoracic radiotherapy plans and mitigating RP, a common complication of thoracic radiotherapy. We examined and compared the RP prediction models developed in these studies with the radiomics features employed in RP models.

METHODS

We systematically searched Google Scholar, Embase, PubMed, and MEDLINE for studies published up to April 19, 2024. Sixteen studies met the inclusion criteria. We compared the RP prediction models developed in these studies and the radiomics features employed.

RESULTS

Radiomics, as a single-factor evaluation, achieved an area under the receiver operating characteristic curve (AUROC) of 0.73, accuracy of 0.69, sensitivity of 0.64, and specificity of 0.74. Dosiomics achieved an AUROC of 0.70. Clinical and dosimetric factors showed lower performance, with AUROCs of 0.59 and 0.58. Combining clinical and radiomic factors yielded an AUROC of 0.78, while combining dosiomic and radiomics factors produced an AUROC of 0.81. Triple combinations, including clinical, dosimetric, and radiomics factors, achieved an AUROC of 0.81. The study identifies key radiomics features, such as the Gray Level Co-occurrence Matrix (GLCM) and Gray Level Size Zone Matrix (GLSZM), which enhance the predictive accuracy of RP models.

CONCLUSIONS

Radiomics-based hybrid models are highly effective in predicting RP. These models, combining traditional predictive factors with radiomic features, particularly GLCM and GLSZM, offer a clinically feasible approach for identifying patients at higher RP risk. This approach enhances clinical outcomes and improves patient quality of life.

PROTOCOL REGISTRATION

The protocol of this study was registered on PROSPERO (CRD42023426565).

Establishing a 4D-CT lung function related volumetric dose model to reduce radiation pneumonia

In order to study how to use pulmonary functional imaging obtained through 4D-CT fusion for radiotherapy planning, and transform traditional dose volume parameters into functional dose volume parameters, a functional dose volume parameter model that may reduce level 2 and above radiation pneumonia was obtained. 41 pulmonary tumor patients who underwent 4D-CT in our department from 2020 to 2023 were included. MIM Software (MIM 7.0.7; MIM Software Inc., Cleveland, OH, USA) was used to register adjacent phase CT images in the 4D-CT series. The three-dimensional displacement vector of CT pixels was obtained when changing from one respiratory state to another respiratory state, and this three-dimensional vector was quantitatively analyzed. Thus, a color schematic diagram reflecting the degree of changes in lung CT pixels during the breathing process, namely the distribution of ventilation function strength, is obtained. Finally, this diagram is fused with the localization CT image. Select areas with Jacobi > 1.2 as high lung function areas and outline them as fLung. Import the patient's DVH image again, fuse the lung ventilation image with the localization CT image, and obtain the volume of fLung different doses (V60, V55, V50, V45, V40, V35, V30, V25, V20, V15, V10, V5). Analyze the functional dose volume parameters related to the risk of level 2 and above radiation pneumonia using R language and create a predictive model. By using stepwise regression and optimal subset method to screen for independent variables V35, V30, V25, V20, V15, and V10, the prediction formula was obtained as follows: Risk = 0.23656-0.13784 * V35 + 0.37445 * V30-0.38317 * V25 + 0.21341 * V20-0.10209 * V15 + 0.03815 * V10. These six independent variables were analyzed using a column chart, and a calibration curve was drawn using the calibrate function. It was found that the Bias corrected line and the Apparent line were very close to the Ideal line, The consistency between the predicted value and the actual value is very good. By using the ROC function to plot the ROC curve and calculating the area under the curve: 0.8475, 95% CI 0.7237-0.9713, it can also be determined that the accuracy of the model is very high. In addition, we also used Lasso method and random forest method to filter out independent variables with different results, but the calibration curve drawn by the calibration function confirmed poor prediction performance. The function dose volume parameters V35, V30, V25, V20, V15, and V10 obtained through 4D-CT are key factors affecting radiation pneumonia. Establishing a predictive model can provide more accurate lung restriction basis for clinical radiotherapy planning.

Dosiomics and radiomics-based prediction of pneumonitis after radiotherapy and immune checkpoint inhibition: The relevance of fractionation

OBJECTIVES

Post-therapy pneumonitis (PTP) is a relevant side effect of thoracic radiotherapy and immunotherapy with checkpoint inhibitors (ICI). The influence of the combination of both, including dose fractionation schemes on PTP development is still unclear. This study aims to improve the PTP risk estimation after radio(chemo)therapy (R(C)T) for lung cancer with and without ICI by investigation of the impact of dose fractionation on machine learning (ML)-based prediction.

MATERIALS AND METHODS

Data from 100 patients who received fractionated R(C)T were collected. 39 patients received additional ICI therapy. Computed Tomography (CT), RT segmentation and dose data were extracted and physical doses were converted to 2-Gy equivalent doses (EQD2) to account for different fractionation schemes. Features were reduced using Pearson intercorrelation and the Boruta algorithm within 1000-fold bootstrapping. Six single (clinics, Dose Volume Histogram (DVH), ICI, chemotherapy, radiomics, dosiomics) and four combined models (radiomics + dosiomics, radiomics + DVH + Clinics, dosiomics + DVH + Clinics, radiomics + dosiomics + DVH + Clinics) were trained to predict PTP. Dose-based models were tested using physical dose and EQD2. Four ML-algorithms (random forest (rf), logistic elastic net regression, support vector machine, logitBoost) were trained and tested using 5-fold nested cross validation and Synthetic Minority Oversampling Technique (SMOTE) for resampling in R. Prediction was evaluated using the area under the receiver operating characteristic curve (AUC) on the test sets of the outer folds.

RESULTS

The combined model of all features using EQD2 surpassed all other models (AUC = 0.77, Confidence Interval CI 0.76-0.78). DVH, clinical data and ICI therapy had minor impact on PTP prediction with AUC values between 0.42 and 0.57. All EQD2-based models outperformed models based on physical dose.

CONCLUSIONS

Radiomics + dosiomics based ML models combined with clinical and dosimetric models were found to be suited best for PTP prediction after R(C)T and could improve pre-treatment decision making. Different RT dose fractionation schemes should be considered for dose-based ML approaches.

A dynamic nomogram predicting symptomatic pneumonia in patients with lung cancer receiving thoracic radiation

PURPOSE

The most common and potentially fatal side effect of thoracic radiation therapy is radiation pneumonitis (RP). Due to the lack of effective treatments, predicting radiation pneumonitis is crucial. This study aimed to develop a dynamic nomogram to accurately predict symptomatic pneumonitis (RP ≥ 2) following thoracic radiotherapy for lung cancer patients.

METHODS

Data from patients with pathologically diagnosed lung cancer at the Zhongshan People's Hospital Department of Radiotherapy for Thoracic Cancer between January 2017 and June 2022 were retrospectively analyzed. Risk factors for radiation pneumonitis were identified through multivariate logistic regression analysis and utilized to construct a dynamic nomogram. The predictive performance of the nomogram was validated using a bootstrapped concordance index and calibration plots.

RESULTS

Age, smoking index, chemotherapy, and whole lung V5/MLD were identified as significant factors contributing to the accurate prediction of symptomatic pneumonitis. A dynamic nomogram for symptomatic pneumonitis was developed using these risk factors. The area under the curve was 0.89(95% confidence interval 0.83-0.95). The nomogram demonstrated a concordance index of 0.89(95% confidence interval 0.82-0.95) and was well calibrated. Furthermore, the threshold values for high- risk and low- risk were determined to be 154 using the receiver operating curve.

CONCLUSIONS

The developed dynamic nomogram offers an accurate and convenient tool for clinical application in predicting the risk of symptomatic pneumonitis in patients with lung cancer undergoing thoracic radiation.

Identification of high-risk population of pneumoconiosis using deep learning segmentation of lung 3D images and radiomics texture analysis

OBJECTION

The aim of this study is to develop an early-warning model for identifying high-risk populations of pneumoconiosis by combining lung 3D images and radiomics lung texture features.

METHODS

A retrospective study was conducted, including 600 dust-exposed workers and 300 confirmed pneumoconiosis patients. Chest computed tomography (CT) images were divided into a training set and a test set in a 2:1 ratio. Whole-lung segmentation was performed using deep learning models for feature extraction of radiomics. Two feature selection algorithms and five classification models were used. The optimal model was selected using a 10-fold cross-validation strategy, and the calibration curve and decision curve were evaluated. To verify the applicability of the model, the diagnostic efficiency and accuracy between the model and human interpretation were compared. Additionally, the risk probabilities for different risk groups defined by the model were compared at different time intervals.

RESULTS

Four radiomics features were ultimately used to construct the predictive model. The logistic regression model was the most stable in both the training set and testing set, with an area under curve (AUC) of 0.964 (95 % confidence interval [CI], 0.950-0.976) and 0.947 (95 %CI, 0.925-0.964). In the training and testing sets, the Brier scores were 0.092 and 0.14, respectively, with threshold probability ranges of 2 %-99 % and 2 %-85 %. These results indicate that the model exhibits good calibration and clinical benefit. The comparison between the model and human interpretation showed that the model was not inferior in terms of diagnostic efficiency and accuracy. Additionally, the high-risk population identified by the model was diagnosed as pneumoconiosis two years later.

CONCLUSION

This study provides a meticulous and quantifiable method for detecting and assessing the risk of pneumoconiosis, building upon accurate diagnosis. Employing risk scoring and probability estimation, not only enhances the efficiency of diagnostic physicians but also provides a valuable reference for controlling the occurrence of pneumoconiosis.

Exploring non-invasive precision treatment in non-small cell lung cancer patients through deep learning radiomics across imaging features and molecular phenotypes

BACKGROUND

Accurate prediction of tumor molecular alterations is vital for optimizing cancer treatment. Traditional tissue-based approaches encounter limitations due to invasiveness, heterogeneity, and molecular dynamic changes. We aim to develop and validate a deep learning radiomics framework to obtain imaging features that reflect various molecular changes, aiding first-line treatment decisions for cancer patients.

METHODS

We conducted a retrospective study involving 508 NSCLC patients from three institutions, incorporating CT images and clinicopathologic data. Two radiomic scores and a deep network feature were constructed on three data sources in the 3D tumor region. Using these features, we developed and validated the 'Deep-RadScore,' a deep learning radiomics model to predict prognostic factors, gene mutations, and immune molecule expression levels.

FINDINGS

The Deep-RadScore exhibits strong discrimination for tumor molecular features. In the independent test cohort, it achieved impressive AUCs: 0.889 for lymphovascular invasion, 0.903 for pleural invasion, 0.894 for T staging; 0.884 for EGFR and ALK, 0.896 for KRAS and PIK3CA, 0.889 for TP53, 0.895 for ROS1; and 0.893 for PD-1/PD-L1. Fusing features yielded optimal predictive power, surpassing any single imaging feature. Correlation and interpretability analyses confirmed the effectiveness of customized deep network features in capturing additional imaging phenotypes beyond known radiomic features.

INTERPRETATION

This proof-of-concept framework demonstrates that new biomarkers across imaging features and molecular phenotypes can be provided by fusing radiomic features and deep network features from multiple data sources. This holds the potential to offer valuable insights for radiological phenotyping in characterizing diverse tumor molecular alterations, thereby advancing the pursuit of non-invasive personalized treatment for NSCLC patients.

Deep-Learning Model Prediction of Radiation Pneumonitis Using Pretreatment Chest Computed Tomography and Clinical Factors

Objectives: This study aimed to build a comprehensive deep-learning model for the prediction of radiation pneumonitis using chest computed tomography (CT), clinical, dosimetric, and laboratory data. Introduction: Radiation therapy is an effective tool for treating patients with lung cancer. Despite its effectiveness, the risk of radiation pneumonitis limits its application. Although several studies have demonstrated models to predict radiation pneumonitis, no reliable model has been developed yet. Herein, we developed prediction models using pretreatment chest CT and various clinical data to assess the likelihood of radiation pneumonitis in lung cancer patients. Methods: This retrospective study analyzed 3-dimensional (3D) lung volume data from chest CT scans and 27 features including dosimetric, clinical, and laboratory data from 548 patients who were treated at our institution between 2010 and 2021. We developed a neural network, named MergeNet, which processes lung 3D CT, clinical, dosimetric, and laboratory data. The MergeNet integrates a convolutional neural network with subsequent fully connected layers. A support vector machine (SVM) and light gradient boosting machine (LGBM) model were also implemented for comparison. For comparison, the convolution-only neural network was implemented as well. Three-dimensional Resnet-10 network and 4-fold cross-validation were used. Results: Classification performance was quantified by using the area under the receiver operative characteristic curve (AUC) metrics. MergeNet showed the AUC of 0.689. SVM, LGBM, and convolution-only networks showed AUCs of 0.525, 0.541, and 0.550, respectively. Application of DeLong test to pairs of receiver operating characteristic curves respectively yielded P values of .001 for the MergeNet-SVM pair and 0.001 for the MergeNet-LGBM pair. Conclusion: The MergeNet model, which incorporates chest CT, clinical, dosimetric, and laboratory data, demonstrated superior performance compared to other models. However, since its prediction performance has not yet reached an efficient level for clinical application, further research is required. Contribution: This study showed that MergeNet may be an effective means to predict radiation pneumonitis. Various predictive factors can be used together for the radiation pneumonitis prediction task via the MergeNet.

Dosiomics and radiomics improve the prediction of post-radiotherapy neutrophil-lymphocyte ratio in locally advanced non-small cell lung cancer

PURPOSE

To develop and validate a dosiomics and radiomics model based on three-dimensional (3D) dose distribution map and computed tomography (CT) images for the prediction of the post-radiotherapy (post-RT) neutrophil-to-lymphocyte ratio (NLR).

METHODS

This work retrospectively collected 242 locally advanced non-small cell lung cancer (LA-NSCLC) patients who were treated with definitive radiotherapy from 2012 to 2016. The NLR collected one month after the completion of RT was defined as the primary outcome. Clinical characteristics and two-dimensional dosimetric factors calculated from the dose-volume histogram (DVH) were included. A total of 4165 dosiomics and radiomics features were extracted from the 3D dose maps and CT images within five different anatomical regions of interest (ROIs), respectively. Then, a three-step feature selection method was proposed to progressively filter features from coarse to fine: (i) model-based ranking according to individual feature's performance, (ii) maximum relevance and minimum redundancy (mRMR), (iii) select from model based on feature importance calculated with an ensemble of several decision trees. The selected feature subsets were utilized to develop the prediction model with GBDT. All patients were divided into a development set and an independent testing set (2:1). Five-fold cross-validation was applied to the development set for both feature selection and model training procedure. Finally, a fusion model combining dosiomics, radiomics and clinical features was constructed to further improve the prediction results. The area under receiver operating characteristic curve (ROC) were used to evaluate the model performance.

RESULTS

The clinical-based and DVH-based models showed limited predictive power with AUCs of 0.632 (95% CI: 0.490-0.773) and 0.634 (95% CI: 0.497-0.771), respectively, in the independent testing set. The 9 feature-based dosiomics and 3 feature-based radiomics models showed improved AUCs of 0.738 (95% CI: 0.628-0.849) and 0.689 (95% CI: 0.566-0.813), respectively. The dosiomics & radiomics & clinical fusion model further improved the model's generalization ability with an AUC of 0.765 (95% CI: 0.656-0.874).

CONCLUSIONS

Dosiomics and radiomics can benefit the prediction of post-RT NLR of LA-NSCLC patients. This can provide a reference for evaluating radiotherapy-related inflammation.

The Association of Gross Tumor Volume and Its Radiomics Features with Brain Metastases Development in Patients with Radically Treated Stage III Non-Small Cell Lung Cancer

PURPOSE

To identify clinical risk factors, including gross tumor volume (GTV) and radiomics features, for developing brain metastases (BM) in patients with radically treated stage III non-small cell lung cancer (NSCLC).

METHODS

Clinical data and planning CT scans for thoracic radiotherapy were retrieved from patients with radically treated stage III NSCLC. Radiomics features were extracted from the GTV, primary lung tumor (GTVp), and involved lymph nodes (GTVn), separately. Competing risk analysis was used to develop models (clinical, radiomics, and combined model). LASSO regression was performed to select radiomics features and train models. Area under the receiver operating characteristic curves (AUC-ROC) and calibration were performed to assess the models' performance.

RESULTS

Three-hundred-ten patients were eligible and 52 (16.8%) developed BM. Three clinical variables (age, NSCLC subtype, and GTVn) and five radiomics features from each radiomics model were significantly associated with BM. Radiomic features measuring tumor heterogeneity were the most relevant. The AUCs and calibration curves of the models showed that the GTVn radiomics model had the best performance (AUC: 0.74; 95% CI: 0.71-0.86; sensitivity: 84%; specificity: 61%; positive predictive value [PPV]: 29%; negative predictive value [NPV]: 95%; accuracy: 65%).

CONCLUSION

Age, NSCLC subtype, and GTVn were significant risk factors for BM. GTVn radiomics features provided higher predictive value than GTVp and GTV for BM development. GTVp and GTVn should be separated in clinical and research practice.

Computed tomography and radiation dose images-based deep-learning model for predicting radiation pneumonitis in lung cancer patients after radiation therapy

PURPOSE

To develop a deep learning model that combines CT and radiation dose (RD) images to predict the occurrence of radiation pneumonitis (RP) in lung cancer patients who received radical (chemo)radiotherapy.

METHODS

CT, RD images and clinical parameters were obtained from 314 retrospectively-collected patients (training set) and 35 prospectively-collected patients (test-set-1) who were diagnosed with lung cancer and received radical radiotherapy in the dose range of 50 Gy and 70 Gy. Another 194 (60 Gy group, test-set-2) and 158 (74 Gy group, test-set-3) patients from the clinical trial RTOG 0617 were used for external validation. A ResNet architecture was used to develop a prediction model that combines CT and RD features. Thereafter, the CT and RD weights were adjusted by using 40 patients from test-set-2 or 3 to accommodate cohorts with different clinical settings or dose delivery patterns. Visual interpretation was implemented using a gradient-weighted class activation map (grad-CAM) to observe the area of model attention during the prediction process. To improve the usability, ready-to-use online software was developed.

RESULTS

The discriminative ability of a baseline trained model had an AUC of 0.83 for test-set-1, 0.55 for test-set-2, and 0.63 for test-set-3. After adjusting CT and RD weights of the model using a subset of the RTOG-0617 subjects, the discriminatory power of test-set-2 and 3 improved to AUC 0.65 and AUC 0.70, respectively. Grad-CAM showed the regions of interest to the model that contribute to the prediction of RP.

CONCLUSION

A novel deep learning approach combining CT and RD images can effectively and accurately predict the occurrence of RP, and this model can be adjusted easily to fit new cohorts.