Intratumoral and peritumoral radiomics nomograms for the preoperative prediction of lymphovascular invasion and overall survival in non-small cell lung cancer

CT-based intratumoral and peritumoral radiomics nomogram to predict spread through air spaces in lung adenocarcinoma with diameter ≤ 3 cm: A multicenter study

Purpose

The aim of this study was to explore and develop a preoperative and noninvasive model for predicting spread through air spaces (STAS) status in lung adenocarcinoma (LUAD) with diameter ≤ 3 cm.

Methods

This multicenter retrospective study included 640 LUAD patients. Center I included 525 patients (368 in the training cohort and 157 in the validation cohort); center II included 115 patients (the test cohort). We extracted radiomics features from the intratumor, extended tumor and peritumor regions. Multivariate logistic regression and boruta algorithm were used to select clinical independent risk factors and radiomics features, respectively. We developed a clinical model and four radiomics models (the intratumor model, extended tumor model, peritumor model and fusion model). A nomogram based on prediction probability value of the optimal radiomics model and clinical independent risk factors was developed to predict STAS status.

Results

Maximum diameter and nodule type were clinical independent risk factors. The extended tumor model achieved satisfactory STAS status discrimination performance with the AUC of 0.74, 0.71 and 0.80 in the three cohorts, respectively, performed better than other radiomics models. The integrated discrimination improvement value revealed that the nomogram outperformed compared to the clinical model with the value of 12 %. Patients with high nomogram score (≥ 77.31) will be identified as STAS-positive.

Conclusions

Peritumoral information is significant to predict STAS status. The nomogram based on the extended tumor model and clinical independent risk factors provided good preoperative prediction of STAS status in LUAD with diameter ≤ 3 cm, aiding surgical decision-making.

Explainable PET-based intratumoral and peritumoral machine learning model for predicting visceral pleural invasion in clinical-stage IA non-small cell lung cancer: A two-center study

AIM

The aim of this study was to develop a PET-based machine learning model for predicting visceral pleural invasion (VPI) in patients with clinical stage IA non-small cell lung cancer.

MATERIALS AND METHODS

A total of 294 patients and 69 patients from two institutions who underwent the 18F-FDG-PET scan were retrospectively analyzed. We extracted PET-based radiomics features from the gross tumor volume (GTV) and gross tumor volume incorporating peritumoral 4, 8 and 12 mm regions (GPTV4, GPTV8, GPTV12), respectively. Then four models were respectively established by using machine learning algorithms. The performance of the models was assessed by the receiver operating characteristic (ROC) curve and decision curve analyses (DCA). Shapley additive explanation (SHAP) was employed to explain the machine learning (ML) models and visualize variable predictions.

RESULTS

Compared with GTV, GPTV4, and GPTV12 radiomics models, the radiomics model based on GPTV8 using random forest (RF) among the 10 features demonstrated better prediction performance, with the AUC of 0.879, 0.846, and 0.745 in the training, internal validation, and external validation sets, respectively. The results of the SHAP method showed that the GLRLM_ShortRunLowGreyLevel Emphasis features were the most important factors in VPI. At the patient level, SHAP force plots provided a deep understanding for predicting VPI.

CONCLUSION

The PET-based intratumoral and peritumoral model based on machine learning offers an innovative tool for preoperative prediction of VPI in patients with lung adenocarcinoma. By employing the SHAP method, clinicians may gain a clearer insight into the factors contributing to VPI, which could enhance clinical decision-making of prognosis assessment.

Pericardial Fat and Primary Tumor Radiomics for Predicting Occult N2 Disease and Survival in Clinical Stage I Non-Small Cell Lung Cancer: Multicenter Study With Biologic Correlation

Background: Occult N2 disease significantly affects clinical stage I non-small cell lung cancer (NSCLC) prognosis. Pericardial fat characteristics also have prognostic associations. Objective: To develop and test a model incorporating pericardial fat and tumor radiomic features on CT for detecting occult N2 disease in clinical stage I NSCLC, explore the model's prognostic role, and investigate its biologic basis through radiogenomics analyses. Methods: This retrospective study included patients who underwent clinical stage I NSCLC resection at three hospitals [center 1 (January 2016 to December 2022), stratified randomly by 6:2:2 ratio into training, tuning, and internal test sets; centers 2 and 3 (January 2019 to December 2023), serving as external test sets]. Pericardial fat and primary tumors were segmented on preoperative CT to extract radiomic features and generate tumor and fat rad-scores, respectively. Multivariable analysis was performed to create a hybrid model for predicting occult N2 disease at surgery. Performance was evaluated in external test sets. Associations with recurrence-free survival (RFS) and overall survival (OS) were evaluated using log-rank tests in the internal test set; follow-up data were unavailable in external test sets. Biologic mechanisms were explored through RNA and gene expression analysis in a separate set of patients with NSCLC obtained from a public radiogenomics database. Results: From the three centers, 1662 patients (mean age, 58.6 years; 663 men, 999 women) were included. Following multivariable analysis, the hybrid model included nodule density, fat rad-score, and tumor rad-score. The model had AUC, accuracy, sensitivity, and specificity for occult N2 disease of 0.921, 89.7%, 59.3%, and 93.1%, and 0.913, 91.8%, 56.2%, and 95.5% in external test sets 1 and 2, respectively. High-risk compared with low-risk patients, applying the model in the internal test set, showed worse RFS (p<.001) and OS (p<.001). In 122 patients in radiogenomics analysis, high-risk status was associated with activation of molecular pathways and increased activated dendritic cell and mast cell infiltration. Conclusion: A model incorporating tumor and pericardial fat radiomics showed good performance in predicting occult N2 disease as well as associations with survival and with RNA and gene expression. Clinical Impact: The model could help guide NSCLC management.

Deep learning feature-based model for predicting lymphovascular invasion in urothelial carcinoma of bladder using CT images

OBJECTIVES

Lymphovascular invasion significantly impacts the prognosis of urothelial carcinoma of the bladder. Traditional lymphovascular invasion detection methods are time-consuming and costly. This study aims to develop a deep learning-based model to preoperatively predict lymphovascular invasion status in urothelial carcinoma of bladder using CT images.

METHODS

Data and CT images of 577 patients across four medical centers were retrospectively collected. The largest tumor slices from the transverse, coronal, and sagittal planes were selected and used to train CNN models (InceptionV3, DenseNet121, ResNet18, ResNet34, ResNet50, and VGG11). Deep learning features were extracted and visualized using Grad-CAM. Principal Component Analysis reduced features to 64. Using the extracted features, Decision Tree, XGBoost, and LightGBM models were trained with 5-fold cross-validation and ensembled in a stacking model. Clinical risk factors were identified through logistic regression analyses and combined with DL scores to enhance lymphovascular invasion prediction accuracy.

RESULTS

The ResNet50-based model achieved an AUC of 0.818 in the validation set and 0.708 in the testing set. The combined model showed an AUC of 0.794 in the validation set and 0.767 in the testing set, demonstrating robust performance across diverse data.

CONCLUSION

We developed a robust radiomics model based on deep learning features from CT images to preoperatively predict lymphovascular invasion status in urothelial carcinoma of the bladder. This model offers a non-invasive, cost-effective tool to assist clinicians in personalized treatment planning.

CRITICAL RELEVANCE STATEMENT

We developed a robust radiomics model based on deep learning features from CT images to preoperatively predict lymphovascular invasion status in urothelial carcinoma of the bladder.

KEY POINTS

We developed a deep learning feature-based stacking model to predict lymphovascular invasion in urothelial carcinoma of the bladder patients using CT. Max cross sections from three dimensions of the CT image are used to train the CNN model. We made comparisons across six CNN networks, including ResNet50.

Intratumoral and peritumoral radiomics model for the preoperative prediction of cribriform component in invasive lung adenocarcinoma: a multicenter study

PURPOSE

This study aimed to investigate the predictive value of intratumoral and peritumoral radiomics model for the cribriform component (CC) of invasive lung adenocarcinoma (LUAD).

MATERIALS AND METHODS

The 144 patients with invasive LUAD from our center were randomly divided into training set (n = 100) and internal validation set (n = 44) in a ratio of 7:3, and 75 patients from center 2 were regarded as the external validation set. Clinical risk factors were examined using univariate and multivariate logistic regression to construct the clinical model. We extracted radiomics features from gross tumor volume (GTV), gross and peritumoral volume (GPTV), and peritumoral volume (PTV), respectively. Radiomics models were constructed with selected features. A combined model based on the optimal Radscore and clinically independent predictors was constructed, and its predictive performance was assessed by receiver operating characteristic curve (ROC), calibration curve, and decision curve analysis (DCA).

RESULTS

The area under curves (AUCs) of the GTV model were 0.882 (95% CI 0.817-0.948), 0.794 (95% CI 0.656-0.932), and 0.766 (95% CI 0.657-0.875) in the training, internal validation, and external validation sets, and the PTV model had AUCs of 0.812 (95% CI 0.725-0.899), 0.749 (95% CI 0.597-0.902), and 0.670 (95% CI 0.543-0.798) in the training, internal validation, and external validation sets, respectively. However, the GPTV radiomics model showed better predictive performance compared with the GTV and PTV radiomics models, with the AUCs of 0.950 (95% CI 0.911-0.989), 0.844 (95% CI 0.728-0.959), and 0.815 (95% CI 0.713-0.917) in the training, internal validation and external validation sets, respectively. In the clinical model, tumor shape, lobulation sign and maximal diameter were the independent predictors of CC in invasive LUAD. The combined model including independent clinical predictors and GPTV-Radscore show the considerable instructive to clinical practice, with the AUCs of 0.954(95% CI 0.918-0.990), 0.861(95% CI 0.752-0.970), and 0.794(95% CI 0.690-0.898) in training, internal validation, and external validation sets, respectively. DCA showed that the combined model had good clinical value and correction effect.

CONCLUSION

Radiomics model is a very powerful tool for predicting CC growth pattern in invasive LUAD and can help clinicians make the strategies of treatment and surveillance in patients with invasive LUAD.

Development of an Intra- and Peritumoral Radiomics Nomogram Using Digital Breast Tomosynthesis for Preoperative Assessment of Ki-67 Expression in Invasive Breast Cancer

RATIONALE AND OBJECTIVES

This study aimed to develop a radiomics nomogram model using preoperative digital breast tomosynthesis (DBT) images to predict Ki-67 expression in patients with invasive breast cancer (IBC).

MATERIALS AND METHODS

This retrospective study involved a cohort of 289 patients with IBC, who were randomly divided into a training dataset (N= 202) and a validation dataset (N= 87). Ki-67 expression was categorized into low and high groups using a 14% threshold. Radiomics features from both the intra- and peritumoral regions of DBT images were used to develop the radiomics model, referred to as Radscore. Clinical and nomogram models were constructed using multivariate logistic regression. The performance of the established models was evaluated using receiver operating characteristic (ROC) curve analysis, calibration curve analysis, decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discrimination improvement (IDI).

RESULTS

The clinical model was constructed using tumor size and DBT-reported lymph node metastasis (DBT_reported_LNM). By integrating Radscore_Combine-which incorporates both intra- and peritumoral radiomics features-along with tumor size and DBT_reported_LNM into the nomogram, the model achieved the highest area under the curve (AUC) values of 0.819 and 0.755 in the training and validation datasets, respectively. The notable improvement shown by the NRI and IDI suggests that Radscore_Combine could serve as a valuable biomarker for predicting Ki-67 expression effectively.

CONCLUSION

The nomogram offers a non-invasive method to predict Ki-67 expression in IBC patients, which could aid in creating personalized treatment plans.

Habitat Radiomics and Deep Learning Features Based on CT for Predicting Lymphovascular Invasion in T1-stage Lung Adenocarcinoma: A Multicenter Study

RATIONALE AND OBJECTIVES

The research aims to examine how CT-derived habitat radiomics can be used to predict lymphovascular invasion (LVI) in patients with T1-stage lung adenocarcinoma (LUAD), and compare its effectiveness to traditional radiomics and deep learning (DL) models.

MATERIALS AND METHODS

We retrospectively analyzed 349 T1-stage LUAD patients from three centers from January 2021 to March 2024. The K-means algorithm was utilized to cluster CT images and apparent diffusion coefficient maps. Following features selection, we constructed three types of models, namely radiomics, habitat, and DL to identify patients with LVI. The evaluation of all models was conducted by employing the area under the receiver operating characteristic curve (AUC), calibration curves and decision curve analysis.

RESULTS

349 eligible patients were divided into an internal training set of 210 and an external test set of 139. We identified four distinct habitats, with the AUC for the overall habitat area outperforming that of the four sub-areas. Within the test set, the habitat model reached a higher AUC of 0.941 in contrast to the radiomics model at 0.918 and the deep learning model at 0.896.

CONCLUSION

CT-based habitat radiomics shows promise in predicting LVI in T1-stage LUAD patients, with the habitat signature demonstrating superior performance and significant advantages in identifying patients who are LVI-positive.

Predicting lymphovascular invasion in stage IA lung adenocarcinoma: a CT-based classification and regression tree model

BACKGROUND

Lymphovascular invasion (LVI) is a significant histopathological marker associated with poor prognosis in patients. However, there is a notable lack of reliable, non-invasive preoperative tools to predict LVI accurately.

PURPOSE

To develop and validate a computed tomography (CT)-based classification and regression tree (CART) model for the preoperative prediction of LVI in patients with clinical stage IA lung adenocarcinoma (LUAD).

MATERIALS AND METHODS

This multicenter cohort study recruited patients who underwent resection and had a preoperative CT examination. An internal cohort (n = 525) is included to construct the LVI classification and regression tree model (LVI-CART). An external cohort (n = 115) and a public cohort (n = 57) are then used to fully validate the predictive performance of the LVI-CART. Kaplan-Meier survival analysis and univariable Cox regression analyses were conducted to investigate the relationship between predicted LVI status and survival.

RESULTS

The LVI-CART model includes two features, diameter and nodule type, and shows acceptable performance in predicting pathological LVI, with area under the curve values of 0.719, 0.756, and 0.835 in the internal validation set, external validation set and test set, respectively. A predicted LVI positive relative to the median value in the outcomes cohort was found to be independently associated with 1-, 3-year RFS and 1-, 3-, 5-year OS (all p-values < 0.05).

CONCLUSIONS

The LVI-CART model could be used to preoperatively predict LVI and identify patients with poor prognosis in clinical IA LUAD. The model is like to be simple and easily applicable to risk stratification.

KEY POINTS

Question Lymphovascular invasion is a critical histopathological indicator of poor prognosis, necessitating reliable non-invasive preoperative predictive tools. Findings The classification and regression tree model for predicting lymphovascular invasion (LVI-CART) model demonstrates adequate predictive ability for pathological LVI, portending poor recurrence-free and overall survival. Clinical relevance The LVI-CART model provides clinicians with an easy-to-use method for preoperative identification of patients with clinical stage IA lung adenocarcinoma who are LVI-positive. It also provides a framework for a comprehensive assessment of patient survival risk.

Prediction of radiosensitivity in non-small cell lung cancer based on computed tomography and tumor genomics: a multiple real world cohort study

BACKGROUND

The varying degrees of radiotherapy sensitivity of tumors limit the efficacy of tumor radiotherapy. In this research, based on single cell sequence data we used radiomics to help identify and screen feature signatures to distinguish varying radiosensitivity in different regions of the target area of non-small cell lung cancer can provide a new pattern to assess sensitivity of radiotherapy and assist in clinical decision-making.

METHODS

This retrospective study included CT radiology data from 454 patients diagnosed with non-small cell lung cancer in multiple real-world cohorts prior to radiotherapy. The tumor primary target area was delineated on a training set (n = 154) and segmented to obtain a radiogenomic single signature. The radiogenomic signature LCDigital-RT, which can predict radiosensitivity, was developed by combining transcriptome sequencing signature radiosensitivity index and validated on two independent external validation sets (n = 74) and (n = 160). Besides, we also described the single-cell landscape of non-small cell lung cancer with different radiosensitivity, attempting to explain the potential biological mechanism at the single-cell level.

RESULTS

By constructing solely from the single radiomics feature signature, pre LCDigital-RT can effectively identify populations with differences in radiation sensitivity in non-small cell lung cancer, with AUCs of 0.759, 0.728 and 0.745 for the training and two external validation sets, respectively. However, LCDigital-RT has a greater advantage, with a training set AUC of 0.837, which has been well validated in the JXCH cohort (AUC = 0.789) and GDPH cohort (AUC = 0.791). With the help of LCDigital-RT, patients can be divided into radiation sensitive and radiation resistant groups, and there is a significant difference in the characteristics of primary tumor lesions between the two groups. We have also enriched the interpretability of our radiogenomic features in biology at the single-cell level, demonstrating their enormous value in clinical translational research.

CONCLUSIONS

We have developed an LCDigital RT prediction tool that will help predict populations at risk of radiation sensitivity differences. By visualizing the thermal map of the primary tumor area, we can assist in the development of radiotherapy plans, reduce the occurrence of radiation toxicity events, and improve radiotherapy efficacy. At the same time, it provides a reference basis for evaluating radiation sensitivity from imaging, genetics, and other aspects.

Quantification of Intratumoral Heterogeneity Based on Habitat Analysis for Preoperative Assessment of Lymphovascular Invasion in Colorectal Cancer

RATIONALE AND OBJECTIVES

Preoperative knowledge of the status of lymphovascular invasion (LVI) status in colorectal cancer (CRC) patients can provide valuable information for choosing appropriate treatment strategies. This study aimed to explore the value of heterogeneity features derived from the habitat analysis of 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) images in predicting LVI.

MATERIALS AND METHODS

Pretreatment 18F-FDG PET/computed tomography (CT) images from 177 patients diagnosed with CRC were retrospectively obtained (training cohort, n=106; validation cohort, n=71). Conventional radiomics features and habitat-derived tumor heterogeneity features were extracted from 18F-FDG PET scans. The output probabilities of the imaging-based random forest model were used to generate a radiomics score (Radscore) and intratumoral heterogeneity score (ITHscore). Multivariate logistic regression analysis was used to determine the independent risk factors for LVI. On this basis, four LVI status classification models were developed using (a) clinical variables (Clinical model), (b) tumor heterogeneity features (ITHscore model), (c) radiomics features (Radscore model), and (d) clinical variables, tumor heterogeneity features, and radiomics features (Combined model). The area under the curve (AUC) and decision curve analysis were used to evaluate model performance.

RESULTS

Among all of the variables, the PET/CT-reported lymph node status, ITHscore, and Radscore were retained as predictors related to the risk of LVI in CRC patients (P<0.05). The predictive effect of the ITHscore model (AUC: 0.712) was better than that of the Radscore model (AUC: 0.650) and Clinical model (AUC: 0.652) in the validation cohort. The Combined model achieved better classification effects and clinical usefulness, and the AUCs of the training and validation cohorts were 0.857 and 0.798, respectively. A nomogram of the Combined model was established, and the calibration plot was well fitted (P>0.05). In addition, the results of Spearman's rank correlation tests showed that there was no significant correlation between the ITHscore and Radscore (R=0.044, P=0.655 in the training cohort; R=0.067, P=0.580 in the validation cohort).

CONCLUSION

Our results showed that the ITHscore is a novel and stable quantitative indicator of LVI and is helpful for effectively facilitating the risk stratification of LVI in CRC patients after integrating clinical variables and radiomics features.

Intratumoral and peritumoral PET/CT-based radiomics for non-invasively and dynamically predicting immunotherapy response in NSCLC

BACKGROUND

We aimed to develop a machine learning model based on intratumoral and peritumoral 18F-FDG PET/CT radiomics to non-invasively and dynamically predict the response to immunotherapy in non-small cell lung cancer (NSCLC).

METHODS

This retrospective study included 296 NSCLC patients, including a training cohort (N = 183), a testing cohort (N = 78), and a TCIA radiogenomic cohort (N = 35). The extreme gradient boosting algorithm was employed to develop the radiomic models.

RESULTS

The COMB-Radscore, which was developed by combining radiomic features from PET, CT, and PET/CT images, had the most satisfactory predictive performance with AUC (ROC) 0.894 and 0.819 in the training and testing cohorts, respectively. Survival analysis has demonstrated that COMB-Radscore is an independent prognostic factor for progression-free survival and overall survival. Moreover, COMB-Radscore demonstrates excellent dynamic predictive performance, with an AUC (ROC) of 0.857, enabling the earlier detection of potential disease progression in patients compared to radiological evaluation solely relying on tumor size. Further radiogenomic analysis showed that the COMB-Radscore was associated with infiltration abundance and functional status of CD8 + T cells.

CONCLUSIONS

The radiomic model holds promise as a precise, personalized, and dynamic decision support tool for the treatment of NSCLC patients.

Intratumoral and peritumoral CT radiomics in predicting anaplastic lymphoma kinase mutations and survival in patients with lung adenocarcinoma: a multicenter study

BACKGROUND

To explore the value of intratumoral and peritumoral radiomics in preoperative prediction of anaplastic lymphoma kinase (ALK) mutation status and survival in patients with lung adenocarcinoma.

METHODS

We retrospectively collected data from 505 eligible patients with lung adenocarcinoma from four hospitals (training and external validation sets 1-3). The CT-based radiomics features were extracted separately from the gross tumor volume (GTV) and GTV incorporating peritumoral 3-, 6-, 9-, 12-, and 15-mm regions (GPTV3, GPTV6, GPTV9, GPTV12, and GPTV15), and screened the most relevant features to construct radiomics models to predict ALK (+). The combined model incorporated radiomics scores (Rad-scores) of the best radiomics model and clinical predictors was constructed. Performance was evaluated using receiver operating characteristic (ROC) analysis. Progression-free survival (PFS) outcomes were examined using the Cox proportional hazards model.

RESULTS

In the four sets, 21.19% (107/505) patients were ALK (+). The GPTV3 radiomics model using a support vector machine algorithm achieved the best predictive performance, with the highest average AUC of 0.811 in the validation sets. Clinical TNM stage and pleural indentation were independent predictors. The combined model incorporating the GPTV3-Rad-score and clinical predictors achieved higher performance than the clinical model alone in predicting ALK (+) in three validation sets [AUC: 0.855 (95% CI: 0.766-0.919) vs. 0.648 (95% CI: 0.543-0.745), P = 0.001; 0.882 (95% CI: 0.801-0.962) vs. 0.634 (95% CI: 0.548-0.714), P < 0.001; 0.810 (95% CI: 0.727-0.877) vs. 0.663 (95% CI: 0.570-0.748), P = 0.006]. The prediction score of the combined model could stratify PFS outcomes in patients receiving ALK-TKI therapy (HR: 0.37; 95% CI: 0.15-0.89; P = 0.026) and immunotherapy (HR: 2.49; 95% CI: 1.22-5.08; P = 0.012).

CONCLUSION

The presented combined model based on GPTV3 effectively mined tumor features to predict ALK mutation status and stratify PFS outcomes in patients with lung adenocarcinoma.

Multi-omics models for predicting prognosis in non-small cell lung cancer patients following chemotherapy and radiotherapy: A multi-center study

BACKGROUND AND PURPOSE

Quantifying tumor heterogeneity from various dimensions is crucial for precise treatment. This study aimed to develop and validate multi-omics models based on the computed tomography images, pathological images, dose and clinical information to predict treatment response and overall survival of non-small cell lung cancer (NSCLC) patients undergoing chemotherapy and radiotherapy.

MATERIALS AND METHODS

This retrospective study included 220 NSCLC patients from three centers. Following feature extraction and selection, single-omics and multi-omics models were built for treatment response and overall survival prediction. The performance of treatment response models was evaluated using the area under the curve (AUC) and box plots. For overall survival analysis, the model's evaluation included AUC, concordance index (C-index), Kaplan-Meier curves, and calibration curves. Shapley values were used to assess the contribution of different features to multi-omics models.

RESULTS

Multi-omics models consistently exhibited superior discriminative ability compared to single-omics models in predicting both treatment response and overall survival. For treatment response, the three all-modality models achieved AUC values of 0.87, 0.91, and 0.82 in the external validation set, respectively. In overall survival analysis, the three all-modality models demonstrated AUC values and C-index of 0.73/0.72, 0.80/0.77, 0.79/0.78 in the external validation set, respectively.

CONCLUSION

Multi-omics prediction models demonstrated superior predictive ability with robustness and interpretability. By predicting treatment response and overall survival in NSCLC patients, these models have the potential to assist clinician optimizing treatment plans, supporting individualized treatment strategies, improving the tumor control probability and prolonging the patients' survival.

A PET/CT-based 3D deep learning model for predicting spread through air spaces in stage I lung adenocarcinoma

PURPOSE

This study evaluates a three-dimensional (3D) deep learning (DL) model based on fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) for predicting the preoperative status of spread through air spaces (STAS) in patients with clinical stage I lung adenocarcinoma (LUAD).

METHODS

A retrospective analysis of 162 patients with stage I LUAD was conducted, splitting data into training and test sets (4:1). Six 3D DL models were developed, and the top-performing PET and CT models (ResNet50) were fused for optimal prediction. The model's clinical utility was assessed through a two-stage reader study.

RESULTS

The fused PET/CT model achieved an area under the curve (AUC) of 0.956 (95% CI 0.9230-0.9881) in the training set and 0.889 (95% CI 0.7624-1.0000) in the test set. Compared to three physicians, the model demonstrated superior sensitivity and specificity. After the artificial intelligence (AI) assistance's participation, the diagnostic accuracy of the physicians improved during their subsequent reading session.

CONCLUSION

Our DL model demonstrates potential as a resource to aid physicians in predicting STAS status and preoperative treatment planning for stage I LUAD, though prospective validation is required.

Explainable PET-Based Habitat and Peritumoral Machine Learning Model for Predicting Progression-free Survival in Clinical Stage IA Pure-Solid Non-small Cell Lung Cancer: A Two-center Study

RATIONALE AND OBJECTIVES

This study aimed to develop and validate machine learning (ML) models utilizing positron emission tomography (PET)-habitat of the tumor and its peritumoral microenvironment to predict progression-free survival (PFS) in patients with clinical stage IA pure-solid non-small cell lung cancer (NSCLC).

MATERIALS AND METHODS

234 Patients who underwent lung resection for NSCLC from two hospitals were reviewed. Radiomic features were extracted from both intratumoral, peritumoral and habitat regions on PET. Univariate and multivariate logistic regression analyses were employed to determine significant clinical variables. Subsequently, a radiomics nomogram was developed by combining the radiomics signature with these identified clinical variables. Kaplan-Meier (KM) analysis was performed to investigate the prognostic value of the nomogram. Shapley Additive Explanations (SHAP) were used to interpret the ML models.

RESULTS

The combination model which contained peritumoral 5 mm and habitat regions radiomics features, clinical variables obtained a strong well-performance, achieving area under the curve (AUC) of 0.905 (95% confidence interval (CI) 0.854-0.957) in the train set and 0.875 (95% CI 0.789-0.962) in the internal validation set. The radiomics signature was significantly associated with PFS, the model significantly discerned high and low-risk patients, and exhibited a significant benefit in the clinical use showed low-risk score given have far longer RFS than those with high-risk score (log-rank P＜0.001).

CONCLUSION

The habitat and peritumoral radiomics signatures serve as an independent biomarker for predicting PFS in patients with early-stage NSCLC, effectively stratified survival risk among patients with clinical stage IA pure-solid non-small cell lung cancer.

Enhanced CT-Based Intratumoral and Peritumoral Radiomics Nomograms Predict High-Grade Patterns of Invasive Lung Adenocarcinoma

RATIONALE AND OBJECTIVES

Extraction of intratumoral and peritumoral radiomics features combined with clinical factors to establish nomograms to predict high-grade patterns (micropapillary and solid) of invasive adenocarcinoma of the lung (IAC).

MATERIALS AND METHODS

A retrospective study was conducted on 463 patients with pathologically confirmed IAC. Patients were randomized in a 7:3 ratio into a training cohort (n = 324) and a testing cohort (n = 139). A total of 2154 CT-based radiomic features were extracted from each of the four regions: gross tumor volume (GTV) and gross peritumoral tumor volume (GPTV3, GPTV6, GPTV9) containing peri-tumor regions of 3 mm, 6 mm, and 9 mm. A radiomics nomogram was constructed based on the optimal radiomics model and clinically independent predictors.

RESULTS

The GPTV3 radiomics model showed better predictive performance in the testing group compared to the GTV (0.840), GPTV6 (0.843), and GPTV9 (0.734) models, with an AUC value of 0.889 in the testing group. In the clinical model, tumor density and the presence of a spiculation sign were identified as independent predictors. The nomogram, which combined these independent predictors with the GPTV3-Radscore, proved to be clinically useful.

CONCLUSION

The GPTV3 radiomics model was superior to the GTV, GPTV6, and GPTV9 radiomics models in predicting high-grade patterns (HGP) of IAC. In addition, nomograms based on GPTV3 radiomics features and clinically independent predictors can further improve the prediction efficiency.

Longitudinal Changes of CT-radiomic and Systemic Inflammatory Features Predict Survival in Advanced Non-Small Cell Lung Cancer Patients Treated With Immune Checkpoint Inhibitors

PURPOSE

This study aims to determine whether longitudinal changes in CT radiomic features (RFs) and systemic inflammatory indices outperform single-time-point assessment in predicting survival in advanced non-small cell lung cancer (NSCLC) treated with immune checkpoint inhibitors (ICIs).

MATERIALS AND METHODS

We retrospectively acquired pretreatment (T0) and first disease assessment (T1) RFs and systemic inflammatory indices from a single-center cohort of stage IV NSCLC patients and computed their delta (Δ) variation as [(T1-T0)/T0]. RFs from the primary tumor were selected for building baseline-radiomic (RAD) and Δ-RAD scores using the linear combination of standardized predictors detected by LASSO Cox regression models. Cox models were generated using clinical features alone or combined with baseline and Δ blood parameters and integrated with baseline-RAD and Δ-RAD. All models were 3-fold cross-validated. A prognostic index (PI) of each model was tested to stratify overall survival (OS) through Kaplan-Meier analysis.

RESULTS

We included 90 ICI-treated NSCLC patients (median age 70 y [IQR=42 to 85], 63 males). Δ-RAD outperformed baseline-RAD for predicting OS [c-index: 0.632 (95%CI: 0.628 to 0.636) vs. 0.605 (95%CI: 0.601 to 0.608) in the test splits]. Integrating longitudinal changes of systemic inflammatory indices and Δ-RAD with clinical data led to the best model performance [Integrated-Δ model, c-index: 0.750 (95% CI: 0.749 to 0.751) in training and 0.718 (95% CI: 0.715 to 0.721) in testing splits]. PI enabled significant OS stratification within all the models ( P -value <0.01), reaching the greatest discriminative ability in Δ models (high-risk group HR up to 7.37, 95% CI: 3.9 to 13.94, P <0.01).

CONCLUSION

Δ-RAD improved OS prediction compared with single-time-point radiomic in advanced ICI-treated NSCLC. Integrating Δ-RAD with a longitudinal assessment of clinical and laboratory data further improved the prognostic performance.

PET radiomics-based lymphovascular invasion prediction in lung cancer using multiple segmentation and multi-machine learning algorithms

The current study aimed to predict lymphovascular invasion (LVI) using multiple machine learning algorithms and multi-segmentation positron emission tomography (PET) radiomics in non-small cell lung cancer (NSCLC) patients, offering new avenues for personalized treatment strategies and improving patient outcomes. One hundred and twenty-six patients with NSCLC were enrolled in this study. Various automated and semi-automated PET image segmentation methods were applied, including Local Active Contour (LAC), Fuzzy-C-mean (FCM), K-means (KM), Watershed, Region Growing (RG), and Iterative thresholding (IT) with different percentages of the threshold. One hundred five radiomic features were extracted from each region of interest (ROI). Multiple feature selection methods, including Minimum Redundancy Maximum Relevance (MRMR), Recursive Feature Elimination (RFE), and Boruta, and multiple classifiers, including Multilayer Perceptron (MLP), Logistic Regression (LR), XGBoost (XGB), Naive Bayes (NB), and Random Forest (RF), were employed. Synthetic Minority Oversampling Technique (SMOTE) was also used to determine if it boosts the area under the ROC curve (AUC), accuracy (ACC), sensitivity (SEN), and specificity (SPE). Our results indicated that the combination of SMOTE, IT (with 45% threshold), RFE feature selection and LR classifier showed the best performance (AUC = 0.93, ACC = 0.84, SEN = 0.85, SPE = 0.84) followed by SMOTE, FCM segmentation, MRMR feature selection, and LR classifier (AUC = 0.92, ACC = 0.87, SEN = 1, SPE = 0.84). The highest ACC belonged to the IT segmentation (with 45 and 50% thresholds) alongside Boruta feature selection and the NB classifier without SMOTE (ACC = 0.9, AUC = 0.78 and 0.76, SEN = 0.7, and SPE = 0.94, respectively). Our results indicate that selection of appropriate segmentation method and machine learning algorithm may be helpful in successful prediction of LVI in patients with NSCLC with high accuracy using PET radiomics analysis.

Predicting Lymphovascular Invasion in Non-small Cell Lung Cancer Using Deep Convolutional Neural Networks on Preoperative Chest CT

RATIONALE AND OBJECTIVES

Lymphovascular invasion (LVI) plays a significant role in precise treatments of non-small cell lung cancer (NSCLC). This study aims to build a non-invasive LVI prediction diagnosis model by combining preoperative CT images with deep learning technology.

MATERIALS AND METHODS

This retrospective observational study included a series of consecutive patients who underwent surgical resection for non-small cell lung cancer (NSCLC) and received pathologically confirmed diagnoses. The cohort was randomly divided into a training group comprising 70 % of the patients and a validation group comprising the remaining 30 %. Four distinct deep convolutional neural network (DCNN) prediction models were developed, incorporating different combination of two-dimensional (2D) and three-dimensional (3D) CT imaging features as well as clinical-radiological data. The predictive capabilities of the models were evaluated by receiver operating characteristic curves (AUC) values and confusion matrices. The Delong test was utilized to compare the predictive performance among the different models.

RESULTS

A total of 3034 patients with NSCLC were recruited in this study including 106 LVI+ patients. In the validation cohort, the Dual-head Res2Net_3D23F model achieved the highest AUC of 0.869, closely followed by the models of Dual-head Res2Net_3D3F (AUC, 0.868), Dual-head Res2Net_3D (AUC, 0.867), and EfficientNet-B0_2D (AUC, 0.857). There was no significant difference observed in the performance of the EfficientNet-B0_2D model when compared to the Dual-head Res2Net_3D3F and Dual-head Res2Net_3D23F.

CONCLUSION

Findings of this study suggest that utilizing deep convolutional neural network is a feasible approach for predicting pathological LVI in patients with NSCLC.

Intratumoral and peritumoral radiomics for preoperative prediction of neoadjuvant chemotherapy effect in breast cancer based on 18F-FDG PET/CT

OBJECTIVE

To investigate the value of 18F-FDG PET/CT-based intratumoral and peritumoral radiomics in predicting the efficacy of neoadjuvant chemotherapy (NAC) for breast cancer.

METHODS

190 patients who met the inclusion and exclusion criteria from 2017 to 2022 were studied. Features were extracted from the PET/CT intratumoral and peritumoral regions, feature selection was performed through the correlation analysis, t-tests, and least absolute shrinkage and selection operator regression (LASSO). Four classifiers, support vector machine (SVM), k-nearest neighbor (KNN), logistic regression (LR), and naive bayes (NB) were used to build the prediction models. The receiver operating characteristic (ROC) curves were plotted to measure the predictive performance of the models. Concurrent stratified analysis was conducted to establish subtype-specific features for each molecular subtype.

RESULTS

Compared to intratumoral features alone, intratumoral + peritumoral features achieved higher AUC values in each classifier. The SVM model constructed with intratumoral + peritumoral features achieved the highest AUC values in both the train and test set (train set: 0.95 and test set: 0.83). Subtype-specific features improve performance in predicting the efficacy of NAC (luminal group: 0.90; HER2 + group: 0.86; triple negative group: 0.92).

CONCLUSION

Intratumoral and peritumoral radiomics models based on 18F-FDG PET/CT can reliably forecast the efficacy of NAC, thereby assisting clinical decision-making.

Development and validation of an inflammation-nutrition indices-based nomogram for predicting early recurrence in patients with stage IB lung adenocarcinoma

To explore the inflammation-nutrition indices and related clinical factors affecting early recurrence in patients with stage IB LUAD. A retrospective analysis was conducted on clinical and pathological data of patients diagnosed with stage IB LUAD who underwent radical surgery in our hospital from January 2016 to January 2021. Using R software, patients were randomly divided into training (n = 140) and validation (n = 59) cohorts in a 7:3 ratio. Univariate and multivariate Cox regression analyses were performed to identify risk factors for RFS and construct a predictive model. The performance of the model was evaluated using the area under the receiver operating characteristic curve (AUC), concordance index (C-index), and calibration curve. Clinical utility of the model was assessed using decision curve analysis (DCA). Multivariate Cox regression analysis revealed that vascular invasion, visceral pleural invasion, predominant pattern, preoperative NLR > 2.33, preoperative PLR > 127.62, and preoperative PNI ≤ 48.3 were independent risk factors for RFS. The C-index of the nomogram model constructed based on these independent risk factors was 0.825 (95% CI: 0.762-0.881) in the training cohort and 0.772 (95% CI: 0.667-0.876) in the validation cohort. The ROC curves showed AUCs of 0.902, 0.881, and 0.877 for 1-year, 2-year, and 3-year RFS in the training cohort and AUCs of 0.782, 0.825, and 0.732 in the validation cohort respectively. Calibration curve and decision curve analysis indicated good clinical value of the model. The nomogram model based on inflammation-nutrition indices has predictive value for early recurrence in patients with stage IB LUAD.

Evaluating peritumoral and intratumoral radiomics signatures for predicting lymph node metastasis in surgically resectable non-small cell lung cancer

Background

Whether lymph node metastasis in non-small cell lung cancer is critical to clinical decision-making. This study was to develop a non-invasive predictive model for preoperative assessing lymph node metastasis in patients with non-small cell lung cancer (NSCLC) using radiomic features from chest CT images.

Materials & methods

In this retrospective study, 247 patients with resectable non-small cell lung cancer (NSCLC) were enrolled. These individuals underwent preoperative chest CT scans that identified lung nodules, followed by lobectomies and either lymph node sampling or dissection. We extracted both intratumoral and peritumoral radiomic features from the CT images, which were used as covariates to predict the lymph node metastasis status. By using ROC curves, Delong tests, Calibration curve, and DCA curves, intra-tumoral-peri-tumoral model performance were compared with models using only intratumoral features or clinical information. Finally, we constructed a model that combined clinical information and radiomic features to increase clinical applicability.

Results

This study enrolled 247 patients (117 male and 130 females). In terms of predicting lymph node metastasis, the intra-tumoral-peri-tumoral model (0.953, 95%CI 0.9272-0.9792) has a higher AUC compared to the intratumoral radiomics model (0.898, 95%CI 0.8553-0.9402) and the clinical model (0.818, 95%CI 0.7653-0.8709). The DeLong test shows that the performance of the Intratumoral and Peritumoral radiomics models is superior to that of the Intratumoral or clinical feature model (p <0.001). In addition, to increase the clinical applicability of the model, we combined the intratumoral-peritumoral model and clinical information to construct a nomogram. Nomograms still have good predictive performance.

Conclusion

The radiomics-based model incorporating both peritumoral and intratumoral features from CT images can more accurately predict lymph node metastasis in NSCLC than traditional methods.

Analysis of postoperative recurrence-free survival in non-small cell lung cancer patients based on consensus clustering

AIMS

This study aims to assess whether consensus clustering, based on computed tomography (CT) radiomics from both intratumoral and peritumoral regions, can effectively stratify the risk of non-small cell lung cancer (NSCLC) patients and predict their postoperative recurrence-free survival (RFS).

MATERIALS AND METHODS

A retrospective analysis was conducted on the data of surgical patients diagnosed with NSCLC between December 2014 and April 2020. After preprocessing CT images, radiomic features were extracted from a 9-mm region encompassing both the tumor and its peritumoral area. Consensus clustering was utilized to analyze the radiomics features and categorize patients into distinct clusters. A comparison of the differences in clinical pathological characteristics was conducted among the clusters. Kaplan-Meier survival analysis was employed to investigate differences in survival among the clusters.

RESULTS

A total of 266 patients were included in this study, and consensus clustering identified three clusters (Cluster 1: n=111, Cluster 2: n=61, Cluster 3: n=94). Multiple clinical risk factors, including pathological TNM staging, programmed cell death ligand 1 (PD-L1), and epidermal growth factor receptor (EGFR) expression status exhibit significant differences among the three clusters. Kaplan-Meier survival analysis demonstrated significant variations in RFS across the clusters (P<0.001). The 3-year cumulative recurrence-free survival rates were 76.5% (95% CI: 68.6-84.4) for Cluster 1, 45.9% (95% CI: 33.4-58.4) for Cluster 2, and 41.5% (95% CI: 31.6-51.5) for Cluster 3.

CONCLUSIONS

Consensus clustering of CT radiomics based on intratumoral and peritumoral regions can stratify the risk of postoperative recurrence in patients with NSCLC.

Predicting lymphovascular invasion in N0 stage non-small cell lung cancer: A nomogram based on Dual-energy CT imaging and clinical findings

PURPOSE

To construct a nomogram for predicting lymphovascular invasion (LVI) in N0 stage non-small cell lung cancer (NSCLC) using dual-energy computed tomography (DECT) findings combined with clinical findings.

METHODS

We retrospectively recruited 135 patients with N0 stage NSCLC from two hospitals underwent DECT before surgery and were divided into development cohort (n = 107) and validation cohort (n = 28). The clinical findings (baseline characteristics, biochemical markers, serum tumor markers and Immunohistochemical markers), DECT-derived parameters (iodine concentration [IC], effective atomic number [Eff-Z] and normalized iodine concentration [NIC], iodine enhancement [IE] and NIC ratio [NICr]) and Fractal dimension (FD) were collected and measured. A nomogram was constructed using significant findings to predict LVI in N0 stage NSCLC and was externally validated.

RESULTS

Multivariable analysis revealed that lymphocyte count (LYMPH, odds ratio [OR]: 3.71, P=0.014), IC in arterial phase (ICa, OR: 1.25, P=0.021), NIC in venous phase (NICv, OR: 587.12, P=0.009) and FD (OR: 0.01, P=0.033) were independent significant factors for predicting LVI in N0 stage NSCLC, and were used to construct a nomogram. The nomogram exhibited robust predictive capabilities in both the development and validation cohort, with AUCs of 0.819 (95 % CI: 72.6-90.4) and 0.844 (95 % CI: 68.2-95.8), respectively. The calibration plots showed excellent agreement between the predicted probabilities and the actual rates of positive LVI, on external validation.

CONCLUSIONS

Combination of clinical and DECT imaging findings could aid in predicting LVI in N0 stage NSCLC using significant findings of LYMPH, ICa, NICv and FD.

Prognostic Value and Pathological Correlation of Peritumoral Radiomics in Surgically Resected Non-Small Cell Lung Cancer

RATIONALE AND OBJECTIVES

To determine the additional value of peritumoral radiomics in predicting overall survival (OS) in surgically resected non-small cell lung cancer (NSCLC) and its correlation with pathological findings.

METHODS

A total of 526 patients with surgically resected NSCLC were included (191 training, 160 internal validation, and 175 external validation cohorts). CT images were used to segment the gross tumor volume (GTV) and peritumoral volume (PTV) within distances of 3, 6, 9 mm from the tumor boundary (PTV3, PTV6, and PTV9), and radiomic features were extracted. Four prognostic models for OS (GTV, GTV + PTV3, GTV + PTV6, and GTV + PTV9) were constructed using the training cohort. The prognostic ability and feature importance were evaluated using the validation cohorts. Pathological findings were compared between the two patient groups (n = 30 for each) having the top 30 and bottom 30 values of the most important peritumoral feature.

RESULTS

The GTV+ PTV3 models exhibited the highest predictive ability, which was higher than that of the GTV model in the internal validation cohort (C-index: 0.666 vs. 0.616, P = 0.027) and external validation cohort (C-index: 0.705 vs. 0.656, P = 0.048). The most important feature was GLDM_Dependence_Entropy, extracted from PTV3. High peritumoral GLDM_Dependence_Entropy was associated with a high proportion of invasive histological types, tumor spread through air spaces, and tumor-infiltrating lymphocytes (all P < 0.05).

CONCLUSION

The GTV and PTV3 combination demonstrated a higher prognostic ability, compared to GTV alone. Peritumoral radiomic features may be associated with various pathological prognostic factors.

2.5D peritumoural radiomics predicts postoperative recurrence in stage I lung adenocarcinoma

Objective

Radiomics can non-invasively predict the prognosis of a tumour by applying advanced imaging feature algorithms.The aim of this study was to predict the chance of postoperative recurrence by modelling tumour radiomics and peritumour radiomics and clinical features in patients with stage I lung adenocarcinoma (LUAD).

Materials and methods

Retrospective analysis of 190 patients with postoperative pathologically confirmed stage I LUAD from centre 1, who were divided into training cohort and internal validation cohort, with centre 2 added as external validation cohort. To develop a combined radiation-clinical omics model nomogram incorporating clinical features based on images from low-dose lung cancer screening CT plain for predicting postoperative recurrence and to evaluate the performance of the nomogram in the training cohort, internal validation cohort and external validation cohort.

Results

A total of 190 patients were included in the model in centre 1 and randomised into a training cohort of 133 and an internal validation cohort of 57 in a ratio of 7:3, and 39 were included in centre 2 as an external validation cohort. In the training cohort (AUC=0.865, 95% CI 0.824-0.906), internal validation cohort (AUC=0.902, 95% CI 0.851-0.953) and external validation cohort (AUC=0.830,95% CI 0.751-0.908), the combined radiation-clinical omics model had a good predictive ability. The combined model performed significantly better than the conventional single-modality models (clinical model, radiomic model), and the calibration curve and decision curve analysis (DCA) showed high accuracy and clinical utility of the nomogram.

Conclusion

The combined preoperative radiation-clinical omics model provides good predictive value for postoperative recurrence in stage ILUAD and combines the model's superiority in both internal and external validation cohorts, demonstrating its potential to aid in postoperative treatment strategies.

Application of a Comprehensive Model Based on CT Radiomics and Clinical Features for Postoperative Recurrence Risk Prediction in Non-small Cell Lung Cancer

RATIONALE AND OBJECTIVES

We constructed a comprehensive model by combining the radiomics and clinical features of tumors to predict the recurrence risk of patients with operable stage IA-IIIA non-small cell lung cancer (NSCLC). Our aim was to improve the accuracy of prognostic prediction and provide personalized treatment plans to enhance patient outcomes.

MATERIALS AND METHODS

We retrospectively analyzed 152 surgically treated patients with pathologically confirmed stage IA-IIIA NSCLC. These patients were randomly divided into a training cohort and a test cohort in an 8:2 ratio. Using the 3D Slicer image computing platform, we manually delineated the regions of interest (ROI) for all lesions and extracted radiomics features using Python. We used the Least Absolute Shrinkage and Selection Operator (LASSO) to select the radiomics features, while the COX multivariate regression model was employed to identify independent clinical risk factors for recurrence. Finally, we utilized logistic regression (LR) to build the model and validated it using the receiver operating characteristic curve (ROC). The predictive performance of the model was evaluated using the concordance index (C-index), and the clinical value of the model was compared through decision curve analysis (DCA).

RESULTS

We extracted a total of 1562 radiomics features. After feature selection, we retained 29 features. The COX multivariate regression model demonstrated that the N stage was an independent risk factor for postoperative recurrence. In the training and test cohorts, the area under the curve (AUC) values of the radiomics-clinical comprehensive model were 0.972 and 0.937, respectively, while the C-index values were 0.815 and 0.847. These values surpassed those of the standalone clinical model or radiomics model.

CONCLUSION

Our study demonstrates that a comprehensive model based on CT radiomics and clinical features can effectively stratify the risk of postoperative recurrence in patients with operable NSCLC. It provides a powerful tool for accurately stratifying the risk of high-risk patients after surgery.

EfficientNet-Based System for Detecting EGFR-Mutant Status and Predicting Prognosis of Tyrosine Kinase Inhibitors in Patients with NSCLC

We aimed to develop and validate a deep learning-based system using pre-therapy computed tomography (CT) images to detect epidermal growth factor receptor (EGFR)-mutant status in patients with non-small cell lung cancer (NSCLC) and predict the prognosis of advanced-stage patients with EGFR mutations treated with EGFR tyrosine kinase inhibitors (TKI). This retrospective, multicenter study included 485 patients with NSCLC from four hospitals. Of them, 339 patients from three centers were included in the training dataset to develop an EfficientNetV2-L-based model (EME) for predicting EGFR-mutant status, and the remaining patients were assigned to an independent test dataset. EME semantic features were extracted to construct an EME-prognostic model to stratify the prognosis of EGFR-mutant NSCLC patients receiving EGFR-TKI. A comparison of EME and radiomics was conducted. Additionally, we included patients from The Cancer Genome Atlas lung adenocarcinoma dataset with both CT images and RNA sequencing data to explore the biological associations between EME score and EGFR-related biological processes. EME obtained an area under the curve (AUC) of 0.907 (95% CI 0.840-0.926) on the test dataset, superior to the radiomics model (P = 0.007). The EME and radiomics fusion model showed better (AUC, 0.941) but not significantly increased performance (P = 0.895) compared with EME. In prognostic stratification, the EME-prognostic model achieved the best performance (C-index, 0.711). Moreover, the EME-prognostic score showed strong associations with biological pathways related to EGFR expression and EGFR-TKI efficacy. EME demonstrated a non-invasive and biologically interpretable approach to predict EGFR status, stratify survival prognosis, and correlate biological pathways in patients with NSCLC.

Radiomics features based on dual-area CT predict the expression levels of fatty acid binding protein 4 and outcome in hepatocellular carcinoma

RATIONALE AND OBJECTIVES

To evaluate the predictive value of tumor and peritumor radiomics in the fatty acid binding protein 4 (FABP4) expression levels and overall survival in patients with hepatocellular carcinoma.

MATERIALS AND METHODS

The genomic data of HCC patients were obtained from The Cancer Genome Atlas. The Dual-area CT images of corresponding patients were downloaded from The Cancer Imaging Archive, for radiomics feature extraction, model construction and prognosis analysis. Simultaneously, using patients from Sichuan Provincial People's Hospital, the prognostic value of the radiomics model in HCC patients was validated.

RESULTS

In the TCIA database, the area under the curve (AUC) values of the volumes of interest (VOI)whole model in the training set and internal validation set were 0.812 and 0.754, respectively, and the AUC value of VOIwhole+periphery in the training set and internal validation set were 0.866 and 0.779, respectively. In the VOIwhole and the VOIwhole+periphery model of the independent cohort, there were significant differences in OS between the high and low rad-score groups (P = 0.009, P = 0.021, respectively). Significant positive correlations can be observed between FABP4 expression and correlations with rad-score of VOIwhole model (r = 0.691) and VOIwhole+periphery model (r = 0.732) in the independent cohort.

CONCLUSION

Radiomics models of tumor and peritumor Dual-area CT images could predict stably the expression levels of FABP4 and may be helping in personalized treatment strategies.

Impact of tumour stroma-immune interactions on survival prognosis and response to neoadjuvant chemotherapy in bladder cancer

BACKGROUND

The tumour stroma is associated with unfavourable prognosis in diverse solid tumours, but its prognostic and predictive value in bladder cancer (BCa) is unclear.

METHODS

In this multicentre, retrospective study, we included 830 patients with BCa from six independent cohorts. Differences in overall survival (OS) and cancer-specific survival (CSS) were investigated between high-tumour stroma ratio (TSR) and low-TSR groups. Multi-omics analyses, including RNA sequencing, immunohistochemistry, and single-cell RNA sequencing, were performed to study stroma-immune interactions. TSR prediction models were developed based on pelvic CT scans, and the best performing model was selected based on receiver operator characteristic analysis.

FINDINGS

Compared to low-TSR tumours, high-TSR tumours were significantly associated with worse OS (HR = 1.193, 95% CI: 1.046-1.361, P = 0.008) and CSS (HR = 1.337, 95% CI: 1.139-1.569, P < 0.001), and lower rate of pathological complete response (pCR) to neoadjuvant chemotherapy (NAC). High-TSR tumours exhibited higher infiltration of immunosuppressive cells, including Tregs and tumour-associated neutrophils, while low-TSR tumours exhibited higher infiltration of immune-activating cells such as CD8+ Teff and XCR1+ dendritic cells. The TSR prediction model was developed by combining the intra-tumour and tumour base radiomics features, and showed good performance to predict high-TSR, as indicted by area under the curve of 0.871 (95% CI: 0.821-0.921), 0.821 (95% CI: 0.731-0.911), and 0.801 (95% CI: 0.737-0.865) in the training, internal validation, and external validation cohorts, respectively. In patients with low predicted TSR, 92.3% (12/13) achieved pCR, while only 35.3% (6/17) of patients with high predicted TSR achieved pCR.

INTERPRETATION

The tumour stroma was found to be significantly associated with clinical outcomes in patients with BCa as a result of tumour stroma-immune interactions. The radiomics prediction model provided non-invasive evaluation of TSR and was able to predict pCR in patients receiving NAC for BCa.

FUNDING

This work was supported by National Natural Science Foundation of China (Grant No. 82373254 and 81961128027), Guangdong Provincial Natural Science Foundation (Grant No. 2023A1515010258), Science and Technology Planning Project of Guangdong Province (Grant No. 2023B1212060013). Science and Technology Program of Guangzhou (SL2022A04J01754), Sun Yat-Sen Memorial Hospital Clinical Research 5010 Program (Grant No. SYS-5010Z-202401).

Nomogram using intratumoral and peritumoral radiomics for the preoperative prediction of visceral pleural invasion in clinical stage IA lung adenocarcinoma

BACKGROUND

Accurate prediction of visceral pleural invasion (VPI) in lung adenocarcinoma before operation can provide guidance and help for surgical operation and postoperative treatment. We investigate the value of intratumoral and peritumoral radiomics nomograms for preoperatively predicting the status of VPI in patients diagnosed with clinical stage IA lung adenocarcinoma.

METHODS

A total of 404 patients from our hospital were randomly assigned to a training set (n = 283) and an internal validation set (n = 121) using a 7:3 ratio, while 81 patients from two other hospitals constituted the external validation set. We extracted 1218 CT-based radiomics features from the gross tumor volume (GTV) as well as the gross peritumoral tumor volume (GPTV5, 10, 15), respectively, and constructed radiomic models. Additionally, we developed a nomogram based on relevant CT features and the radscore derived from the optimal radiomics model.

RESULTS

The GPTV10 radiomics model exhibited superior predictive performance compared to GTV, GPTV5, and GPTV15, with area under the curve (AUC) values of 0.855, 0.842, and 0.842 in the three respective sets. In the clinical model, the solid component size, pleural indentation, solid attachment, and vascular convergence sign were identified as independent risk factors among the CT features. The predictive performance of the nomogram, which incorporated relevant CT features and the GPTV10-radscore, outperformed both the radiomics model and clinical model alone, with AUC values of 0.894, 0.828, and 0.876 in the three respective sets.

CONCLUSIONS

The nomogram, integrating radiomics features and CT morphological features, exhibits good performance in predicting VPI status in lung adenocarcinoma.

Development of an Intratumoral and Peritumoral Radiomics Nomogram Using Digital Breast Tomosynthesis for Preoperative Assessment of Lymphovascular Invasion in Invasive Breast Cancer

RATIONALE AND OBJECTIVES

This study aimed to create a nomogram model that combines clinical factors with radiomics analysis of both intra- and peritumoral regions extracted from preoperative digital breast tomosynthesis (DBT) images, in order to develop a reliable method for predicting the lymphovascular invasion (LVI) status in invasive breast cancer (IBC) patients.

MATERIALS AND METHODS

A total of 178 patients were randomly split into a training dataset (N = 124) and a validation dataset (N = 54). Comprehensive clinical data, encompassing DBT features, were gathered for all cases. Radiomics features were extracted and selected from intra- and peritumoral region to establish radiomics signature (Radscore). To construct the clinical model and nomogram model, univariate and multivariate logistic regression analyses were utilized to identify independent risk factors. To assess and validate these models, various analytical methods were employed, including receiver operating characteristic (ROC) curve analysis, calibration curve analysis, decision curve analysis (DCA), net reclassification improvement (NRI), and integrated discriminatory improvement (IDI).

RESULTS

The clinical model is constructed based on two independent risk factors: tumor margin and the DBT-reported lymph node metastasis (DBT_reported_LNM). Incorporating Radscore_Combine (utilizing both intra- and peritumoral radiomics features), tumor margin, and DBT_reported_LNM into the nomogram achieved a reliable predictive performance, with area under the curve (AUC) values of 0.906 and 0.905 in both datasets, respectively. The significant improvement demonstrated by the NRI and IDI indicates that the Radscore_Combine could be a valuable biomarker for effectively predicting the status of LVI.

CONCLUSION

The nomogram demonstrated a reliable ability to predict LVI in IBC patients.

Reproducibility of radiomics quality score: an intra- and inter-rater reliability study

OBJECTIVES

To investigate the intra- and inter-rater reliability of the total radiomics quality score (RQS) and the reproducibility of individual RQS items' score in a large multireader study.

METHODS

Nine raters with different backgrounds were randomly assigned to three groups based on their proficiency with RQS utilization: Groups 1 and 2 represented the inter-rater reliability groups with or without prior training in RQS, respectively; group 3 represented the intra-rater reliability group. Thirty-three original research papers on radiomics were evaluated by raters of groups 1 and 2. Of the 33 papers, 17 were evaluated twice with an interval of 1 month by raters of group 3. Intraclass coefficient (ICC) for continuous variables, and Fleiss' and Cohen's kappa (k) statistics for categorical variables were used.

RESULTS

The inter-rater reliability was poor to moderate for total RQS (ICC 0.30-055, p < 0.001) and very low to good for item's reproducibility (k - 0.12 to 0.75) within groups 1 and 2 for both inexperienced and experienced raters. The intra-rater reliability for total RQS was moderate for the less experienced rater (ICC 0.522, p = 0.009), whereas experienced raters showed excellent intra-rater reliability (ICC 0.91-0.99, p < 0.001) between the first and second read. Intra-rater reliability on RQS items' score reproducibility was higher and most of the items had moderate to good intra-rater reliability (k - 0.40 to 1).

CONCLUSIONS

Reproducibility of the total RQS and the score of individual RQS items is low. There is a need for a robust and reproducible assessment method to assess the quality of radiomics research.

CLINICAL RELEVANCE STATEMENT

There is a need for reproducible scoring systems to improve quality of radiomics research and consecutively close the translational gap between research and clinical implementation.

KEY POINTS

• Radiomics quality score has been widely used for the evaluation of radiomics studies. • Although the intra-rater reliability was moderate to excellent, intra- and inter-rater reliability of total score and point-by-point scores were low with radiomics quality score. • A robust, easy-to-use scoring system is needed for the evaluation of radiomics research.

Development and validation of a deep learning signature for predicting lymphovascular invasion and survival outcomes in clinical stage IA lung adenocarcinoma: A multicenter retrospective cohort study

PURPOSE

The presence of lymphovascular invasion (LVI) influences the management and outcomes of patients with clinical stage IA lung adenocarcinoma. The objective was the development of a deep learning (DL) signature for the prediction of LVI and stratification of prognosis.

METHODS

A total of 2077 patients from three centers were retrospectively enrolled and divided into a training set (n = 1515), an internal validation set (n = 381), and an external set (n = 181). A -three-dimensional residual neural network was used to extract the DL signature and three models, namely, the clinical, DL, and combined models, were developed. Diagnostic efficiency was assessed by ROC curves and AUC values. Kaplan-Meier curves and Cox proportional hazards regression analyses were conducted to evaluate links between various factors and disease-free survival.

RESULTS

The DL model could effectively predict LVI, shown by AUC values of 0.72 (95 %CI: 0.68-0.76) and 0.63 (0.54-0.73) in the internal and external validation sets, respectively. The incorporation of DL signature and clinical-radiological factors increased the AUC to 0.74 (0.71-0.78) and 0.77 (0.70-0.84) in comparison with the DL and clinical models (AUC of 0.71 [0.68-0.75], 0.71 [0.61-0.81]) in the internal and external validation sets, respectively. Pathologic LVI, LVI predicted by both DL and combined models were associated with unfavorable prognosis (all p < 0.05).

CONCLUSION

The effectiveness of the DL signature in the diagnosis of LVI and prognosis prediction in patients with clinical stage IA lung adenocarcinoma was demonstrated. These findings suggest the potential of the model in clinical decision-making.

CT-Based Intratumoral and Peritumoral Radiomics Nomograms for the Preoperative Prediction of Spread Through Air Spaces in Clinical Stage IA Non-small Cell Lung Cancer

The study aims to investigate the value of intratumoral and peritumoral radiomics and clinical-radiological features for predicting spread through air spaces (STAS) in patients with clinical stage IA non-small cell lung cancer (NSCLC). A total of 336 NSCLC patients from our hospital were randomly divided into the training cohort (n = 236) and the internal validation cohort (n = 100) at a ratio of 7:3, and 69 patients from the other two external hospitals were collected as the external validation cohort. Univariate and multivariate analyses were used to select clinical-radiological features and construct a clinical model. The GTV, PTV5, PTV10, PTV15, PTV20, GPTV5, GPTV10, GPTV15, and GPTV20 models were constructed based on intratumoral and peritumoral (5 mm, 10 mm, 15 mm, 20 mm) radiomics features. Additionally, the radscore of the optimal radiomics model and clinical-radiological predictors were used to construct a combined model and plot a nomogram. Lastly, the ROC curve and AUC value were used to evaluate the diagnostic performance of the model. Tumor density type (OR = 6.738) and distal ribbon sign (OR = 5.141) were independent risk factors for the occurrence of STAS. The GPTV10 model outperformed the other radiomics models, and its AUC values were 0.887, 0.876, and 0.868 in the three cohorts. The AUC values of the combined model constructed based on GPTV10 radscore and clinical-radiological predictors were 0.901, 0.875, and 0.878. DeLong test results revealed that the combined model was superior to the clinical model in the three cohorts. The nomogram based on GPTV10 radscore and clinical-radiological features exhibited high predictive efficiency for STAS status in NSCLC.

Leveraging multimodal MRI-based radiomics analysis with diverse machine learning models to evaluate lymphovascular invasion in clinically node-negative breast cancer

Objective

This study aimed to investigate and validate the effectiveness of diverse radiomics models for preoperatively differentiating lymphovascular invasion (LVI) in clinically node-negative breast cancer (BC).

Methods

This study included 198 patients diagnosed with clinically node-negative bc and pathologically confirmed LVI status from January 2018-July 2023. The training dataset consisted of 138 patients, while the validation dataset included 60. Radiomics features were extracted from multimodal magnetic resonance imaging obtained from T1WI, T2WI, DCE, DWI, and ADC sequences. Dimensionality reduction and feature selection techniques were applied to the extracted features. Subsequently, machine learning approaches, including logistic regression, support vector machine, classification and regression trees, k-nearest neighbors, and gradient boosting machine models (GBM), were constructed using the radiomics features. The best-performing radiomic model was selected based on its performance using the confusion matrix. Univariate and multivariable logistic regression analyses were conducted to identify variables for developing a clinical-radiological (Clin-Rad) model. Finally, a combined model incorporating both radiomics and clinical-radiological model features was created.

Results

A total of 6195 radiomic features were extracted from multimodal magnetic resonance imaging. After applying dimensionality reduction and feature selection, seven valuable radiomics features were identified. Among the radiomics models, the GBM model demonstrated superior predictive efficiency and robustness, achieving area under the curve values (AUC) of 0.881 (0.823,0.940) and 0.820 (0.693,0.947) in the training and validation datasets, respectively. The Clin-Rad model was developed based on the peritumoral edema and DWI rim sign. In the training dataset, it achieved an AUC of 0.767 (0.681, 0.854), while in the validation dataset, it achieved an AUC of 0.734 (0.555-0.913). The combined model, which incorporated radiomics and the Clin-Rad model, showed the highest discriminatory capability. In the training dataset, it had an AUC value of 0.936 (0.892, 0.981), and in the validation dataset, it had an AUC value of 0.876 (0.757, 0.995). Additionally, decision curve analysis of the combined model revealed its optimal clinical efficacy.

Conclusion

The combined model, integrating radiomics and clinical-radiological features, exhibited excellent performance in distinguishing LVI status. This non-invasive and efficient approach holds promise for aiding clinical decision-making in the context of clinically node-negative BC.

Radiomics Nomogram Based on Optimal Volume of Interest Derived from High-Resolution CT for Preoperative Prediction of IASLC Grading in Clinical IA Lung Adenocarcinomas: A Multi-Center, Large-Population Study

The novel grading system developed by the International Association for the Study of Lung Cancer (IASLC) for clinical stage IA lung adenocarcinomas has demonstrated remarkable prognostic capabilities. Notably, tumors classified as grade 3 have been associated with poor prognostic outcomes, thereby playing a crucial role in the formulation of personalized surgical strategies. The objective of this study is to develop a radiomics nomogram that utilizes the optimal volume of interest (VOI) derived from high-resolution CT (HRCT) scans to accurately predict the presence of grade 3 tumors in patients with clinical IA lung adenocarcinomas.In this multi-center, large-population study, clinical, pathological, and HRCT imaging data from 1418 patients who were pathologically diagnosed with lung adenocarcinomas were retrospectively collected. The data was obtained from four hospital databases between January 2018 and May 2022. From this patient cohort, 1206 individuals were screened from three databases and randomly divided into training and internal validation datasets in a 7:3 ratio. An additional dataset consisting of 212 individuals was used for external validation dataset. Radiomics features were extracted from HRCT images at various scales, including VOI-2mm, VOI entire, VOI +2mm, and VOI +4mm. To reduce dimensionality, select relevant features, and build radiomics signatures, the maximal redundancy minimal relevance (mRMR) and least absolute shrinkage and selection operator (LASSO) algorithm were utilized. Univariate and multivariate logistic regression analyses were conducted to identify independent clinic-radiological (Clin-Rad) predictors. Receiver operating characteristic (ROC) curves and corresponding area under the curve (AUC) were used to evaluate the diagnostic efficiency. A nomogram predicting the risk of grade 3 in clinical stage IA lung adenocarcinoma was constructed based on multivariate logistic regression, combining independent predictors and the optimal radiomics signatures.Multivariate logistic regression revealed that males exhibited a higher prevalence of grade 3 tumors, and solid nodules were frequently observed through radiological assessments. The utilization of radiomics features extracted from the VOI entire resulted in significant improvements in predictive performance, as evidenced by AUC values of 0.900 (0.880-0.942), 0.885 (0.824-0.946), and 0.888 (0.782-0.993) for the training, internal validation, and external validation datasets, respectively. Furthermore, the nomogram that combined VOI entire -based radiomics signatures and Clin-Rad characteristics, exhibited remarkable predictive performance. This was indicated by AUC values of 0.910(0.873-0.942), 0.891 (0.845-0.937), and 0.905 (0.846-0.964) for the training, internal validation, and external validation datasets, respectively.The extraction of radiomics features from both the indented and peri-tumoral regions does not offer any additional benefits in predicting grade 3 tumors according to the IASLC system. However, when combining the VOI entire-based radiomics model with Clin-Rad characteristics, the resulting integrated nomogram exhibited remarkable predictive performance.

CT Radiomic Nomogram Using Optimal Volume of Interest for Preoperatively Predicting Invasive Mucinous Adenocarcinomas in Patients with Incidental Pulmonary Nodules: A Multicenter, Large-Scale Study

INTRODUCTION

This study evaluated the efficacy of radiomic analysis with optimal volumes of interest (VOIs) on computed tomography images to preoperatively differentiate invasive mucinous adenocarcinoma (IMA) from non-mucinous adenocarcinoma (non-IMA) in patients with incidental pulmonary nodules (IPNs).

METHODS

This multicenter, large-scale retrospective study included 1383 patients with IPNs, 110 (8%) of whom were pathologically diagnosed with IMA postoperatively. Radiomic features were extracted from multi-scale VOI subgroups (VOI-2 mm, VOIentire, VOI + 2 mm, and VOI + 4 mm). Resampling methods, specifically, the synthetic minority oversampling technique, addressed the imbalance between the majority (IMA) and minority (non-IMA) groups. Radiomic features were identified using the least absolute shrinkage and selection operator algorithm. Radscores were calculated by linearly combining the selected features with their weights. A combined nomogram integrating the optimal VOI-based radiomic model with the image-finding classifier was constructed.

RESULTS

Bubble lucency and lower lobe predominance were significant in establishing an image-finding classifier to differentiate between IMA and non-IMA in IPNs, achieving an area under the curve (AUC) value of 0.684 (0.568-0.801). Across all radiomic models, IMA had a higher Radscore than did non-IMA. Specifically, the VOI + 2 mm-based radiomic model exhibited the highest performance, with an AUC of 0.832 (0.753-0.911). The combined nomogram outperformed the recognized image-finding classifier and radiomic models, achieving an AUC of 0.850 (0.776-0.925).

CONCLUSION

A nomogram that combines a recognized image-finding classifier with an optimal VOI-based radiomic model effectively predicts IMA in IPNs, aiding physicians in developing comprehensive treatment strategies.

Nomograms integrating CT radiomic and deep learning signatures to predict overall survival and progression-free survival in NSCLC patients treated with chemotherapy

OBJECTIVES

This study aims to establish nomograms to accurately predict the overall survival (OS) and progression-free survival (PFS) in patients with non-small cell lung cancer (NSCLC) who received chemotherapy alone as the first-line treatment.

MATERIALS AND METHODS

In a training cohort of 121 NSCLC patients, radiomic features were extracted, selected from intra- and peri-tumoral regions, and used to build signatures (S1 and S2) using a Cox regression model. Deep learning features were obtained from three convolutional neural networks and utilized to build signatures (S3, S4, and S5) that were stratified into over- and under-expression subgroups for survival risk using X-tile. After univariate and multivariate Cox regression analyses, a nomogram incorporating the tumor, node, and metastasis (TNM) stages, radiomic signature, and deep learning signature was established to predict OS and PFS, respectively. The performance was validated using an independent cohort (61 patients).

RESULTS

TNM stages, S2 and S3 were identified as the significant prognosis factors for both OS and PFS; S2 (OS: (HR (95%), 2.26 (1.40-3.67); PFS: (HR (95%), 2.23 (1.36-3.65)) demonstrated the best ability in discriminating patients with over- and under-expression. For the OS nomogram, the C-index (95% CI) was 0.74 (0.70-0.79) and 0.72 (0.67-0.78) in the training and validation cohorts, respectively; for the PFS nomogram, the C-index (95% CI) was 0.71 (0.68-0.81) and 0.72 (0.66-0.79). The calibration curves for the 3- and 5-year OS and PFS were in acceptable agreement between the predicted and observed survival. The established nomogram presented a higher overall net benefit than the TNM stage for predicting both OS and PFS.

CONCLUSION

By integrating the TNM stage, CT radiomic signature, and deep learning signatures, the established nomograms can predict the individual prognosis of NSCLC patients who received chemotherapy. The integrated nomogram has the potential to improve the individualized treatment and precise management of NSCLC patients.

Preoperative CT-based radiomics combined with tumour spread through air spaces can accurately predict early recurrence of stage I lung adenocarcinoma: a multicentre retrospective cohort study

OBJECTIVE

To develop and validate a prediction model for early recurrence of stage I lung adenocarcinoma (LUAD) that combines radiomics features based on preoperative CT with tumour spread through air spaces (STAS).

MATERIALS AND METHODS

The most recent preoperative thin-section chest CT scans and postoperative pathological haematoxylin and eosin-stained sections were retrospectively collected from patients with a postoperative pathological diagnosis of stage I LUAD. Regions of interest were manually segmented, and radiomics features were extracted from the tumour and peritumoral regions extended by 3 voxel units, 6 voxel units, and 12 voxel units, and 2D and 3D deep learning image features were extracted by convolutional neural networks. Then, the RAdiomics Integrated with STAS model (RAISm) was constructed. The performance of RAISm was then evaluated in a development cohort and validation cohort.

RESULTS

A total of 226 patients from two medical centres from January 2015 to December 2018 were retrospectively included as the development cohort for the model and were randomly split into a training set (72.6%, n = 164) and a test set (27.4%, n = 62). From June 2019 to December 2019, 51 patients were included in the validation cohort. RAISm had excellent discrimination in predicting the early recurrence of stage I LUAD in the training cohort (AUC = 0.847, 95% CI 0.762-0.932) and validation cohort (AUC = 0.817, 95% CI 0.625-1.000). RAISm outperformed single modality signatures and other combinations of signatures in terms of discrimination and clinical net benefits.

CONCLUSION

We pioneered combining preoperative CT-based radiomics with STAS to predict stage I LUAD recurrence postoperatively and confirmed the superior effect of the model in validation cohorts, showing its potential to assist in postoperative treatment strategies.

Radiomics under 2D regions, 3D regions, and peritumoral regions reveal tumor heterogeneity in non-small cell lung cancer: a multicenter study

PURPOSE

Lung cancer has significant genetic and phenotypic heterogeneity, leading to poor prognosis. Radiomic features have emerged as promising predictors of the tumor phenotype. However, the role of underlying information surrounding the cancer remains unclear.

MATERIALS AND METHODS

We conducted a retrospective study of 508 patients with NSCLC from three institutions. Radiomics models were built using features from six tumor regions and seven classifiers to predict three prognostically significant tumor phenotypes. The models were evaluated and interpreted by the mean area under the receiver operating characteristic curve (AUC) under nested cross-validation and Shapley values. The best-performing predictive models corresponding to six tumor regions and three tumor phenotypes were identified for further comparative analysis. In addition, we designed five experiments with different voxel spacing to assess the sensitivity of the experimental results to the spatial resolution of the voxels.

RESULTS

Our results demonstrated that models based on 2D, 3D, and peritumoral region features yielded mean AUCs and 95% confidence intervals of 0.759 and [0.747-0.771] for lymphovascular invasion, 0.889 and [0.882-0.896] for pleural invasion, and 0.839 and [0.829-0.849] for T-staging in the testing cohort, which was significantly higher than all other models. Similar results were obtained for the model combining the three regional features at five voxel spacings.

CONCLUSION

Our study revealed the predictive role of the developed methods with multi-regional features for the preoperative assessment of prognostic factors in NSCLC. The analysis of different voxel spacing and model interpretability strengthens the experimental findings and contributes to understanding the biological significance of the radiological phenotype.

Radiomics and artificial intelligence for precision medicine in lung cancer treatment

Lung cancer is the leading cause of cancer-related deaths worldwide. It exhibits, at the mesoscopic scale, phenotypic characteristics that are generally indiscernible to the human eye but can be captured non-invasively on medical imaging as radiomic features, which can form a high dimensional data space amenable to machine learning. Radiomic features can be harnessed and used in an artificial intelligence paradigm to risk stratify patients, and predict for histological and molecular findings, and clinical outcome measures, thereby facilitating precision medicine for improving patient care. Compared to tissue sampling-driven approaches, radiomics-based methods are superior for being non-invasive, reproducible, cheaper, and less susceptible to intra-tumoral heterogeneity. This review focuses on the application of radiomics, combined with artificial intelligence, for delivering precision medicine in lung cancer treatment, with discussion centered on pioneering and groundbreaking works, and future research directions in the area.