Radiomics and artificial intelligence for precision medicine in lung cancer treatment

Research advances in predicting the expansion of hypertensive intracerebral hemorrhage based on CT images: an overview

Hematoma expansion (HE) is an important risk factor for death or poor prognosis in patients with hypertensive intracerebral hemorrhage (HICH). Accurately predicting the risk of HE in patients with HICH is of great clinical significance for timely intervention and improving patient prognosis. Many imaging signs reported in literatures showed the important clinical value for predicting HE. In recent years, the development of radiomics and artificial intelligence has provided new methods for HE prediction with high accuracy. Therefore, this article reviews the latest research progress in CT imaging, radiomics, and artificial intelligence of HE, in order to help identify high-risk patients for HE in clinical practice.

Integrating IASLC grading and radiomics for predicting postoperative outcomes in stage IA invasive lung adenocarcinoma

BACKGROUND

The International Association for the Study of Lung Cancer (IASLC) Pathology Committee introduced a histologic grading system for invasive lung adenocarcinoma (LUAD) in 2020. The IASLC grading system, hinging on the evaluation of predominant and high-grade histologic patterns, has proven to be practical and prognostic for invasive LUAD. However, there are still limitations in evaluating the prognosis of stage IA LUAD. Radiomics may serve as a valuable complement.

PURPOSE

To establish a model that integrates IASLC grading and radiomics, aimed at predicting the prognosis of stage IA LUAD.

METHODS

We conducted a retrospective analysis of 628 patients diagnosed with stage IA LUAD who underwent surgical resection between January 2015 and December 2018 at our institution. The patients were randomly divided into the training set (n = 439) and testing set (n = 189) at a ratio of 7:3. Overall survival (OS) and disease-free survival (DFS) were taken as the end points. Radiomics features were obtained by PyRadiomics. Feature selection was performed using the least absolute shrinkage and selection operator (LASSO). The prediction models for OS and DFS were developed using multivariate Cox regression analysis, and the models were visualized through nomogram plots. The model's performance was evaluated using area under the curves (AUC), concordance index (C-index), calibration curves, and survival decision curve analysis (DCA).

RESULTS

In total, nine radiomics features were selected for the OS prediction model, and 15 radiomics features were selected for the DFS prediction model. Patients with high radiomics scores were associated with a worse prognosis (p < 0.001). We built separate prediction models using radiomics or IASLC alone, as well as a combined prediction model. In the prediction of OS, we observed that the combined model (C-index: 0.812 ± 0.024, 3 years AUC: 0.692, 5 years AUC: 0.792) achieved superior predictive performance than the radiomics (C-index: 0.743 ± 0.038, 3 years AUC: 0.633, 5 years AUC: 0.768) and IASLC grading (C-index: 0.765 ± 0.042, 3 years AUC: 0.658, 5 years AUC: 0.743) models alone. Similar results were obtained in the models for DFS.

CONCLUSION

The combination of radiomics and IASLC pathological grading proves to be an effective approach for predicting the prognosis of stage IA LUAD. This has substantial clinical relevance in guiding treatment decisions for early-stage LUAD.

Role of peritumoral tissue analysis in predicting characteristics of hepatocellular carcinoma using ultrasound-based radiomics

Predicting the biological characteristics of hepatocellular carcinoma (HCC) is essential for personalized treatment. This study explored the role of ultrasound-based radiomics of peritumoral tissues for predicting HCC features, focusing on differentiation, cytokeratin 7 (CK7) and Ki67 expression, and p53 mutation status. A cohort of 153 patients with HCC underwent ultrasound examinations and radiomics features were extracted from peritumoral tissues. Subgroups were formed based on HCC characteristics. Predictive modeling was carried out using the XGBOOST algorithm in the differentiation subgroup, logistic regression in the CK7 and Ki67 expression subgroups, and support vector machine learning in the p53 mutation status subgroups. The predictive models demonstrated robust performance, with areas under the curves of 0.815 (0.683-0.948) in the differentiation subgroup, 0.922 (0.785-1) in the CK7 subgroup, 0.762 (0.618-0.906) in the Ki67 subgroup, and 0.849 (0.667-1) in the p53 mutation status subgroup. Confusion matrices and waterfall plots highlighted the good performance of the models. Comprehensive evaluation was carried out using SHapley Additive exPlanations plots, which revealed notable contributions from wavelet filter features. This study highlights the potential of ultrasound-based radiomics, specifically the importance of peritumoral tissue analysis, for predicting HCC characteristics. The results warrant further validation of peritumoral tissue radiomics in larger, multicenter studies.

Deep learning-based multi-model prediction for disease-free survival status of patients with clear cell renal cell carcinoma after surgery: a multicenter cohort study

BACKGROUND

Although separate analysis of individual factor can somewhat improve the prognostic performance, integration of multimodal information into a single signature is necessary to stratify patients with clear cell renal cell carcinoma (ccRCC) for adjuvant therapy after surgery.

METHODS

A total of 414 patients with whole slide images, computed tomography images, and clinical data from three patient cohorts were retrospectively analyzed. The authors performed deep learning and machine learning algorithm to construct three single-modality prediction models for disease-free survival of ccRCC based on whole slide images, cell segmentation, and computed tomography images, respectively. A multimodel prediction signature (MMPS) for disease-free survival were further developed by combining three single-modality prediction models and tumor stage/grade system. Prognostic performance of the prognostic model was also verified in two independent validation cohorts.

RESULTS

Single-modality prediction models performed well in predicting the disease-free survival status of ccRCC. The MMPS achieved higher area under the curve value of 0.742, 0.917, and 0.900 in three independent patient cohorts, respectively. MMPS could distinguish patients with worse disease-free survival, with HR of 12.90 (95% CI: 2.443-68.120, P <0.0001), 11.10 (95% CI: 5.467-22.520, P <0.0001), and 8.27 (95% CI: 1.482-46.130, P <0.0001) in three different patient cohorts. In addition, MMPS outperformed single-modality prediction models and current clinical prognostic factors, which could also provide complements to current risk stratification for adjuvant therapy of ccRCC.

CONCLUSION

Our novel multimodel prediction analysis for disease-free survival exhibited significant improvements in prognostic prediction for patients with ccRCC. After further validation in multiple centers and regions, the multimodal system could be a potential practical tool for clinicians in the treatment for ccRCC patients.

Predicting postoperative rehemorrhage in hypertensive intracerebral hemorrhage using noncontrast CT radiomics and clinical data with an interpretable machine learning approach

In hypertensive intracerebral hemorrhage (HICH) patients, while emergency surgeries effectively reduce intracranial pressure and hematoma volume, their significant risk of causing postoperative rehemorrhage necessitates early detection and management to improve patient prognosis. This study sought to develop and validate machine learning (ML) models leveraging clinical data and noncontrast CT radiomics to pinpoint patients at risk of postoperative rehemorrhage, equipping clinicians with an early detection tool for prompt intervention. The study conducted a retrospective analysis on 609 HICH patients, dividing them into training and external verification cohorts. These patients were categorized into groups with and without postoperative rehemorrhage. Radiomics features from noncontrast CT images were extracted, standardized, and employed to create several ML models. These models underwent internal validation using both radiomics and clinical data, with the best model's feature significance assessed via the Shapley additive explanations (SHAP) method, then externally validated. In the study of 609 patients, postoperative rehemorrhage rates were similar in the training (18.8%, 80/426) and external verification (17.5%, 32/183) cohorts. Six significant noncontrast CT radiomics features were identified, with the support vector machine (SVM) model outperforming others in both internal and external validations. SHAP analysis highlighted five critical predictors of postoperative rehemorrhage risk, encompassing three radiomics features from noncontrast CT and two clinical data indicators. This study highlights the effectiveness of an SVM model combining radiomics features from noncontrast CT and clinical parameters in predicting postoperative rehemorrhage among HICH patients. This approach enables timely and effective interventions, thereby improving patient outcomes.

Predicting Microwave Ablation Early Efficacy in Pulmonary Malignancies via Δ Radiomics Models

OBJECTIVE

This study aimed to explore the value of preoperative and postoperative computed tomography (CT)-based radiomic signatures and Δ radiomic signatures for evaluating the early efficacy of microwave ablation (MWA) for pulmonary malignancies.

METHODS

In total, 115 patients with pulmonary malignancies who underwent MWA treatment were categorized into response and nonresponse groups according to relevant guidelines and consensus. Quantitative image features of the largest pulmonary malignancies were extracted from CT noncontrast scan images preoperatively (time point 0, TP0) and immediately postoperatively (time point 1, TP1). Critical features were selected from TP0 and TP1 and as Δ radiomics signatures for building radiomics models. In addition, a combined radiomics model (C-RO) was developed by integrating radiomics parameters with clinical risk factors. Prediction performance was assessed using the area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA).

RESULTS

The radiomics model using Δ features outperformed the radiomics model using TP0 and TP1 features, with training and validation AUCs of 0.892, 0.808, and 0.787, and 0.705, 0.825, and 0.778, respectively. By combining the TP0, TP1, and Δ features, the logistic regression model exhibited the best performance, with training and validation AUCs of 0.945 and 0.744, respectively. The DCA confirmed the clinical utility of the Δ radiomics model.

CONCLUSIONS

A combined prediction model, including TP0, TP1, and Δ radiometric features, can be used to evaluate the early efficacy of MWA in pulmonary malignancies.

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

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

Artificial Intelligence-Based Treatment Decisions: A New Era for NSCLC

Non-small cell lung cancer (NSCLC) is the leading cause of cancer-related mortality among women and men, in developed countries, despite the public health interventions including tobacco-free campaigns, screening and early detection methods, recent therapeutic advances, and ongoing intense research on novel antineoplastic modalities. Targeting oncogenic driver mutations and immune checkpoint inhibition has indeed revolutionized NSCLC treatment, yet there still remains the unmet need for robust and standardized predictive biomarkers to accurately inform clinical decisions. Artificial intelligence (AI) represents the computer-based science concerned with large datasets for complex problem-solving. Its concept has brought a paradigm shift in oncology considering its immense potential for improved diagnosis, treatment guidance, and prognosis. In this review, we present the current state of AI-driven applications on NSCLC management, with a particular focus on radiomics and pathomics, and critically discuss both the existing limitations and future directions in this field. The thoracic oncology community should not be discouraged by the likely long road of AI implementation into daily clinical practice, as its transformative impact on personalized treatment approaches is undeniable.

CCL18, CHI3L1, ANG2, IL-6 systemic levels are associated with the extent of lung damage and radiomic features in SARS-CoV-2 infection

OBJECTIVE AND DESIGN

We aimed to identify cytokines whose concentrations are related to lung damage, radiomic features, and clinical outcomes in COVID-19 patients.

MATERIAL OR SUBJECTS

Two hundred twenty-six patients with SARS-CoV-2 infection and chest computed tomography (CT) images were enrolled.

METHODS

CCL18, CHI3L1/YKL-40, GAL3, ANG2, IP-10, IL-10, TNFα, IL-6, soluble gp130, soluble IL-6R were quantified in plasma samples using Luminex assays. The Mann-Whitney U test, the Kruskal-Wallis test, correlation and regression analyses were performed. Mediation analyses were used to investigate the possible causal relationships between cytokines, lung damage, and outcomes. AVIEW lung cancer screening software, pyradiomics, and XGBoost classifier were used for radiomic feature analyses.

RESULTS

CCL18, CHI3L1, and ANG2 systemic levels mainly reflected the extent of lung injury. Increased levels of every cytokine, but particularly of IL-6, were associated with the three outcomes: hospitalization, mechanical ventilation, and death. Soluble IL-6R showed a slight protective effect on death. The effect of age on COVID-19 outcomes was partially mediated by cytokine levels, while CT scores considerably mediated the effect of cytokine levels on outcomes. Radiomic-feature-based models confirmed the association between lung imaging characteristics and CCL18 and CHI3L1.

CONCLUSION

Data suggest a causal link between cytokines (risk factor), lung damage (mediator), and COVID-19 outcomes.

Construction of disulfidptosis-based immune response prediction model with artificial intelligence and validation of the pivotal grouping oncogene c-MET in regulating T cell exhaustion

Background

Given the lack of research on disulfidptosis, our study aimed to dissect its role in pan-cancer and explore the crosstalk between disulfidptosis and cancer immunity.

Methods

Based on TCGA, ICGC, CGGA, GSE30219, GSE31210, GSE37745, GSE50081, GSE22138, GSE41613, univariate Cox regression, LASSO regression, and multivariate Cox regression were used to construct the rough gene signature based on disulfidptosis for each type of cancer. SsGSEA and Cibersort, followed by correlation analysis, were harnessed to explore the linkage between disulfidptosis and cancer immunity. Weighted correlation network analysis (WGCNA) and Machine learning were utilized to make a refined prognosis model for pan-cancer. In particular, a customized, enhanced prognosis model was made for glioma. The siRNA transfection, FACS, ELISA, etc., were employed to validate the function of c-MET.

Results

The expression comparison of the disulfidptosis-related genes (DRGs) between tumor and nontumor tissues implied a significant difference in most cancers. The correlation between disulfidptosis and immune cell infiltration, including T cell exhaustion (Tex), was evident, especially in glioma. The 7-gene signature was constructed as the rough model for the glioma prognosis. A pan-cancer suitable DSP clustering was made and validated to predict the prognosis. Furthermore, two DSP groups were defined by machine learning to predict the survival and immune therapy response in glioma, which was validated in CGGA. PD-L1 and other immune pathways were highly enriched in the core blue gene module from WGCNA. Among them, c-MET was validated as a tumor driver gene and JAK3-STAT3-PD-L1/PD1 regulator in glioma and T cells. Specifically, the down-regulation of c-MET decreased the proportion of PD1+ CD8+ T cells.

Conclusion

To summarize, we dissected the roles of DRGs in the prognosis and their relationship with immunity in pan-cancer. A general prognosis model based on machine learning was constructed for pan-cancer and validated by external datasets with a consistent result. In particular, a survival-predicting model was made specifically for patients with glioma to predict its survival and immune response to ICIs. C-MET was screened and validated for its tumor driver gene and immune regulation function (inducing t-cell exhaustion) in glioma.

Developing a clinical-radiomic prediction model for 3-year cancer-specific survival in lung cancer patients treated with stereotactic body radiation therapy

PURPOSE

The study aims to develop and validate a combined model for predicting 3-year cancer-specific survival (CSS) in lung cancer patients treated with stereotactic body radiation therapy (SBRT) by integrating clinical and radiomic parameters.

METHODS

Clinical data and pre-treatment CT images were collected from 102 patients treated with lung SBRT. Multivariate logistic regression and the least absolute shrinkage and selection operator were used to determine the clinical and radiomic factors associated with 3-year CSS. Three prediction models were developed using clinical factors, radiomic factors, and a combination of both. The performance of the models was assessed using receiver operating characteristic curve and calibration curve. A nomogram was also created to visualize the 3-year CSS prediction.

RESULTS

With a 36-month follow-up, 40 patients (39.2%) died of lung cancer and 62 patients (60.8%) survived. Three clinical factors, including gender, clinical stage, and lymphocyte ratio, along with three radiomic features, were found to be independent factors correlated with 3-year CSS. The area under the curve values for the clinical, radiomic, and combined model were 0.839 (95% CI 0.735-0.914), 0.886 (95% CI 0.790-0.948), and 0.914 (95% CI 0.825-0.966) in the training cohort, and 0.757 (95% CI 0.580-0.887), 0.818 (95% CI 0.648-0.929), and 0.843 (95% CI 0.677-0.944) in the validation cohort, respectively. Additionally, the calibration curve demonstrated good calibration performance and the nomogram created from the combined model showed potential for clinical utility.

CONCLUSION

A clinical-radiomic model was developed to predict the 3-year CSS for lung cancer patients treated with SBRT.

A pairwise radiomics algorithm-lesion pair relation estimation model for distinguishing multiple primary lung cancer from intrapulmonary metastasis

Background

Distinguishing multiple primary lung cancer (MPLC) from intrapulmonary metastasis (IPM) is critical for their disparate treatment strategy and prognosis. This study aimed to establish a non-invasive model to make the differentiation pre-operatively.

Methods

We retrospectively studied 168 patients with multiple lung cancers (307 pairs of lesions) including 118 cases for modeling and internal validation, and 50 cases for independent external validation. Radiomic features on computed tomography (CT) were extracted to calculate the absolute deviation of paired lesions. Features were then selected by correlation coefficients and random forest classifier 5-fold cross-validation, based on which the lesion pair relation estimation (PRE) model was developed. A major voting strategy was used to decide diagnosis for cases with multiple pairs of lesions. Cases from another institute were included as the external validation set for the PRE model to compete with two experienced clinicians.

Results

Seven radiomic features were selected for the PRE model construction. With major voting strategy, the mean area under receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the training versus internal validation versus external validation cohort to distinguish MPLC were 0.983 versus 0.844 versus 0.793, 0.942 versus 0.846 versus 0.760, 0.905 versus 0.728 versus 0.727, and 0.962 versus 0.910 versus 0.769, respectively. AUCs of the two clinicians were 0.619 and 0.580.

Conclusions

The CT radiomic feature-based lesion PRE model is potentially an accurate diagnostic tool for the differentiation of MPLC and IPM, which could help with clinical decision making.

Non-linear modifications enhance prediction of pathological response to pre-operative PD-1 blockade in lung cancer: A longitudinal hybrid radiological model

Major pathologic remission (MPR, residual tumor <10%) is a promising clinical endpoint for prognosis analysis in patients with lung cancer receiving pre-operative PD-1 blockade therapy. Most of the current biomarkers for predicting MPR such as PD-L1 and tumor mutation burden (TMB) need to be obtained invasively. They cannot overcome the spatiotemporal heterogeneity or provide dynamic monitoring solutions. Radiomics and artificial intelligence (AI) models provide a practical tool enabling non-invasive follow-up observation of tumor structural information through high-throughput data analysis. Currently, AI-based models mainly focus on the single baseline scan or pipeline, namely sole radiomics or deep learning (DL). This work merged the delta-radiomics based on the slope of classic radiomics indexes within a time interval and the features extracted by deep networks from the subtraction between the baseline and follow-up images. The subtracted images describing the tumor changes were based on the transformation generated by registration. Stepwise optimization of components was performed by repeating experiments among various combinations of DL networks, registration methods, feature selection algorithms, and classifiers. The optimized model could predict MPR with a cross-validation AUC of 0.91 and an external validation AUC of 0.85. A core set of 27 features (eight classic radiomics, 15 delta-radiomics, one classic DL features, and three delta-DL features) was identified. The changes in delta-radiomics indexes during the treatment were fitted with mathematic models. The fitting results revealed that over half of the features were of non-linear dynamics. Therefore, non-linear modifications were made on eight features by replacing the original features with non-linear fitting parameters, and the modified model achieved an improved power. The dynamic hybrid model serves as a novel and promising tool to predict the response of lesions to PD-1 blockade, which implies the importance of introducing the non-linear dynamic effects and DL approaches to the original delta-radiomics in the future.

Artificial Intelligence and Lung Cancer: Impact on Improving Patient Outcomes

Lung cancer remains one of the leading causes of cancer-related deaths worldwide, emphasizing the need for improved diagnostic and treatment approaches. In recent years, the emergence of artificial intelligence (AI) has sparked considerable interest in its potential role in lung cancer. This review aims to provide an overview of the current state of AI applications in lung cancer screening, diagnosis, and treatment. AI algorithms like machine learning, deep learning, and radiomics have shown remarkable capabilities in the detection and characterization of lung nodules, thereby aiding in accurate lung cancer screening and diagnosis. These systems can analyze various imaging modalities, such as low-dose CT scans, PET-CT imaging, and even chest radiographs, accurately identifying suspicious nodules and facilitating timely intervention. AI models have exhibited promise in utilizing biomarkers and tumor markers as supplementary screening tools, effectively enhancing the specificity and accuracy of early detection. These models can accurately distinguish between benign and malignant lung nodules, assisting radiologists in making more accurate and informed diagnostic decisions. Additionally, AI algorithms hold the potential to integrate multiple imaging modalities and clinical data, providing a more comprehensive diagnostic assessment. By utilizing high-quality data, including patient demographics, clinical history, and genetic profiles, AI models can predict treatment responses and guide the selection of optimal therapies. Notably, these models have shown considerable success in predicting the likelihood of response and recurrence following targeted therapies and optimizing radiation therapy for lung cancer patients. Implementing these AI tools in clinical practice can aid in the early diagnosis and timely management of lung cancer and potentially improve outcomes, including the mortality and morbidity of the patients.