Conventional response evaluation criteria primarily rely on tumor size reduction to assess the efficacy of neoadjuvant therapy (NAT) in esophageal cancer. However, these morphological assessments often fail to capture the underlying biological changes, as some patients achieve substantial pathological response without marked tumor shrinkage. To address this limitation, we introduced the concept of tumor burden (TB), integrating both tumor volume (TV) and tumor regression grade (TRG), and proposed four novel parameters—TV change ratio (TVCR), pre-NAT mean cross-sectional area (Pre-NT MCSA), post-NAT MCSA (Post-NT MCSA), and post-NAT mean cross-sectional TB (Post-NT MCTB)—to better quantify residual disease. We conducted a retrospective study of 204 esophageal cancer patients who underwent NAT followed by minimally invasive esophagectomy between September 2019 and September 2024. The relationships between the four novel parameters and resection status were evaluated. Subsequently, a radiomics-based random forest model was developed to predict post-NAT TB (Post-NT TB) from pre- and post-NAT CT images. Among the four parameters, only Post-NT MCTB showed a significant association with resection status (p < 0.01), confirming its clinical relevance in reflecting residual TB. The radiomics-based model demonstrated robust predictive performance, achieving a coefficient of determination (R²) of 0.85 in the test set. Post-NT MCTB serves as a clinically relevant indicator of residual TB after NAT, surpassing traditional volume-based assessment. This radiomics-based framework enables dynamic and quantitative evaluation that may assist preoperative planning in esophageal cancer.

Locally advanced head and neck squamous cell carcinoma (HNSCC) exhibits significant heterogeneity to neoadjuvant targeted therapy and chemotherapy, making personalized treatment selection challenging. This study aims to develop and validate a Transformer-based multimodal fusion model based on multimodal magnetic resonance imaging (MRI) and pathological whole slide image (WSI) to improve the prediction of neoadjuvant targeted therapy and chemotherapy response in locally advanced HNSCC. A total of 201 patients with stage III-IV HNSCC receiving neoadjuvant targeted therapy and chemotherapy was recruited from two medical centers. For feature extraction: Macro-level imaging features were extracted from T1WI, T2WI, and contrast-enhanced T1WI (CE-T1WI) using a ResNet50-based deep learning model; Micro-level cellular features were extracted from hematoxylin and eosin (H E)-stained WSIs via Term Frequency-Inverse Document Frequency (TF-IDF) analysis, which aggregates patch-level pathological information into slide-level representations. Based on a Transformer fusion framework with multi-head self-attention mechanisms, the multimodal fusion model dynamically weights and fuses cross-scale features across modalities. The predictive performance of each model was evaluated using the area under the curve (AUC), calibration curve, and decision curve analysis (DCA). The visualization of deep learning model was utilized to enhance interpretability. The multimodal fusion model outperformed single-modal models in predicting the overall response rate (ORR) of targeted therapy and chemotherapy in locally advanced HNSCC, with AUC values of 0.862 (95

IntroductionTo explore the value of enhanced computed tomography (CT) -derived extracellular volume (ECV) combined with systemic immune-inflammation index (SII) in predicting tumor budding (TB) grading of rectal cancer.Materials and MethodsThe clinical and imaging data of 177 rectal cancer patients were retrospectively analyzed, and we divided them into a low-grade and medium-high group according to pathological TB count. ECV and SII values between the two groups were compared. Intra-class correlation coefficient (ICC) was used to detect the consistency of measurements among observers. Binary logistic regression was used to analyze the correlations between variables and TB grading of rectal cancer. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic efficiency of statistically significant parameters and their combination. Area under the curve (AUC), its 95% confidence interval, and the corresponding Youden index, sensitivity, and specificity were calculated.ResultsAmong the 177 rectal cancer patients, 108 were low-grade and 69 were medium-high grade. ECV values measured by two physicians showed good consistency (ICC = 0.98). ECV value of low-grade (21.76% ± 4.89%) was lower than that of medium-high grade TB group (27.91% ± 4.77%) (P < .001). SII value was lower in low-grade group (492.14 ± 239.56) than in medium-high grade TB group (825.02 ± 529.38). In the multivariate analysis, ECV value [odds ratio (OR): 1.339 (95% CI: 1.194-1.502)] and SII value [OR: 1.004 (95% CI: 1.002-1.005)] were independent risk factors for predicting TB grading. In the training set, AUCs of ECV, SII, and their combination in evaluating TB grading of rectal cancer were 0.838 (95% CI: 0.760-0.905), 0.755 (95% CI: 0.663-0.829), and 0.889 (95% CI: 0.832-0.943), respectively. In the test set, the corresponding AUCs were 0.741 (95% CI: 0.626-0.870), 0.716 (95% CI: 0.554-0.849), and 0.815 (95% CI: 0.711-0.913). Decision curve analysis (DCA) showed that the combination had higher clinical value than using ECV or SII alone.ConclusionThe combination of ECV and SII can non-invasively evaluate TB grading of rectal cancer before surgery, potentially providing a reference for preoperative risk stratification as a decision-support tool.

RATIONALE AND OBJECTIVES:To investigate the ability of a radiomics model of T2-weighted imaging (T2WI) tumor and peritumoral regions for predicting noninvasively tumor budding grading in rectal patients. MATERIALS AND METHODS:This retrospective study included 478 patients diagnosed with rectal cancer from three different institutions and randomly divided into a training cohort (n=309), an internal testing cohort (n=78), and an external validation cohort (n=91). The model that achieved the balance between discrimination and stability in the training and internal testing cohort was selected as the optimal peritumor model. A nomogram was constructed by an ensemble model combined peritumoral model with tumoral model and independent clinical predictor. The Receiver Operating Characteristic curve (ROC), the Area Under the ROC curve (AUC), decision curve analysis (DCA), calibration analysis, and the DeLong test were adopted to evaluate the performance of models. The Shapley Additive Explanations (SHAP) algorithm was used to visually interpret the weights of each feature that constructed the models. RESULTS:The peritumoral model constructed by features extracted from the 5 mm extension distance achieved the balance between discrimination and stability in the training and internal test cohort, with AUC values of 0.841 and 0.761, respectively. The ensemble model showed certain discriminative ability in the external validation cohort, with AUC of 0.853. CONCLUSION:The ensemble model that integrated radiomics features from tumor and peritumoral regions and independent clinical predictor can serve as a promising noninvasive assessment tool for clinicians to predict TB grading in rectal cancer patients preoperatively.

Introduction Lymphovascular invasion (LVI), an aggressive pathological manifestation of breast cancer, is closely associated with increased risk of distant metastasis and poor prognosis. This study proposes a novel modeling strategy that integrates MRI-derived microvascular atlas parameters with the TwinsSVT deep learning architecture to enable noninvasive prediction of LVI status in breast cancer patients and to explore its biological interpretability.Materials and Methods A total of 436 breast cancer patients from two medical centers, all pathologically confirmed postoperatively, were retrospectively enrolled. All patients underwent high-resolution multi-b-value diffusion-weighted imaging (DWI) prior to surgery. From the MRI data, four types of microvascular simulation parameter maps were reconstructed within tumor regions: apparent diffusion coefficient (ADC), mean flow velocity (v_m), velocity dispersion (v_s), and angiographic branching index (ANB), aiming to characterize intratumoral microcirculation and vascular structural complexity. These functional parametric maps were individually input into separate encoder branches of the TwinsSVT model to extract multi-scale spatial features. A multi-layer Transformer fusion module was then employed to capture structural interactions across modalities, thereby constructing a multi-parametric fusion model. Model performance was evaluated using metrics including area under the curve (AUC) and F1 score.Results Compared with single-parameter models, the multi-parametric fusion model demonstrated significantly improved predictive performance, with AUCs of 0.881 (95% CI: 0.781-0.982) and 0.859 (95% CI: 0.764-0.953) in internal and external validation cohorts, respectively. Grad-CAM visualizations revealed that the model predominantly focused on tumor margins and regions of high vascular density, suggesting a strong correlation between the model's attention and actual pathological structures.Conclusion The deep learning model constructed based on MRI-derived microvascular simulation atlases enables noninvasive preoperative prediction of LVI status in breast cancer patients. By effectively capturing structural information and offering biological interpretability, the model holds promise as a robust imaging-based tool for precision subtyping and clinical decision support.

OBJECTIVE:Deep learning signatures (DLS) extracted from CT images can noninvasively reflect tumor heterogeneity and have shown promise in prognostic modeling for esophageal squamous cell carcinoma (ESCC). To develop and validate a CT-based DL model combined with nutritional biomarkers to predict 3-year overall survival (OS) in ESCC, and to investigate transcriptomic differences between DLS-based risk groups. MATERIALS AND METHODS:This retrospective multicenter study included 662 postoperative ESCC patients from three hospitals and 16 additional patients from The Cancer Genome Atlas (TCGA). DL features extraction from CT images based on the Crossformer architecture. Skeletal muscle index was measured at the L3 vertebra to assess low skeletal muscle mass (LSMM). Cox regression was used to build clinical, DL, and combined models. Model performance was evaluated using the concordance index (C-index). Transcriptomic analysis of the TCGA cohort was performed to identify metabolic pathway differences between DLS-based risk groups. RESULTS:The DL model achieved a C-index of 0.743 (95% CI: 0.683-0.803) in the internal validation cohort and 0.692 (95% CI: 0.576-0.809) in the external cohort. Pathological T and N stages, Neuroaggression, Vascular invasion, and LSMM were identified as independent clinical predictors. The combined model achieved a C-index of 0.753 (95% CI: 0.697-0.808) internally and 0.725 (95% CI: 0.613-0.838) externally. DLS-based risk stratification revealed significant differences in metabolic activity between groups, supporting its biological relevance. CONCLUSION:The combined model enables preoperative OS prediction in ESCC. DLS-based stratification reflects transcriptomic metabolic heterogeneity and enhances the biological interpretability of imaging features. CRITICAL RELEVANCE STATEMENT:This study developed a CT-based DLS and combined it with nutritional markers for prognostic modeling in ESCC. Transcriptomic analysis of DLS-based groups revealed metabolic heterogeneity, enhancing the biological interpretability of the DL model. KEY POINTS:A combined DLS and nutritional model enables individualized preoperative survival prediction in ESCC. DLS-based risk groups defined by the DLS exhibited transcriptomic differences in key metabolic pathways, revealing biological underpinnings of imaging-based phenotypes. Attention map visualization revealed consistent spatial focus on morphologically distinct tumor regions, enhancing the interpretability of deep learning predictions.

BACKGROUND The most common complication in D2 gastrectomy for gastric cancer is the anastomotic stricture, with reported incidence between 3% and 15%. Appropriate tissue perfusion is essential at all time points during the inflammatory, proliferative, and remodeling phases of anastomotic healing. Portal venous phase computed tomography (CT), a study universally indicated for preoperative staging and postoperative re-evaluation, is an objective assessment of visceral hemodynamic status that may impact anastomotic healing. Yet the predictive value of preoperative CT portal venous parameters for risk of anastomotic stricture has not yet been investigated, and reliable preoperative stratification tools are urgently needed. Deep learning-based portal venous CT parameters have been particularly well-studied for liver cirrhosis and portal hypertension, but this is the first time they have been studied in predicting gastrointestinal anastomotic complications. AIM To prospectively explore the predictive power of preoperative enhanced CT portal venous phase quantitative parameters for anastomotic stricture after D2 gastrectomy for gastric cancer to establish a combined prediction model. METHODS Clinical data from 199 patients who underwent D2 gastrectomy for gastric cancer at our institution between January 2022 and June 2024 were retrospectively analyzed. Patients were categorized into stricture group (n = 23) and non-stricture group (n = 176) based on whether anastomotic stricture occurred within 12 months postoperatively. On preoperative enhanced CT portal venous phase images, main portal vein (PV) CT value (PV-HU), splenic vein CT value, hepatic parenchyma, and abdominal aorta were measured. Derived parameters including normalized PV-HU (nPV-HU) and PV-to-abdominal aorta CT value (aorta-HU) ratio were calculated. Multivariate logistic regression analysis was employed to identify independent predictive factors. Receiver operating characteristic curves were constructed to evaluate predictive performance, and a combined prediction model was developed. RESULTS The postoperative anastomotic stricture rate was 11.6% (23/199). The stricture group demonstrated significantly lower PV-HU, splenic vein CT value, PV-to-aorta-HU ratio, and nPV-HU compared to the non-stricture group (P < 0.05). Multivariate logistic regression analysis revealed that body mass index >= 25 kg/m2, diabetes mellitus, total gastrectomy, 25 mm stapler diameter, decreased nPV-HU, and decreased PV-to-aorta-HU ratio were independent risk factors for anastomotic stricture (P < 0.05). For predicting anastomotic stricture, nPV-HU yielded an area under the curve (AUC) of 0.812, with an optimal cutoff value of 42.5%, sensitivity of 78.3%, and specificity of 72.7%; PV-to-aorta-HU ratio yielded an AUC of 0.768, with an optimal cutoff value of 1.45, sensitivity of 73.9%, and specificity of 71.0%. The combined prediction model incorporating clinical factors and CT parameters achieved an AUC of 0.893, with sensitivity of 87.0%, specificity of 80.1%, and negative predictive value of 97.9%. CONCLUSION Preoperative enhanced CT portal venous phase quantitative parameters nPV-HU and PV-to-aorta-HU ratio can effectively predict the risk of anastomotic stricture following D2 gastrectomy for gastric cancer. The prediction model combining clinical risk factors demonstrates high diagnostic performance and provides a strong imaging-based rationale for preoperative risk stratification, potentially guiding personalized perioperative management strategies.

Abstract Background Immunotherapy has revolutionized the treatment landscape for head and neck squamous cell carcinoma (HNSCC) and PD-L1 combined positivity score (CPS) scoring is recommended as a biomarker for immunotherapy. Therefore, this study aimed to develop an MRI-based deep learning score (DLS) to non-invasively assess PD-L1 expression status in HNSCC patients and evaluate its potential effeciency in predicting prognostic stratification following treatment with immune checkpoint inhibitors (ICI). Methods In this study, we collected data from four patient cohorts comprising a total of 610 HNSCC patients from two separate institutions. We developed deep learning models based on the ResNet-101 convolutional neural network to analyze three MRI sequences (T1WI, T2WI, and contrast-enhanced T1WI). Tumor regions were manually segmented, and features extracted from different MRI sequences were fused using a transformer-based model incorporating attention mechanisms. The model’s performance in predicting PD-L1 expression was evaluated using the area under the curve (AUC), sensitivity, specificity, and calibration metrics. Survival analyses were conducted using Kaplan-Meier survival curves and log-rank tests to evaluate the prognostic significance of the DLS. Results The DLS demonstrated high predictive accuracy for PD-L1 expression, achieving an AUC of 0.981, 0.860 and 0.803 in the training, internal and external validation cohort. Patients with higher DLS scores demonstrated significantly improved progression-free survival (PFS) in both the internal validation cohort (hazard ratio: 0.491; 95% CI, 0.270–0.892; P = 0.005) and the external validation cohort (hazard ratio: 0.617; 95% CI, 0.391–0.973; P = 0.040). In the ICI-treated cohort, the DLS achieved an AUC of 0.739 for predicting durable clinical benefit (DCB). Conclusions The proposed DLS offered a non-invasive and accurate approach for assessing PD-L1 expression in patients with HNSCC and effectively stratified HNSCC patients to benefit from immunotherapy based on PFS.

RATIONALE AND OBJECTIVES:Accurate determination of human epidermal growth factor receptor 2 (HER2) expression is critical for guiding targeted therapy in breast cancer. This study aimed to develop and validate a deep learning (DL)-based decision-making visual biomarker system (DM-VBS) for predicting HER2 status using radiomics and DL features derived from magnetic resonance imaging (MRI) and mammography (MG). MATERIALS AND METHODS:Radiomics features were extracted from MRI, and DL features were derived from MG. Four submodels were constructed: Model I (MRI-radiomics) and Model III (mammography-DL) for distinguishing HER2-zero/low from HER2-positive cases, and Model II (MRI-radiomics) and Model IV (mammography-DL) for differentiating HER2-zero from HER2-low/positive cases. These submodels were integrated into a XGBoost model for ternary classification of HER2 status. Radiologists assessed imaging features associated with HER2 expression, and model performance was validated using two independent datasets from The Cancer Image Archive. RESULTS:A total of 550 patients were divided into training, internal validation, and external validation cohorts. Models I and III achieved an area under the curve (AUC) of 0.800-0.850 for distinguishing HER2-zero/low from HER2-positive cases, while Models II and IV demonstrated AUC values of 0.793-0.847 for differentiating HER2-zero from HER2-low/positive cases. The DM-VBS achieved average accuracy of 85.42%, 80.4%, and 89.68% for HER2-zero, -low, and -positive patients in the validation cohorts, respectively. Imaging features such as lesion size, number of lesions, enhancement type, and microcalcifications significantly differed across HER2 statuses, except between HER2-zero and -low groups. CONCLUSION:DM-VBS can predict HER2 status and assist clinicians in making treatment decisions for breast cancer.

Background:The treatment of locally advanced oesophageal squamous cell carcinoma (LAESCC) without distant metastasis remains a subject of debate. Neoadjuvant immunochemotherapy (NIC) combined with surgery is the preferred initial approach for managing LAESCC. However, information on the clinical efficacy and survival of patients with LAESCC treated with NIC followed by surgery is limited. Methods:This retrospective analysis aimed to identify predictors NIC treatment effectiveness and on patient survival. We developed a Cox proportional hazards model and Kaplan-Meier curve to estimate progression-free survival (PFS) and overall survival (OS) following NIC treatment and surgery. Results:Overall, 225 patients with LAESCC were divided into training (157) and test set (68) (7:3). After a median follow-up of 2.86 years, death was observed as a positive event in 41 patients (26.1%). It is statistically significant to construct a prediction model combining radiomics features pre- and post-NIC with clinical features to predict the PFS and OS of LAESCC. The combined model showed the highest performance in predicting both disease-free survival and OS compared with the clinical or radiomics models. multivariate Cox regression analysis identified smoking (HR = 1.417, 95% confidence interval [CI]: 0.875-2.293, p = 0.156), Ki67(HR = 2.426, 95% confidence interval [CI]: 1.506-3.908, p = 0.000) and postRad-S1 (HR = 1.867, 95% CI: 1.053-3.311, p = 0.033) as significant independent covariates associated with high PFS. While Ki67 and postRad-S2 were prognostic factors significantly associated with OS (HR = 1.521, 95% CI: 0.821-2.818, p = 0.183; HR = 1.912, 95% CI: 1.001-3.654, p = 0.050, respectively). Conclusion:For patients with LAESCC treated with NIC followed by surgery, the combined model effectively evaluated the efficacy of NIC and predicted PFS and OS. Additionally, different independent predictors were associated with PFS and OS, providing clues for future studies.

Introduction To evaluate the predictive efficacy of enhanced Computed Tomograph(CT) radiomics combined with clinical features for assessing treatment response to neoadjuvant chemotherapy plus immunotherapy in esophageal squamous cell carcinoma (ESCC) patients. Methods We retrospectively analyzed 189 pathologically confirmed esophageal squamous cell carcinoma patients (treated between January 2020 and October 2024) who underwent neoadjuvant chemoimmunotherapy. Patients were stratified into remission and non-remission groups based on pathological response and randomly divided into training (n = 114) and testing (n = 75) sets (6:4 ratio). Clinical predictors were identified using logistic regression to construct a clinical model. Radiomic features were extracted from manually delineated tumor regions on contrast-enhanced CT scans, and a radiomics model was developed. A combined model integrating clinical variables and radiomics probabilities was then built and presented as a nomogram. Model performance was assessed using receiver operating characteristic (ROC) curves (AUC, Area Under the Curve) comparison via Delong test), calibration curves, and decision curve analysis (DCA). Results Multivariable analysis identified treatment cycle number as a significant clinical predictor. Ten radiomic features were selected for the final model. In the training set, the clinical model achieved an AUC of 0.705 (95% CI 0.607-0.802), while the radiomics and combined models showed superior performance with AUCs of 0.905 (95% CI 0.843-0.967) and 0.914 (95% CI 0.857-0.970), respectively. Similar trends were observed in the testing set, where the combined model (AUC 0.859, 95% CI 0.768-0.950) outperformed both the radiomics (AUC 0.815) and clinical (AUC 0.644) models. Conclusion The enhanced CT radiomics model has better predictive efficacy for remission with neoadjuvant chemotherapy combined with immunotherapy in esophageal squamous cell carcinoma patients, and the combined model has greater predictive value.

Rationale and Objectives To develop and validate a deep learning radiomics nomogram (DLRN) based on T2-weighted MRI to distinguish between borderline ovarian tumors (BOTs) and stage I epithelial ovarian cancer (EOC) preoperatively. Materials and Methods This retrospective multicenter study enrolled 279 patients from three centers, divided into a training set (n = 207) and an external test set (n = 72). The intra- and peritumoral radiomics analysis was employed to develop a combined radiomics model. A deep learning model was constructed based on the largest orthogonal slices of the tumor volume, and a clinical model was constructed using independent clinical predictors. The DLRN was then constructed by integrating deep learning, intra- and peritumoral radiomics, and clinical predictors. For comparison, an original radiomics model based solely on tumor volume (excluding the peritumoral area) was also constructed. All models were validated through 10-fold cross-validation and external testing, and their predictive performance was evaluated by the area under the receiver operating characteristic curve (AUC). Results The DLRN demonstrated superior performance across the 10-fold cross-validation, with the highest AUC of 0.825±0.082. On the external test set, the DLRN significantly outperformed the clinical model and the original radiomics model (AUC = 0.819 vs. 0.708 and 0.670, P = 0.047 and 0.015, respectively). Furthermore, the combined radiomics model performed significantly better than the original radiomics model (AUC = 0.778 vs. 0.670, P = 0.043). Conclusion The DLRN exhibited promising performance in distinguishing BOTs from stage I EOC preoperatively, thus potentially assisting clinical decision-making.

To develop a nomogram model which combined clinical inflammatory indicators and CT radiomics features to predict progression free survival (PFS) in esophageal squamous cell carcinoma (ESCC) after radical operation. 258 ESCC patients receiving surgical operation treatment were retrospectively collected from July 2017 to March 2019. Clinical data, laboratory results, pathology results, pre-operative CT data, and survival outcomes were analyzed. Using cox proportional hazards regression model to assess the relationship between relevant clinicopathological factors and PFS. C-index and calibration curve were used to evaluate the nomogram model. Survival curves were obtained using the Kaplan-Meier and comparisons were made by using the log-rank test. The inflammatory model, radiomics model and nomogram model all have good predictive efficacy for predicting PFS of ESCC patients in both training and test set. Significant differences were found between the nomogram model and inflammatory model and the radiomics model (DeLong test, Z = 3.869 and 3.195, P < 0.001, P = 0.001). Decision curve analysis (DCA) results revealed the net benefit of nomogram model was better than that of inflammatory model and radiomics model. Kaplan-Meier results showed significant difference in PFS between high-risk and low-risk group in Radscore and nomogram model (P < 0.001), and the high-risk group was prone to postoperative recurrence and poor PFS. The nomogram model developed by combining inflammatory indicators and radiomics features, which is helpful for risk stratification and follow-up work, and improving ESCC patients’ prognosis.

RATIONALE AND OBJECTIVES:This study aims to develop and validate a deep learning radiomics signature (DLRS) that integrates radiomics and deep learning features for the non-invasive prediction of microvascular invasion (MVI) in patients with colon cancer (CC). Furthermore, the study explores the potential association between DLRS and tumor immune heterogeneity. MATERIALS AND METHODS:This study is a multi-center retrospective study that included a total of 1007 patients with colon cancer (CC) from three medical centers and The Cancer Genome Atlas (TCGA-COAD) database. Patients from Medical Centers 1 and 2 were divided into a training cohort (n = 592) and an internal validation cohort (n = 255) in a 7:3 ratio. Medical Center 3 (n = 135) and the TCGA-COAD database (n = 25) were used as external validation cohorts. Radiomics and deep learning features were extracted from contrast-enhanced venous-phase CT images. Feature selection was performed using machine learning algorithms, and three predictive models were developed: a radiomics model, a deep learning (DL) model, and a combined deep learning radiomics (DLR) model. The predictive performance of each model was evaluated using multiple metrics, including the area under the curve (AUC), sensitivity, and specificity. Additionally, differential gene expression analysis was conducted on RNA-seq data from the TCGA-COAD dataset to explore the association between the DLRS and tumor immune heterogeneity within the tumor microenvironment. RESULTS:Compared to the standalone radiomics and deep learning models, DLR fusion model demonstrated superior predictive performance. The AUC for the internal validation cohort was 0.883 (95% CI: 0.828-0.937), while the AUC for the external validation cohort reached 0.855 (95% CI: 0.775-0.935). Furthermore, stratifying patients from the TCGA-COAD dataset into high-risk and low-risk groups based on the DLRS revealed significant differences in immune cell infiltration and immune checkpoint expression between the two groups (P < 0.05). CONCLUSION:The contrast-enhanced CT-based DLR fusion model developed in this study effectively predicts the MVI status in patients with CC. This model serves as a non-invasive preoperative assessment tool and reveals a potential association between the DLRS and immune heterogeneity within the tumor microenvironment, providing insights to optimize individualized treatment strategies.

To explore the feasibility of predicting the pathological activity of Crohn’s disease (CD) based on dual-energy CT enterography (DECTE). The clinical, endoscopic, imaging and pathological data of 55 patients with CD scanned by DECTE were retrospectively analyzed; the pathological results were used as a reference standard to classify the diseased bowel segments into active and inactive phases. The normalized iodine concentration (NIC), energy-spectrum curve slope K, dual energy index (DEI), fat fraction (FF) of the arterial phases and venous phases were compared. To assess the parameters’ predictive ability, receiver-operating characteristic curves were used. The Delong test was used to compare the differences between the diagnostic efficiency of each parameter. A total of 84 intestinal segments were included in the study, including 54 active intestinal segments and 30 inactive intestinal segments. The NIC, energy-spectrum curve slope K and DEI were significantly different between active and inactive bowel segments in the arterial and venous phases (P < 0.05), while FF were not significantly different (P > 0.05). The largest area under the curve (AUC) of NIC, energy-spectrum curve slope K and DEI were higher in arterial phase than in venous phase. For identifying the intestinal activity of CD, the maximum AUC of NIC in arterial phase was 0.908, with a sensitivity of 0.833 and a specificity of 0.800, and the DEI in arterial phase had the highest sensitivity (0.944). The NIC, energy-spectrum curve slope K and DEI can effectively distinguish the active and inactive phases of the intestinal segments of CD patients and provide good assistance for determining further treatment.

Abstract Objectives To investigate the potential of intratumoral and peritumoral radiomics derived from T2-weighted MRI to preoperatively predict extrapelvic peritoneal metastasis (EPM) in patients with epithelial ovarian cancer (EOC). Methods In this retrospective study, 488 patients from four centers were enrolled and divided into training (n = 245), internal test (n = 105), and external test (n = 138) sets. Intratumoral and peritumoral models were constructed based on radiomics features extracted from the corresponding regions. A combined intratumoral and peritumoral model was developed via a feature-level fusion. An ensemble model was created by integrating this combined model with specific independent clinical predictors. The robustness and generalizability of these models were assessed using tenfold cross-validation and both internal and external testing. Model performance was evaluated by the area under the receiver operating characteristic curve (AUC). The Shapley Additive Explanation method was employed for model interpretation. Results The ensemble model showed superior performance across the tenfold cross-validation, with the highest mean AUC of 0.844 ± 0.063. On the internal test set, the peritumoral and ensemble models significantly outperformed the intratumoral model (AUC = 0.786 and 0.832 vs. 0.652, p = 0.007 and p < 0.001, respectively). On the external test set, the AUC of the ensemble model significantly exceeded those of the intratumoral and peritumoral models (0.843 vs. 0.750 and 0.789, p = 0.008 and 0.047, respectively). Conclusion Peritumoral radiomics provide more informative insights about EPM than intratumoral radiomics. The ensemble model based on MRI has the potential to preoperatively predict EPM in EOC patients. Critical relevance statement Integrating both intratumoral and peritumoral radiomics information based on MRI with clinical characteristics is a promising noninvasive method to predict EPM to guide preoperative clinical decision-making for EOC patients. Key Points Peritumoral radiomics can provide valuable information about extrapelvic peritoneal metastasis in epithelial ovarian cancer. The ensemble model demonstrated satisfactory performance in predicting extrapelvic peritoneal metastasis. Combining intratumoral and peritumoral MRI radiomics contributes to clinical decision-making in epithelial ovarian cancer. Graphical Abstract

Rationale and Objectives: To assess the efficacy of consensus cluster analysis based on CT radiomics in stratifying risk and predicting postoperative progression-free survival (PFS) in patients diagnosed with esophageal squamous cell carcinoma (ESC). Materials and Methods: We conducted a retrospective study involving 546 patients diagnosed with ESC between January 2016 and March 2021. All patients underwent preoperative enhanced CT examinations. From the enhanced CT images, radiomics features were extracted, and a consensus clustering algorithm was applied to group the patients based on these features. Statistical analysis was performed to examine the relationship between the clustering results and gene protein expression, histopathological features, and patients' 3-year PFS. We applied the Kruskal-Wallis test for continuous data, chi-square or Fisher's exact tests for categorical data, and the log-rank test for PFS. Results: This study identified four groups: Cluster 1 (n = 100, 18.3%), Cluster 2 (n = 197, 36.1%), Cluster 3 (n = 205, 37.5%), and Cluster 4 (n = 44, 8.1%). The cancer gene Breast Cancer Susceptibility Gene 1 (BRCA1) was most highly expressed in Cluster 4 (75%), showing significant differences between the four subtypes with a P-value of 0.035. The expression of programmed death-1 (PD-1) was highest in Cluster 1 (51%), with a P-value of 0.022. Vascular invasion occurred most frequently in Cluster 2 (28.9%), with a P-value of 0.022. The majority of patients with stage T3-4 were in Cluster 2 (67%), with a P-value of 0.003. Kaplan-Meier survival analysis revealed significant differences in PFS between the four groups (P = 0.013). Among them, patients in Cluster 1 had the best prognosis, while those in Cluster 2 had the worst. Conclusion: This study highlights the effectiveness of consensus clustering analysis based on enhanced CT radiomics features in identifying associations between radiomics features, histopathological characteristics, and prognosis in different clusters. These findings provide valuable insights for clinicians in accurately and effectively evaluating the prognosis of esophageal cancer. (c) 2024 The Association of University Radiologists. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Rationale and Objectives: To develop and validate a T2-weighted magnetic resonance imaging (MRI)-based deep learning radiomics nomogram (DLRN) to differentiate between type I and type II epithelial ovarian cancer (EOC). Materials and Methods: This multicenter study incorporated 437 patients from five centers, divided into training ( n = 271), internal validation ( n = 68), and external validation ( n = 98) sets. The deep learning (DL) model was constructed using the largest orthogonal slices of the tumor area. The extracted radiomics features were employed in building the radiomics model. The clinical model was developed based on clinical characteristics. A DLRN was built by integrating the DL signature, radiomics signature, and independent clinical predictors. Model performances were evaluated through receiver operating characteristic (ROC) analysis, Brier score, calibration curve, and decision curve analysis (DCA). The areas under the ROC curve (AUCs) were compared using the DeLong test. A two-tailed P < 0.05 was considered significantly different. Results: The DLRN exhibited satisfactory discrimination between type I and type II EOC with the AUC of 0.888 (95% confidence interval [CI] 0.810, 0.966) and 0.866 (95% CI 0.786, 0.946) in the internal and external validation sets, respectively. These AUCs significantly exceeded those of the clinical model ( P = 0.013 and 0.043, in the internal and external validation sets, respectively). The DLRN demonstrated optimal classification accuracy and clinical application value, according to Brier scores, calibration curves, and DCA. Conclusion: A T2-weighted MRI-based DLRN showed promising potential in differentiating between type I and type II EOC, which could offer assistance in clinical decision-making.

Objective: To explore whether preoperative contrast-enhanced computed tomogrpahy (CT) can predict lymphovascular invasion (LVI) in esophageal squamous cell carcinoma (ESCC), and provide a reliable reference for the formulation of clinical individualized treatment plans. Methods: This retrospective study enrolled 228 patients with surgically resected and pathologically confirmed ESCC, including 36 patients with LVI and 192 patients without LVI. All patients underwent contrast-enhanced CT (CECT) scan within 2 weeks before the operation. Tumor size (including tumor length and maximum tumor thickness), tumor-to-normal wall enhancement ratio (TNR), and gross tumor volume (GTV) were obtained. All clinical features and CECT-derived parameters associated with LVI were analyzed by univariate and multivariate analysis. The independent predictors for LVI were identified, and their combination was built by multivariate logistic regression analysis, using the significant variables from the univariate analysis as inputs. Results: Univariate analysis of clinical features and CECT-derived parameters revealed that age, TNR, and clinical N stage (cN stage) were significantly associated with LVI. The multivariable analysis results demonstrated that age (odds ratio [OR]: 5.32, 95% confidence interval [CI]: 2.224-12.743, P<.001), TNR (OR: 5.399, 95% CI: 1.609-18.110, P  =  .006), and cN stage (cN1: OR: 2.874, 95% CI: 1.182-6.989, P  =  .02; cN2: OR: 6.876, 95% CI: 2.222-21.227) were identified to be independent predictors for LVI. The combination of age, TNR, and cN stage achieved a relatively higher area under the curve (AUC) (0.798), accuracy (ACC) (65.4%), sensitivity (SEN) (69.4%), specificity (SPE) (79.7%), positive predictive value (PPV) (77.4%), and negative predictive value (NPV) (71.6%). Conclusions: The combination of clinical features and CECT-derived parameters may be effective in predicting LVI status preoperatively in ESCC.