Advances in artificial intelligence (AI), multi-omic profiling, and sophisticated imaging technologies have significantly advanced personalized medicine in gastrointestinal surgical oncology. These technological innovations enable precise patient stratification, tailored surgical strategies, and individualized therapeutic approaches, thereby significantly enhancing clinical outcomes. Despite remarkable progress, challenges persist, including the standardization and integration of diverse data types, ethical concerns regarding patient privacy, and rigorous clinical validation of predictive models. Addressing these challenges requires establishing international standards for data interoperability, such as Fast Healthcare Interoperability Resources, and adopting advanced security methods, such as homomorphic encryption, to facilitate secure multi-institutional data sharing. Moreover, ensuring model transparency and explainability through techniques such as explainable AI is critical for fostering trust among clinicians and patients. The successful integration of these advanced technologies necessitates strong multidisciplinary collaboration among surgeons, radiologists, geneticists, pathologists, and oncologists. Ultimately, the continued development and effective implementation of these personalized medical strategies complemented by human expertise promise a transformative shift toward patient-centered care, improving long-term outcomes for patients with gastrointestinal cancer.

While multi-task learning (MTL) has been widely developed for natural image analysis, its potential for enhancing performance in medical imaging remains relatively unexplored. Most methods formulate MTL as a multi-objective problem, inherently forcing all tasks to compete with each other during optimization. In this work, we propose a novel approach by formulating MTL as a multi-level optimization problem, in which the features learned from one task are optimized by benefiting from the other tasks. Specifically, we advocate for a cooperative approach where each task considers the features of others, enabling individual performance enhancement without detriment to others. To achieve this objective, we introduce a novel optimization strategy aimed at seeking flat minima for each sub-problem, fostering the learning of robust sub-models resilient to changes in other sub-models. We demonstrate the advantages of our proposed method through comprehensive parameter and comparison studies on the OrganCMNIST dataset. Additionally, we evaluate its efficacy on three eye-related medical image datasets, comparing its performance against other state-of-the-art MTL approaches. The results highlight the superiority of our method over existing approaches, showcasing its potential for training multi-purpose models in medical image analysis.

Upper gastrointestinal cancers, mainly comprising esophageal and gastric cancers, are among the most prevalent cancers worldwide. There are many new cases of upper gastrointestinal cancers annually, and the survival rate tends to be low. Therefore, timely screening, precise diagnosis, appropriate treatment strategies, and effective prognosis are crucial for patients with upper gastrointestinal cancers. In recent years, an increasing number of studies suggest that artificial intelligence (AI) technology can effectively address clinical tasks related to upper gastrointestinal cancers. These studies mainly focus on four aspects: screening, diagnosis, treatment, and prognosis. In this review, we focus on the application of AI technology in clinical tasks related to upper gastrointestinal cancers. Firstly, the basic application pipelines of radiomics and deep learning in medical image analysis were introduced. Furthermore, we separately reviewed the application of AI technology in the aforementioned aspects for both esophageal and gastric cancers. Finally, the current limitations and challenges faced in the field of upper gastrointestinal cancers were summarized, and explorations were conducted on the selection of AI algorithms in various scenarios, the popularization of early screening, the clinical applications of AI, and large multimodal models.

To propose and validate a CT radiomics model utilizing radiomic features from lymph nodes (LNs) with maximum short axis diameter (MSAD) < 1 cm for predicting small metastatic LN (sMLN) in patients with resectable esophageal squamous cell carcinoma (ESCC).

A total of 196 resectable patients with ESCC undergoing surgery were retrospectively enrolled, among whom 25% had sMLN. 146 out of 196 patients (from hospital 1) were randomly divided into the training (n = 116) and testing cohorts (n = 30) at an 8:2 ratio, while the remaining 50 patients from hospital 2 constituted the external validation cohort. Least absolute shrinkage and selection operator binary logistic regression was employed for radiomics feature dimensionality reduction and selection, and multivariable logistic regression analysis was used to construct the radiomics prediction model. The clinical features were statistically selected to develop the clinical model. And both the selected radiomics and clinical features were used to develop the combined model. The predictive value of models was assessed using the area under the receiver operating characteristic curves (AUC).

The LN radiomics model was constructed with 9 radiomics features, the clinical model was developed with 3 clinical features, and the combined model was developed using both the LN radiomics and clinical features. However, no statistical radiomics features from ESCC were extracted in dimensionality reduction. Compared to the clinical model, the combined model exhibited superior predictive ability (AUC: 0.893 vs. 0.766, P = 0.003), and the LN radiomics model showed slightly better predictive ability (AUC: 0.860 vs. 0.766, P = 0.153). It was validated in the test and external validation cohorts.

The combined model could assist in preoperatively identifying sMLN in resectable ESCC. It is beneficial for more accurate N staging and clinical comprehensive staging of ESCC, thereby facilitating the clinical physician to make more personalized and standardized treatment strategies.

To determine whether intratumoral and peritumoral radiomics derived from dual-phase contrast-enhanced CT imaging could predict lymph node metastasis (LNM) in gastric cancer.

Patients with gastric cancer from January 2017 to January 2022 were retrospectively collected and were randomly divided into training cohort (n = 287) and test cohort (n = 121) with a ratio of 7: 3. Clinical features and traditional radiological features were analyzed to construct clinical model. Radiomics features based on intratumoral (ITV) and peritumoral volumetric (PTV) regions of the tumor were extracted and screened to construct radiomics models. Clinical-radiomics combined model was constructed by the most predictive radiomics features and clinical independent predictors. The correlation between LNM predicted by the best model and 2-year disease-free survival (DFS) was evaluated by the Kaplan-Meier analysis.

CT-LNM and CT-T stage were independent predictors of LNM. Compared with other radiomics models, ITV + PTV on atrial and venous phase (ITV + PTV-AP + VP) radiomics model presented moderate AUCs of 0.679 and 0.670 in the training cohort and validation cohort, respectively. Among the models, clinical-radiomics combined model achieved the highest AUC of 0.894 and 0.872 in the training and test cohorts, and 0.744 and 0.784 in the T1-2 and T3-4 subgroups, respectively. Clinical-radiomics combined model based LNM could stratify patients into high-risk and low-risk groups, and 2-year DFS of high-risk group was significantly lower than that of low-risk group (p < 0.001).

Clinical-radiomics combined model integrating CT-LNM, CT-T stage, and ITV-PTV-AP + VP radiomics features could predict LNM, and this combined model based LNM was associated with 2-year DFS.

This study aimed to assess the effectiveness of various machine learning models in identifying lymph node metastasis in colon cancer patients and to explore the potential benefits of combining clinicoradiological and radiomics features for improved diagnosis. A total of 260 patients with pathologically confirmed colon cancer were retrospectively included in study center 1 and study center 2 from January 2015 to August 2024. A total of 198 patients with colon cancer in center 1 were randomly divided into a training set (n = 138) and an internal testing set (n = 60) at a ratio of 7:3. Patients in center 2 were included in the external testing set (n = 62). Five clinical radiological features were used to establish a clinical model. Radiomics features were extracted from the computed tomography venous phase images, and four classifiers, including logistic regression, support vector machine, decision tree, and k‑nearest neighbor, were used to build machine learning models. In addition, a combined model was constructed by joining clinical, radiological, and radiogenomic features. The performance of these models was evaluated in terms of accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), receiver operating curve (ROC) and calibration curves in the training set, internal testing set, and external testing set to determine the diagnostic model with the highest predictive efficiency and to evaluate the stability of the model. Among the four machine learning models, the SVM model had the best predictive performance, with an area under the ROC (AUC) of 0.813, 0.724, and 0.721 for the training set, internal testing set, and external testing set, respectively. The clinical model, radiomics model, and combined model had an AUC of 0.823, 0.813, 0.817, 0.508, 0.724, 0.751, 0.582, 0.721, and 0.744 in the training set, internal testing set, and external testing set, respectively. In conclusion, the combined model performed significantly better than the clinical model (p = 0.017, 0.038), but there was no significant difference between the radiomics model and the combined model (p = 0.556, 0.614).

A predictive model was developed based on enhanced computed tomography (CT), laboratory test results, and pathological indicators to achieve the convenient and effective prediction of single lymph node metastasis (LNM) in gastric cancer.

Sixty-six consecutive patients (235 regional lymph nodes) with pathologically confirmed gastric cancer who underwent surgery at our hospital between December 2020 and November 2021 were retrospectively reviewed. They were randomly allocated to training (n = 38, number of lymph nodes = 119) and validation (n = 28, number of lymph nodes = 116) datasets. The clinical data, laboratory test results, enhanced CT characteristics, and pathological indicators from gastroscopy-guided needle biopsies were obtained. Multivariable logistic regression with generalised estimation equations (GEEs) was used to develop a predictive model for LNM in gastric cancer. The predictive performance of the model developed using the training and validation datasets was validated using receiver operating characteristic curves.

Lymph node enhancement pattern, Ki67 level, and lymph node long-axis diameter were independent predictors of LNM in gastric cancer (p < 0.01). The GEE-logistic model was associated with LNM (p = 0.001). The area under the curve and accuracy of the model, with 95% confidence intervals, were 0.944 (0.890-0.998) and 0.897 (0.813-0.952), respectively, in the training dataset and 0.836 (0.751-0.921) and 0.798 (0.699-0.876), respectively, in the validation dataset.

The predictive model constructed based on lymph node enhancement pattern, Ki67 level, and lymph node long-axis diameter exhibited good performance in predicting LNM in gastric cancer and should aid the lymph node staging of gastric cancer and clinical decision-making.

The expansion of function-preserving surgery became possible due to a more profound understanding of gastric cancer (GC), and T1N + or T2N + gastric cancer patients might be potential beneficiaries. However, ways to evaluate the possibility of function-preserving pylorus surgery are still unknown.

A total of 288 patients at Renji Hospital and 58 patients at Huadong Hospital, pathologically diagnosed with gastric cancer staging at T1 and T2 with tumors located in the upper two-thirds of the stomach, were retrospectively enrolled from March 2015 to October 2022. Tumor regions of interest (ROIs) were manually delineated on bi-phase CT images, and a nomogram was built and evaluated.

The radiomic features distributed differently between positive and negative pLNm groups. Two radiomic signatures (RS1 and RS2) and one clinical signature were constructed. The radiomic signatures exhibited good performance for discriminating pLNm status in the test set. The three signatures were then combined into an integrated nomogram (IN). The IN showed good discrimination of pLNm in the Renji cohort (AUC 0.918) and the Huadong cohort (AUC 0.649). The verification models showed high values.

For GC patients with T1 and T2 tumors located in the upper two-thirds of the stomach, a nomogram was successfully built for predicting pylorus lymph node metastasis, which would guide the surgical indication extension of conservative gastrectomies.

This study explores the clinical value of a machine learning (ML) model based on ultrasound radiomics features of primary foci, combined with clinicopathologic factors to predict the pathological complete response (pCR) of neoadjuvant chemotherapy (NAC) for patients with breast cancer (BC).

We retrospectively analyzed ultrasound images and clinical information from 231 participants with BC who received NAC. These patients were randomly assigned to training and validation cohorts. Tumor regions of interest (ROI) were delineated, and radiomics features were extracted. Z-score normalization, Pearson correlation analysis, and the least absolute shrinkage selection operator (LASSO) were utilized for further screening ultrasound radiomics and clinical features. Univariate and multivariate logistic regression analysis were performed to identify the CFs that were independently associated with pCR. We compared 10 ML models based on radiomics features: support vector machine (SVM), logistic regression (LR), random forest, extra trees (ET), naïve Bayes (NB), k-nearest neighbor (KNN), multilayer perceptron (MLP), gradient boosting ML (GBM), light GBM (LGBM), and adaptive boost (AB). Diagnostic performance was evaluated using the receiver operating characteristic (ROC) area under the curve (AUC), accuracy, sensitivity, and specificity, and the Rad score was calculated. Subsequently, construction of clinical predictive models and Rad score joint clinical predictive models using ML algorithms for optimal diagnostic performance. The diagnostic process of the ML model was visualized and analyzed using SHapley Additive exPlanation (SHAP).

Out of 231 participants with BC, 98 (42.42%) achieved pCR, and 133 (57.58%) did not. Twelve radiomics features were identified, with the GBM model demonstrating the best predictive performance (AUC of 0.851, accuracy of 0.75, sensitivity of 0.821, and specificity of 0.698). The clinical feature prediction model using the GBM algorithm had an AUC of 0.819 and an accuracy of 0.739. Combining the Rad score with clinical features in the GBM model resulted in superior predictive performance (AUC of 0.939 and an accuracy of 0.87). SHAP analysis indicated that participants with a high Rad score, PR-negative, ER-negative and human epidermal growth factor receptor-2 (HER-2) positive were more possibly to reach pCR. Based on the decision curve analysis, it was shown that the combined model of GBM provided higher clinical benefits.

The GBM model based on ultrasound radiomics features and routine clinical date of BC patients had high performance in predicting pCR. SHAP analysis provided a clear explanation for the prediction results of the GBM model, revealing that patients with a high Rad score, PR-negative status, ER-negative status and HER-2-positive status are more likely to achieve pCR.

Lymph node metastasis is an important route for gastric cancer metastasis. The clinical significance of transverse lymph node metastasis (TLNM) is still unclear.

This study investigates effects of TLNM on the prognosis of GC patients and establishes two nomograms for evaluating the prognosis of GC patients and for predicting the risk clinicopathological factors to TLNM based on a Chinese medical database.

A total of 902 GC patients with lymph node metastasis (LNM) who underwent R0 gastrectomy was included in this study. According to results of Cox proportional hazards analyses and logistic regression analyses, the prognostic and the predictive nomograms were established and validated.

The overall survival of patients with TLNM was significantly worse than those without TLNM (P < 0.001) and similar to patients with extra-gastric LNM (P > 0.05). TLNM independently influenced prognosis of GC patients. Prognostic and predictive nomograms were established and validated. Both nomograms were proven that have high accuracy by calculating each AUC (Area Under Cure) value. Calibration curves aligned well with actual outcomes. DCA (Decision Curve Analyses) analyses indicated the high clinical utility.

These nomograms offer precise survival and TLNM occurrence predictions, which may aid clinical decisions.

To explore the ultrasonographic features and influencing factors of free-floating thrombus (FFT) detachment in the lower extremity deep veins (LEDVs) of patients with fractures.

Clinical data of patients diagnosed with FFT in the LEDVs and implanted with an inferior vena cava filter (IVCF) in our hospital between July 2021 and August 2023 were retrospectively analysed. The patients were divided into the thrombus detachment group (the experimental group, n = 92) and the non-thrombus detachment group (the control group, n = 103) based on the presence of detached thrombus in the IVCF. The effects of thrombus echogenicity, floating degree, thrombus location, thrombin time, D-dimer and fibrinogen on thrombus shedding were analysed. The nomogram method was used to establish the model and predict the probability of delayed postoperative recovery.

The proportions of patients with extremely hypoechoic thrombus and medium and high floating degrees increased in the experimental group compared with those in the control group, and the differences between the two groups were statistically significant (P < 0.05). Extremely hypoechoic thrombus (P = 0.021, 95 % CI: 1.109-13.748) and high (P = 0.001, 95 % CI: 3.854-28.573) and medium floating degrees (P = 0.004, 95 % CI: 1.792-13.453) were risk factors for deep veins FFT (DV FFT) detachment. The results of receiver operating characteristic curve analysis showed that the area under the curve of the model was 0.893, with a 95 % CI of 0.856-0.937, indicating a high prediction accuracy.

Ultrasonographic parameters, including thrombus echogenicity and floating degree, are valuable in predicting DV FFT detachment in patients with traumatic fractures, providing references for IVCF implantation.

The aim of this study was to explore whether machine learning model based on computed tomography (CT) radiomics and clinical characteristics can differentiate Epstein-Barr virus-associated gastric cancer (EBVaGC) from non-EBVaGC.

Contrast-enhanced CT images were collected from 158 patients with GC (46 EBV-positive, 112 EBV-negative) between April 2018 and February 2023. Radiomics features were extracted from the volumes of interest. A radiomics signature was built based on radiomics features by the least absolute shrinkage and selection operator logistic regression algorithm. Multivariate analyses were used to identify significant clinicoradiological variables. We developed 6 ML models for EBVaGC, including logistic regression, Extreme Gradient Boosting, random forest (RF), support vector machine, Gaussian Naive Bayes, and K-nearest neighbor algorithm. The area under the receiver operating characteristic curve (AUC), the area under the precision-recall curves (AP), calibration plots, and decision curve analysis were applied to assess the effectiveness of each model.

Six ML models achieved AUC of 0.706-0.854 and AP of 0.480-0.793 for predicting EBV status in GC. With an AUC of 0.854 and an AP of 0.793, the RF model performed the best. The forest plot of the AUC score revealed that the RF model had the most stable performance, with a standard deviation of 0.003 for AUC score. RF also performed well in the testing dataset, with an AUC of 0.832 (95% confidence interval: 0.679-0.951), accuracy of 0.833, sensitivity of 0.857, and specificity of 0.824, respectively.

The RF model based on clinical variables and Rad_score can serve as a noninvasive tool to evaluate the EBV status of gastric cancer.

To evaluate the performance of dual-energy CT (DECT)-based radiomics models for identifying high-risk histopathologic phenotypes-serosal invasion (pT4a), lymph node metastasis (LNM), lymphovascular invasion (LVI) and perineural invasion (PNI) in gastric cancer.

This prospective bi-center study recruited histologically confirmed gastric adenocarcinoma patients who underwent triple-phase enhanced DECT before gastrectomy between January 2021 and July 2023. Radiomics features were extracted from polychromatic/monochromatic (40 keV, 100 keV)/iodine images at arterial/venous/delay phase, respectively. Predictive features were selected in the training dataset using logistic regression classifier, and trained models were applied to the external validation dataset. Performances of clinical models, conventional contrast enhanced CT (CECT) models and DECT models were evaluated using areas under the receiver operating characteristic curve (AUCs).

In total, 503 patients were recruited: 396 at training dataset (60.1 ± 10.8 years, 110 females, 286 males) and 107 at validation dataset (61.4 ± 9.5 years, 29 females, 78 males). DECT models dichotomizing pT4a, LNM, LVI, and PNI achieved AUCs of 0.891, 0.817, 0.834, and 0.889, respectively, in the validation dataset, similar with the CECT models. In the training dataset, compared to the CECT model, the DECT model provided increased performance for identifying pT4a, LNM, LVI (all P＜0.05), and similar performance for stratifying PNI (P = 0.104). The DECT models was associated with patient disease-free survival (all P＜0.05).

DECT radiomics can stratify patients preoperatively according to high-risk histopathologic phenotypes for gastric cancer and are associated with patient disease-free survival in the training dataset.

This study aims to evaluate whether radiomic features of pericoronary adipose tissue (PCAT) derived from coronary computed tomography angiography (CCTA) can better predict major adverse cardiovascular events (MACE) in patients with angina pectoris.

A single-center retrospective study included 239 patients with angina pectoris who underwent coronary CT examinations. Participants were divided into MACE (n = 46) and non-MACE (n = 193) groups based on the occurrence of MACE during follow-up, and further allocated into a training cohort (n = 167) and a validation cohort (n = 72) at a 7:3 ratio. Automatic segmentation of PCAT surrounding the proximal segments of the left anterior descending artery (LAD), left circumflex coronary artery (LCX), and right coronary artery (RCA) was performed for all patients. Radiomic features of the coronary arteries were extracted, screened, and integrated while quantifying the fat attenuation index (FAI) for the three vessels. Univariate and multivariate logistic regression analyses were utilized to select clinical predictors of adverse cardiovascular events. Subsequently, machine learning techniques were employed to construct models based on FAI, clinical features, and radiomic characteristics. The predictive performance of each model was assessed and compared using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis for clinical utility.

The radiomics model demonstrated superior performance in predicting MACE in patients with angina pectoris within both the training and validation cohorts, yielding areas under the curve (AUC) of 0.83 and 0.71, respectively, which significantly outperformed the FAI model (AUC = 0.71, 0.54) and the clinical model (AUC = 0.81, 0.67), with statistically significant differences in AUC (p < 0.05). Calibration curves for all three predictive models exhibited good fit (all p > 0.05). Decision curve analysis indicated that the radiomics model provided higher clinical benefit than the traditional clinical and FAI models.

The CCTA-based PCAT radiomics model is an effective tool for predicting MACE in patients with angina pectoris, assisting clinicians in optimizing risk stratification for individual patients. The CCTA-based radiomics model significantly surpasses traditional FAI and clinical models in predicting major adverse cardiovascular events in patients with angina pectoris.

Vascular and nerve infiltration are important indicators for the progression and prognosis of gastric cancer (GC), but traditional imaging methods have some limitations in preoperative evaluation. In recent years, energy spectrum computed tomography (CT) multiparameter imaging technology has been gradually applied in clinical practice because of its advantages in tissue contrast and lesion detail display.

To explore and analyze the value of multiparameter energy spectrum CT imaging in the preoperative assessment of vascular invasion (LVI) and nerve invasion (PNI) in GC patients.

Data from 62 patients with GC confirmed by pathology and accompanied by energy spectrum CT scanning at our hospital between September 2022 and September 2023, including 46 males and 16 females aged 36-71 (57.5 ± 9.1) years, were retrospectively collected. The patients were divided into a positive group (42 patients) and a negative group (20 patients) according to the presence of LVI/PNI. The CT values (CT40 keV, CT70 keV), iodine concentration (IC), and normalized IC (NIC) of lesions in the upper energy spectrum CT images of the arterial phase, venous phase, and delayed phase 40 and 70 keV were measured, and the slopes of the energy spectrum curves [K (40-70)] from 40 to 70 keV were calculated. Arterial phase combined parameter, venous phase combined parameters (VP-ALLs), and delayed phase association parameters were calculated for patients with late-stage disease. The differences in the energy spectrum parameters between the positive and negative groups were compared, receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC), sensitivity, specificity, and optimal threshold were calculated to measure the diagnostic efficiency of each parameter.

In the delayed phase, the CT40 keV, CT70 keV, K (40-70), IC, NIC, and CT70 keV and the NIC in the upper arterial and venous phases of energy spectrum CT were greater in the LVI/PNI-positive group than in the LVI-negative group. The representative parameters for the arterial phase NIC were 0.14 ± 0.04 in the positive group and 0.12 ± 0.04 in the negative group. The venous phase NIC was 0.5 (0.5, 0.6) in the positive group and 0.4 (0.4, 0.5) in the negative group. Last, for the delayed phase NIC, it was 0.6 ± 0.1 in the positive group and 0.5 ± 0.1 in the negative group (all P values are less than 0.05). ROC curve analysis demonstrated that the diagnostic efficacy of each parameter during the venous stage was superior to that during the arterial and delayed stages. Furthermore, the diagnostic efficacy of the combined parameter throughout all three stages was superior to that of any single parameter. The AUC, sensitivity, and specificity of the optimal parameter, VP-ALL, were 0.931 (95% confidence interval: 0.872-0.990), 80.95%, and 95.00%, respectively.

When assessing the condition of LVI and PNI (perineural invasion) in patients with GC prior to surgery, the ability to diagnose these conditions using venous stage parameters was superior to that using arterial stage and delayed stage parameters. Furthermore, the diagnostic accuracy of using a combination of parameters was better than that of using individual parameters alone.

Vascular invasion (VI) is closely related to the metastasis, recurrence, prognosis, and treatment of gastric cancer. Currently, predicting VI preoperatively using traditional clinical examinations alone remains challenging. This study aims to explore the value of radiomics analysis based on preoperative enhanced CT images in predicting VI in gastric cancer.

We retrospectively analyzed 194 patients with gastric adenocarcinoma who underwent enhanced CT examination. Based on pathology analysis, patients were divided into the VI group (n = 43) and the non-VI group (n = 151). Radiomics features were extracted from arterial phase (AP) and portal venous phase (PP) CT images. The radiomics score (Rad-score) was then calculated. Prediction models based on image features, clinical factors, and a combination of both were constructed. The diagnostic efficiency and clinical usefulness of the models were evaluated using receiver operating characteristic (ROC) curves and decision curve analysis (DCA).

The combined prediction model included the Rad-score of AP, the Rad-score of PP, Ki-67, and Lauren classification. In the training group, the area under the curve (AUC) of the combined prediction model was 0.83 (95% CI 0.76-0.89), with a sensitivity of 64.52% and a specificity of 92.45%. In the validation group, the AUC was 0.80 (95% CI 0.67-0.89), with a sensitivity of 66.67% and a specificity of 88.89%. DCA indicated that the combined prediction model might have a greater net clinical benefit than the clinical model alone.

The integrated models, incorporating enhanced CT radiomics features, Ki-67, and clinical factors, demonstrate significant predictive capability for VI. Moreover, the radiomics model has the potential to optimize personalized clinical treatment selection and patient prognosis assessment.

Survival prognosis of patients with gastric cancer (GC) often influences physicians' choice of their follow-up treatment. This study aimed to develop a positron emission tomography (PET)-based radiomics model combined with clinical tumor-node-metastasis (TNM) staging to predict overall survival (OS) in patients with GC.

We reviewed the clinical information of a total of 327 patients with pathological confirmation of GC undergoing 18 F-fluorodeoxyglucose (18 F-FDG) PET scans. The patients were randomly classified into training (n = 229) and validation (n = 98) cohorts. We extracted 171 PET radiomics features from the PET images and determined the PET radiomics scores (RS) using the least absolute shrinkage and selection operator (LASSO) and random survival forest (RSF). A radiomics model, including PET RS and clinical TNM staging, was constructed to predict the OS of patients with GC. This model was evaluated for discrimination, calibration, and clinical usefulness.

On multivariate COX regression analysis, the difference between age, carcinoembryonic antigen (CEA), clinical TNM, and PET RS in GC patients was statistically significant (p < 0.05). A radiomics model was developed based on the results of COX regression. The model had the Harrell's concordance index (C-index) of 0.817 in the training cohort and 0.707 in the validation cohort and performed better than a single clinical model and a model with clinical features combined with clinical TNM staging. Further analyses showed higher PET RS in patients who were older (p < 0.001) and those who had elevated CEA (p < 0.001) and higher clinical TNM (p < 0.001). At different clinical TNM stages, a higher PET RS was associated with a worse survival prognosis.

Radiomics models based on PET RS, clinical TNM, and clinical features may provide new tools for predicting OS in patients with GC.

The study aims to investigate the predictive capability of machine learning algorithms for omental metastasis in locally advanced gastric cancer (LAGC) and to compare the performance metrics of various machine learning predictive models. A retrospective collection of 478 pathologically confirmed LAGC patients was undertaken, encompassing both clinical features and arterial phase computed tomography images. Radiomic features were extracted using 3D Slicer software. Clinical and radiomic features were further filtered through lasso regression. Selected clinical and radiomic features were used to construct omental metastasis predictive models using support vector machine (SVM), decision tree (DT), random forest (RF), K-nearest neighbors (KNN), and logistic regression (LR). The models' performance metrics included accuracy, area under the curve (AUC) of the receiver operating characteristic curve, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). In the training cohort, the RF predictive model surpassed LR, SVM, DT, and KNN in terms of accuracy, AUC, sensitivity, specificity, PPV, and NPV. Compared to the other four predictive models, the RF model significantly improved PPV. In the test cohort, all five machine learning predictive models exhibited lower PPVs. The DT model demonstrated the most significant variation in performance metrics relative to the other models, with a sensitivity of 0.231 and specificity of 0.990. The LR-based predictive model had the lowest PPV at 0.210, compared to the other four models. In the external validation cohort, the performance metrics of the predictive models were generally consistent with those in the test cohort. The LR-based model for predicting omental metastasis exhibited a lower PPV. Among the machine learning algorithms, the RF predictive model demonstrated higher accuracy and improved PPV relative to LR, SVM, KNN, and DT models.

To construct and validate CT radiomics model based on the peritumoral adipose region of gastric adenocarcinoma to preoperatively predict lymph node metastasis (LNM).

293 consecutive gastric adenocarcinoma patients receiving radical gastrectomy with lymph node dissection in two medical institutions were stratified into a development set (from Institution A, n = 237), and an external validation set (from Institution B, n = 56). Volume of interest of peritumoral adipose region was segmented on preoperative portal-phase CT images. The least absolute shrinkage and selection operator method and stepwise logistic regression were used to select features and build radiomics models. Manual classification was performed according to routine CT characteristics. A classifier incorporating the radiomics score and CT characteristics was developed for predicting LNM. Area under the receiver operating characteristic curve (AUC) was used to show discrimination between tumors with and without LNM, and the calibration curves and Brier score were used to evaluate the predictive accuracy. Violin plots were used to show the distribution of radiomics score.

AUC values of radiomics model to predict LNM were 0.938, 0.905, and 0.872 in the training, internal test, and external validation sets, respectively, higher than that of manual classification (0.674, all P values < 0.01). The radiomics score of the positive LNM group were higher than that of the negative group in all sets (both P-values < 0.001). The classifier showed no improved predictive power compared with the radiomics signature alone with AUC values of 0.916 and 0.872 in the development and external validation sets, respectively. Multivariate analysis showed that radiomics score was an independent predictor.

Radiomics model based on peritumoral adipose region could be a useful approach for preoperative LNM prediction in gastric adenocarcinoma.

To determine whether contrast-enhanced CT radiomics features can preoperatively predict lymphovascular invasion (LVI) and perineural invasion (PNI) in gastric cancer (GC).

A total of 148 patients were included in the LVI group, and 143 patients were included in the PNI group. Three predictive models were constructed, including clinical, radiomics, and combined models. A nomogram was developed with clinical risk factors to predict LVI and PNI status. The predictive performance of the three models was mainly evaluated using the mean area under the curve (AUC). The performance of three predictive models was assessed concerning calibration and clinical usefulness.

In the LVI group, the predictive power of the combined model (AUC=0.871, 0.822) outperformed the clinical model (AUC=0.792, 0.728) and the radiomics model (AUC=0.792, 0.728) in both the training and testing cohorts. In the PNI group, the combined model (AUC=0.834, 0.828) also had better predictive power than the clinical model (AUC=0.764, 0.632) and the radiomics model (AUC=0.764, 0.632) in both the training and testing cohorts. The combined models also showed good calibration and clinical usefulness for LVI and PNI prediction.

CECT-based radiomics analysis might serve as a non-invasive method to predict LVI and PNI status in GC.

Accurate segmentation of gastric tumors from CT scans provides useful image information for guiding the diagnosis and treatment of gastric cancer. However, automated gastric tumor segmentation from 3D CT images faces several challenges. The large variation of anisotropic spatial resolution limits the ability of 3D convolutional neural networks (CNNs) to learn features from different views. The background texture of gastric tumor is complex, and its size, shape and intensity distribution are highly variable, which makes it more difficult for deep learning methods to capture the boundary. In particular, while multi-center datasets increase sample size and representation ability, they suffer from inter-center heterogeneity.

In this study, we propose a new cross-center 3D tumor segmentation method named Hierarchical Class-Aware Domain Adaptive Network (HCA-DAN), which includes a new 3D neural network that efficiently bridges an Anisotropic neural network and a Transformer (AsTr) for extracting multi-scale context features from the CT images with anisotropic resolution, and a hierarchical class-aware domain alignment (HCADA) module for adaptively aligning multi-scale context features across two domains by integrating a class attention map with class-specific information. We evaluate the proposed method on an in-house CT image dataset collected from four medical centers and validate its segmentation performance in both in-center and cross-center test scenarios.

Our baseline segmentation network (i.e., AsTr) achieves best results compared to other 3D segmentation models, with a mean dice similarity coefficient (DSC) of 59.26%, 55.97%, 48.83% and 67.28% in four in-center test tasks, and with a DSC of 56.42%, 55.94%, 46.54% and 60.62% in four cross-center test tasks. In addition, the proposed cross-center segmentation network (i.e., HCA-DAN) obtains excellent results compared to other unsupervised domain adaptation methods, with a DSC of 58.36%, 56.72%, 49.25%, and 62.20% in four cross-center test tasks.

Comprehensive experimental results demonstrate that the proposed method outperforms compared methods on this multi-center database and is promising for routine clinical workflows.

This study aimed to develop and validate an artificial intelligence radiopathological model using preoperative CT scans and postoperative hematoxylin and eosin (HE) stained slides to predict the pathological staging of gastric cancer (stage I-II and stage III).

This study included a total of 202 gastric cancer patients with confirmed pathological staging (training cohort: n = 141; validation cohort: n = 61). Pathological histological features were extracted from HE slides, and pathological models were constructed using logistic regression (LR), support vector machine (SVM), and NaiveBayes. The optimal pathological model was selected through receiver operating characteristic (ROC) curve analysis. Machine learnin algorithms were employed to construct radiomic models and radiopathological models using the optimal pathological model. Model performance was evaluated using ROC curve analysis, and clinical utility was estimated using decision curve analysis (DCA).

A total of 311 pathological histological features were extracted from the HE images, including 101 Term Frequency-Inverse Document Frequency (TF-IDF) features and 210 deep learning features. A pathological model was constructed using 19 selected pathological features through dimension reduction, with the SVM model demonstrating superior predictive performance (AUC, training cohort: 0.949; validation cohort: 0.777). Radiomic features were constructed using 6 selected features from 1834 radiomic features extracted from CT scans via SVM machine algorithm. Simultaneously, a radiopathomics model was built using 17 non-zero coefficient features obtained through dimension reduction from a total of 2145 features (combining both radiomics and pathomics features). The best discriminative ability was observed in the SVM_radiopathomics model (AUC, training cohort: 0.953; validation cohort: 0.851), and clinical decision curve analysis (DCA) demonstrated excellent clinical utility.

The radiopathomics model, combining pathological and radiomic features, exhibited superior performance in distinguishing between stage I-II and stage III gastric cancer. This study is based on the prediction of pathological staging using pathological tissue slides from surgical specimens after gastric cancer curative surgery and preoperative CT images, highlighting the feasibility of conducting research on pathological staging using pathological slides and CT images.

Lymph node involvement significantly impacts the survival of gastric cancer patients and is a crucial factor in determining the appropriate treatment. This study aimed to evaluate the potential of enhanced computed tomography (CT)-based radiomics in predicting lymph node metastasis (LNM) and survival in patients with gastric cancer before surgery.

Retrospective analysis of clinical data from 192 patients diagnosed with gastric carcinoma was conducted. The patients were randomly divided into a training cohort (n = 128) and a validation cohort (n = 64). Radiomic features of CT images were extracted using the Pyradiomics software platform, and distinctive features were further selected using a Lasso Cox regression model. Features significantly associated with LNM were identified through univariate and multivariate analyses and combined with radiomic scores to create a nomogram model for predicting lymph node involvement before surgery. The predictive performance of radiomics features, CT-reported lymph node status, and the nomogram model for LNM were compared in the training and validation cohorts by plotting receiver operating characteristic (ROC) curves. High-risk and low-risk groups were identified in both cohorts based on the cut-off value of 0.582 within the radiomics evaluation scheme, and survival rates were compared.

Seven radiomic features were identified and selected, and patients were stratified into high-risk and low-risk groups using a 0.582 cut-off radiomics score. Univariate and multivariate analyses revealed that radiomics features, diabetes mellitus, Nutrition Risk Screening (NRS) 2002 score, and CT-reported lymph node status were significant predictors of LNM in patients with gastric cancer. A predictive nomogram model was developed by combining these predictors with the radiomics score, which accurately predicted LNM in gastric cancer patients before surgery and outperformed other models in terms of accuracy and sensitivity. The AUC values for the training and validation cohorts were 0.82 and 0.722, respectively. The high-risk and low-risk groups in both the training and validation cohorts showed significant differences in survival rates.

The radiomics nomogram, based on contrast-enhanced computed tomography (CECT ), is a promising non-invasive tool for preoperatively predicting LNM in gastric cancer patients and postoperative survival.

We aim to predict the pathological complete response (pCR) of neoadjuvant chemotherapy (NAC) in breast cancer patients by constructing a Nomogram based on radiomics models, clinicopathological features, and ultrasound features.

Ultrasound images of 464 breast cancer patients undergoing NAC were retrospectively analyzed. The patients were further divided into the training cohort and the validation cohort. The radiomics signatures (RS) before NAC treatment (RS1), after 2 cycles of NAC (RS2), and the different signatures between RS2 and RS1 (Delta-RS/RS1) were obtained. LASSO regression and random forest analysis were used for feature screening and model development, respectively. The independent predictors of pCR were screened from clinicopathological features, ultrasound features, and radiomics models by using univariate and multivariate analysis. The Nomogram model was constructed based on the optimal radiomics model and clinicopathological and ultrasound features. The predictive performance was evaluated with the receiver operating characteristic (ROC) curve.

We found that RS2 had better predictive performance for pCR. In the validation cohort, the area under the ROC curve was 0.817 (95%CI: 0.734-0.900), which was higher than RS1 and Delta-RS/RS1. The Nomogram based on clinicopathological features, ultrasound features, and RS2 could accurately predict the pCR value, and had the area under the ROC curve of 0.897 (95%CI: 0.866-0.929) in the validation cohort. The decision curve analysis showed that the Nomogram model had certain clinical practical value.

The Nomogram based on radiomics signatures after two cycles of NAC, and clinicopathological and ultrasound features have good performance in predicting the NAC efficacy of breast cancer.

To explore the value of the features of lymph nodes (LNs) with a short-axis diameter ≥6 mm in predicting lymph node metastasis (LNM) in advanced gastric adenocarcinoma (GAC) based on dual-energy CT (DECT) radiomics.

Data of patients with GAC who underwent radical gastrectomy and LN dissection were retrospectively analyzed. To ensure the correspondence between imaging and pathology, metastatic LNs were only selected from patients with pN3, nonmetastatic LNs were selected from patients with pN0, and the short-axis diameters of the enrolled LNs were all ≥6 mm. The traditional features of LNs were recorded, including short-axis diameter, long-axis diameter, long-to-short-axis ratio, position, shape, density, edge, and the degree of enhancement; univariate and multivariate logistic regression analyses were used to establish a clinical model. Radiomics features at the maximum level of LNs were extracted in venous phase equivalent 120 kV linear fusion images and iodine maps. Intraclass correlation coefficients and the Boruta algorithm were used to screen significant features, and random forest was used to build a radiomics model. To construct a combined model, we included the traditional features with statistical significance in univariate analysis and radiomics scores (Rad-score) in multivariate logistic regression analysis. Receiver operating curve (ROC) curves and the DeLong test were used to evaluate and compare the diagnostic performance of the models. Decision curve analysis (DCA) was used to evaluate the clinical benefits of the models.

This study included 114 metastatic LNs from 36 pN3 cases and 65 nonmetastatic LNs from 28 pN0 cases. The samples were divided into a training set (n=125) and a validation set (n=54) at a ratio of 7:3. Long-axis diameter and LN shape were independent predictors of LNM and were used to establish the clinical model; 27 screened radiomics features were used to build the radiomics model. LN shape and Rad-score were independent predictors of LNM and were used to construct the combined model. Both the radiomics model (area under the curve [AUC] of 0.986 and 0.984) and the combined model (AUC of 0.970 and 0.977) outperformed the clinical model (AUC of 0.772 and 0.820) in predicting LNM in both the training and validation sets. DCA showed superior clinical benefits from radiomics and combined models.

The models based on DECT LN radiomics features or combined traditional features have high diagnostic performance in determining the nature of each LN with a short-axis diameter of ≥6 mm in advanced GAC.

This research aimed to develop a radiomics-clinical nomogram based on enhanced thin-section CT radiomics and clinical features for the purpose of predicting the presence or absence of metastasis in lymph nodes among patients with resectable esophageal squamous cell carcinoma (ESCC).

This study examined the data of 256 patients with ESCC, including 140 cases with lymph node metastasis. Clinical information was gathered for each case, and radiomics features were derived from thin-section contrast-enhanced CT with the help of a 3D slicer. To validate risk factors that are independent of the clinical and radiomics models, least absolute shrinkage and selection operator logistic regression analysis was used. A nomogram pattern was constructed based on the radiomics features and clinical characteristics. The receiver operating characteristic curve and Brier Score were used to evaluate the model's discriminatory ability, the calibration plot to evaluate the model's calibration, and the decision curve analysis to evaluate the model's clinical utility. The confusion matrix was used to evaluate the applicability of the model. To evaluate the efficacy of the model, 1000 rounds of 5-fold cross-validation were conducted.

The clinical model identified esophageal wall thickness and clinical T (cT) stage as independent risk factors, whereas the radiomics pattern was built based on 4 radiomics features chosen at random. Area under the curve (AUC) values of 0.684 and 0.701 are observed for the radiomics approach and clinical model, respectively. The AUC of nomogram combining radiomics and clinical features was 0.711. The calibration plot showed good agreement between the incidence of lymph node metastasis predicted by the nomogram and the actual probability of occurrence. The nomogram model displayed acceptable levels of performance. After 1000 rounds of 5-fold cross-validation, the AUC and Brier score had median values of 0.702 (IQR: 0.65, 7.49) and 0.21 (IQR: 0.20, 0.23), respectively. High-risk patients (risk point >110) were found to have an increased risk of lymph node metastasis [odds ratio (OR) = 5.15, 95% CI, 2.95-8.99] based on the risk categorization.

A successful preoperative prediction performance for metastasis to the lymph nodes among patients with ESCC was demonstrated by the nomogram that incorporated CT radiomics, wall thickness, and cT stage.

This study demonstrates a novel radiomics-clinical nomogram for lymph node metastasis prediction in ESCC, which helps physicians determine lymph node status preoperatively.

Lymphovascular invasion (LVI) and perineural invasion (PNI) are important prognostic factors for gastric cancer (GC) that indicate an increased risk of metastasis and poor outcomes. Accurate preoperative prediction of LVI/PNI status could help clinicians identify high-risk patients and guide treatment decisions. However, prior models using conventional computed tomography (CT) images to predict LVI or PNI separately have had limited accuracy. Spectral CT provides quantitative enhancement parameters that may better capture tumor invasion. We hypothesized that a predictive model combining clinical and spectral CT parameters would accurately preoperatively predict LVI/PNI status in GC patients.

To develop and test a machine learning model that fuses spectral CT parameters and clinical indicators to predict LVI/PNI status accurately.

This study used a retrospective dataset involving 257 GC patients (training cohort, n = 172; validation cohort, n = 85). First, several clinical indicators, including serum tumor markers, CT-TN stages and CT-detected extramural vein invasion (CT-EMVI), were extracted, as were quantitative spectral CT parameters from the delineated tumor regions. Next, a two-step feature selection approach using correlation-based methods and information gain ranking inside a 10-fold cross-validation loop was utilized to select informative clinical and spectral CT parameters. A logistic regression (LR)-based nomogram model was subsequently constructed to predict LVI/PNI status, and its performance was evaluated using the area under the receiver operating characteristic curve (AUC).

In both the training and validation cohorts, CT T3-4 stage, CT-N positive status, and CT-EMVI positive status are more prevalent in the LVI/PNI-positive group and these differences are statistically significant (P < 0.05). LR analysis of the training group showed preoperative CT-T stage, CT-EMVI, single-energy CT values of 70 keV of venous phase (VP-70 keV), and the ratio of standardized iodine concentration of equilibrium phase (EP-NIC) were independent influencing factors. The AUCs of VP-70 keV and EP-NIC were 0.888 and 0.824, respectively, which were slightly greater than those of CT-T and CT-EMVI (AUC = 0.793, 0.762). The nomogram combining CT-T stage, CT-EMVI, VP-70 keV and EP-NIC yielded AUCs of 0.918 (0.866-0.954) and 0.874 (0.784-0.936) in the training and validation cohorts, which are significantly higher than using each of single independent factors (P < 0.05).

The study found that using portal venous and EP spectral CT parameters allows effective preoperative detection of LVI/PNI in GC, with accuracy boosted by integrating clinical markers.

To assess the performance of radiomics-based analysis of contrast-enhanced computerized tomography (CE-CT) images for distinguishing GS from gastric GIST.

Forty-nine patients with GS and two hundred fifty-three with GIST were enrolled in this retrospective study. CT features were evaluated by two associate chief radiologists. Radiomics features were extracted from portal venous phase images using Pyradiomics software. A non-radiomics dataset (combination of clinical characteristics and radiologist-determined CT features) and a radiomics dataset were used to build stepwise logistic regression and least absolute shrinkage and selection operator (LASSO) logistic regression models, respectively. Model performance was evaluated according to sensitivity, specificity, accuracy, and receiver operating characteristic (ROC) curve, and Delong's test was applied to compare the area under the curve (AUC) between different models.

A total of 1223 radiomics features were extracted from portal venous phase images. After reducing dimensions by calculating Pearson correlation coefficients (PCCs), 20 radiomics features, 20 clinical characteristics + CT features were used to build the models, respectively. The AUC values for the models using radiomics features and those using clinical features were more than 0.900 for both the training and validation groups. There were no significant differences in predictive performance between the radiomic and clinical data models according to Delong's test.

A radiomics-based model applied to CE-CT images showed comparable predictive performance to senior physicians in the differentiation of GS from GIST.

This study aims to develop and validate an innovative radiopathomics model that combines radiomics and pathomics features to effectively differentiate between stages I-II and stage III gastric cancer (pathological staging).

Our study included 200 patients with well-defined stages of gastric cancer divided into a training cohort (n = 140) and a test cohort (n = 60). Radiomics features were extracted from contrast-enhanced CT images using PyRadiomics, while pathomics features were obtained from whole slide images of pathological specimens through a fine-tuned deep learning model (ResNet-18). After rigorous feature dimensionality reduction and selection, we constructed radiomics models (SVM_rad, LR_rad, and MLP_rad) and pathomics models (SVM_path, LR_path, and MLP_path) utilizing support vector machine (SVM), logistic regression (LR), and multilayer perceptron (MLP) algorithms. The optimal radiomics and pathomics models were chosen based on comprehensive evaluation criteria such as ROC curves, Hosmer‒Lemeshow tests, and calibration curve tests. Feature patterns extracted from the best-performing radiomics model (MLP_rad) and pathomics model (SVM_rad) were integrated to create a powerful radiopathomics nomogram.

From a pool of 1834 radiomics features extracted from CT images, 14 were selected to construct radiomics models. Among these, the MLP_rad model exhibited the most robust predictive performance (AUC, training cohort: 0.843; test cohort: 0.797). Likewise, 10 pathomics features were chosen from 512 extracted from whole slide images to build pathomics models, with the SVM_path model demonstrating the highest predictive efficiency (AUC, training cohort: 0.937; test cohort: 0.792). The combined radiopathomics nomogram model exhibited optimal discriminative ability (AUC, training cohort: 0.951; test cohort: 0.837), as confirmed by decision curve analysis (DCA), which indicated superior clinical effectiveness.

This study presents a cutting-edge radiopathomics nomogram model designed to predict pathological staging in gastric cancer, distinguishing between stages I-II and stage III. Our research leverages preoperative CT images and histopathological slides to forecast gastric cancer staging accurately, potentially facilitating the estimation of staging before radical gastric cancer surgery in the future.

Lymph node metastasis (LNM) is associated with poor prognosis in patients with duodenal cancer (DC). However, the efficacy and optimal extent of lymph node (LN) dissection have not been thoroughly discussed.

A total of 98 consecutive patients with DC who underwent surgical resection (pancreatoduodenectomy, n = 55; partial resection, n = 32; pancreas-sparing total duodenectomy, n = 9) were retrospectively analyzed. The LN stations located upstream of the lymphatic flow were defined as Np stations according to tumor location, whereas the others were defined as Nd stations. The association between the dissection of each LN station and survival outcome was investigated using the efficacy index (EI; percentage of metastases to lymph nodes in each station multiplied by the 5-year survival rate of metastatic cases).

The survival of patients with LNM at the Nd stations (n = 6) was significantly worse than that of patients with LNM only at the Np stations (n = 20) (relapse-free survival, median survival time [MST], 6.0 vs. 48.4 months, p < 0.001; overall survival, MST, 15.1 vs. 96.0 months, p < 0.001). Multivariate analysis identified LNM at Nd stations as an independent prognostic factor for overall survival (hazard ratio 9.92; p = 0.015). The Np stations had a high EI (range, 8.34-20.88), whereas the Nd stations had an EI of 0, regardless of the tumor location.

LN dissection of the Np stations contributed to acceptable survival, whereas LNM of the Nd stations led to poor survival, possibly reflecting advanced tumor progression to systemic disease in patients with DC.

This two-center study aimed to establish a model for predicting the risk of lymph node metastasis in gastric cancer patients using machine learning (ML) and logistic regression (LR) algorithms, and to evaluate its predictive performance in clinical practice.

Data of a total of 369 patients who underwent radical gastrectomy in the Department of General Surgery of Affiliated Hospital of Xuzhou Medical University (Xuzhou, China) from March 2016 to November 2019 were collected and retrospectively analyzed as the training group. In addition, data of 123 patients who underwent radical gastrectomy in the Department of General Surgery of Jining First People's Hospital (Jining, China) were collected and analyzed as the verification group. Besides, 7 ML and logistic models were developed, including decision tree, random forest, support vector machine (SVM), gradient boosting machine (GBM), naive Bayes, neural network, and LR, in order to evaluate the occurrence of lymph node metastasis in patients with gastric cancer. The ML model was established following 10 cross-validation iterations within the training dataset, and subsequently, each model was assessed using the test dataset. The model's performance was evaluated by comparing the area under the receiver operating characteristic curve of each model.

Compared with the traditional logistic model, among the 7 ML algorithms, except for SVM, the other models exhibited higher accuracy and reliability, and the influences of various risk factors on the model were more intuitive.

For the prediction of lymph node metastasis in gastric cancer patients, the ML algorithm outperformed traditional LR, and the GBM algorithm exhibited the most robust predictive capability.

Whether or not a gastric cancer (GC) patient exhibits lymph node metastasis (LNM) is critical to accurately guiding their treatment and prognostic evaluation, necessitating the ability to reliably predict preoperative LNM status. The present meta-analysis sought to examine the diagnostic value of computed tomography (CT)-based predictive models as a tool to gauge the preoperative LNM status of patients with GC.

Relevant articles were identified in the PubMed, Web of Science, and Wanfang databases. These studies were used to conduct pooled analyses examining sensitivity, specificity, positive likelihood ratio (PLR), and negative likelihood ratio (NLR) values, and area under the curve values were computed for summary receiver operating characteristic curves.

The final meta-analysis incorporated data from 15 studies, all of which were conducted in China, enrolling 3,817 patients with GC (LNM+: 1790; LNM-: 2027). The developed CT-based predictive model exhibited respective pooled sensitivity, specificity, PLR, and NLR values of 84% (95% confidence interval [CI], 0.79-0.87), 81% (95% CI, 0.76-0.85), 4.39 (95% CI, 3.40-5.67), and 0.20 (95% CI, 0.16-0.26). The identified results were not associated with significant potential for publication bias ( P = 0.071). Similarly, CT-based analyses of LN status exhibited respective pooled sensitivity, specificity, PLR, and NLR values of 62% (95% CI, 0.53-0.70), 77% (95% CI, 0.72-0.81), 2.71 (95% CI, 2.20-3.33), and 0.49 (95% CI, 0.40-0.61), with no significant risk of publication bias ( P = 0.984).

Overall, the present meta-analysis revealed that a CT-based predictive model may outperform CT-based analyses alone when assessing the preoperative LNM status of patients with GC, offering superior diagnostic utility.

Precise preoperative evaluation of lymph node metastasis (LNM) is crucial for ensuring effective treatment for rectal cancer (RC). This research aims to develop a clinical-radiomics nomogram based on deep learning techniques, preoperative magnetic resonance imaging (MRI) and clinical characteristics, enabling the accurate prediction of LNM in RC.

Between January 2017 and May 2023, a total of 519 rectal cancer cases confirmed by pathological examination were retrospectively recruited from two tertiary hospitals. A total of 253 consecutive individuals were selected from Center I to create an automated MRI segmentation technique utilizing deep learning algorithms. The performance of the model was evaluated using the dice similarity coefficient (DSC), the 95th percentile Hausdorff distance (HD95), and the average surface distance (ASD). Subsequently, two external validation cohorts were established: one comprising 178 patients from center I (EVC1) and another consisting of 88 patients from center II (EVC2). The automatic segmentation provided radiomics features, which were then used to create a Radscore. A predictive nomogram integrating the Radscore and clinical parameters was constructed using multivariate logistic regression. Receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA) were employed to evaluate the discrimination capabilities of the Radscore, nomogram, and subjective evaluation model, respectively.

The mean DSC, HD95 and ASD were 0.857 ± 0.041, 2.186 ± 0.956, and 0.562 ± 0.194 mm, respectively. The nomogram, which incorporates MR T-stage, CEA, CA19-9, and Radscore, exhibited a higher area under the ROC curve (AUC) compared to the Radscore and subjective evaluation in the training set (0.921 vs. 0.903 vs. 0.662). Similarly, in both external validation sets, the nomogram demonstrated a higher AUC than the Radscore and subjective evaluation (0.908 vs. 0.735 vs. 0.640, and 0.884 vs. 0.802 vs. 0.734).

The application of the deep learning method enables efficient automatic segmentation. The clinical-radiomics nomogram, utilizing preoperative MRI and automatic segmentation, proves to be an accurate method for assessing LNM in RC. This approach has the potential to enhance clinical decision-making and improve patient care.

Research registry, identifier 9158, https://www.researchregistry.com/browse-the-registry#home/registrationdetails/648e813efffa4e0028022796/.

The objective of this study was twofold: firstly, to develop a convolutional neural network (CNN) for automatic segmentation of rectal cancer (RC) lesions, and secondly, to construct classification models to differentiate between different T-stages of RC. Additionally, it was attempted to investigate the potential benefits of rectal filling in improving the performance of deep learning (DL) models.

A retrospective study was conducted, including 317 consecutive patients with RC who underwent MRI scans. The datasets were randomly divided into a training set (n = 265) and a test set (n = 52). Initially, an automatic segmentation model based on T2-weighted imaging (T2WI) was constructed using nn-UNet. The performance of the model was evaluated using the dice similarity coefficient (DSC), the 95th percentile Hausdorff distance (HD95), and the average surface distance (ASD). Subsequently, three types of DL-models were constructed: Model 1 trained on the total training dataset, Model 2 trained on the rectal-filling dataset, and Model 3 trained on the non-filling dataset. The diagnostic values were evaluated and compared using receiver operating characteristic (ROC) curve analysis, confusion matrix, net reclassification index (NRI), and decision curve analysis (DCA).

The automatic segmentation showed excellent performance. The rectal-filling dataset exhibited superior results in terms of DSC and ASD (p = 0.006 and 0.017). The DL-models demonstrated significantly superior classification performance to the subjective evaluation in predicting T-stages for all test datasets (all p < 0.05). Among the models, Model 1 showcased the highest overall performance, with an area under the curve (AUC) of 0.958 and an accuracy of 0.962 in the filling test dataset.

This study highlighted the utility of DL-based automatic segmentation and classification models for preoperative T-stage assessment of RC on T2WI, particularly in the rectal-filling dataset. Compared with subjective evaluation, the models exhibited superior performance, suggesting their noticeable potential for enhancing clinical diagnosis and treatment practices.

To compare computed tomography (CT)- and magnetic resonance imaging (MRI)-based multiparametric radiomics models and validate a multi-modality, multiparametric clinical-radiomics nomogram for individual preoperative prediction of lymph node metastasis (LNM) in rectal cancer (RC) patients.

234 rectal adenocarcinoma patients from our retrospective study cohort were randomly selected as the training (n = 164) and testing (n = 70) cohorts. The radiomics features of the primary tumor were extracted from the non-contrast enhanced computed tomography (NCE-CT), the enhanced computed tomography (CE-CT), the T2-weighted imaging (T2WI) and the gadolinium contrast-enhanced T1-weighted imaging (CE-TIWI) of each patient. Three kinds of models were constructed based on training cohort, including the Clinical model (based on the clinical features), the radiomics models (based on NCE-CT, CE-CT, T2WI, CE-T1WI, CT, MRI, CT combing with MRI) and the clinical-radiomics models (based on CT or MRI radiomics model combing with clinical data) and Clinical-IMG model (based on CT and MRI radiomics model combing with clinical data). The performances of the 11 models were evaluated via the area under the receiver operator characteristic curve (AUC), accuracy, sensitivity, and specificity in the training and validation cohort. Differences in the AUCs among the 11 models were compared using DeLong's test. Finally, the optimal model (Clinical-IMG model) was selected to create a radiomics nomogram. The performance of the nomogram to evaluate clinical efficacy was verified by ROC curves and decision curve analysis (DCA).

The MRI radiomics model in the validation cohort significantly outperformed than CT radiomics model (AUC, 0.785 vs. 0.721, p<0.05). The Clinical-IMG nomogram had the highest prediction efficiency than all other predictive models (p<0.05), of which the AUC was 0.947, the sensitivity was 0.870 and the specificity was 0.884.

MRI radiomics model performed better than both CT radiomics model and Clinical model in predicting LNM of RC. The clinical-radiomics nomogram that combines the radiomics features obtained from both CT and MRI along with preoperative clinical characteristics exhibits the best diagnostic performance.

Pulmonary nodule growth rate assessment is critical in the management of subsolid pulmonary nodules (SSNs) during clinical follow-up. The present study aimed to develop a model to predict the growth rate of SSNs.

A total of 273 growing SSNs with clinical information and 857 computed tomography (CT) scans were retrospectively analyzed. The images were randomly divided into training and validation sets. All images were categorized into fast-growth (volume doubling time (VDT) ≤ 400 days) and slow-growth (VDT > 400 days) groups. Models for predicting the growth rate of SSNs were developed using radiomics and clinical features. The models' performance was evaluated using the area under the curve (AUC) values for the receiver operating characteristic curve.

The fast- and slow-growth groups included 108 and 749 scans, respectively, and 10 radiomics features and three radiographic features (nodule density, presence of spiculation, and presence of vascular changes) were selected to predict the growth rate of SSNs. The nomogram integrating radiomics and radiographic features (AUC = 0.928 and AUC = 0.905, respectively) performed better than the radiographic (AUC = 0.668 and AUC = 0.689, respectively) and radiomics (AUC = 0.888 and AUC = 0.816, respectively) models alone in both the training and validation sets.

The nomogram model developed by combining radiomics with radiographic features can predict the growth rate of SSNs more accurately than traditional radiographic models. It can also optimize clinical treatment decisions for patients with SSNs and improve their long-term management.