Purpose To develop and validate a CT-based radiomics model for differentiating primary gastric lymphoma (PGL) from Borrmann type IV gastric cancer (GC). Materials and methods A total of 136 patients with pathologically confirmed PGL (n = 56) and Borrmann type IV GC (n = 80) were retrospectively enrolled between January 2016 and May 2022. The cohort was randomly partitioned into a training set (n = 95) and a testing set (n = 41) at a 7:3 ratio. Radiomics features were extracted from unenhanced, arterial, venous, double-phase (arterial + venous), and three-phase (unenhanced + arterial + venous) CT images. After feature selection using the Least Absolute Shrinkage and Selection Operator, radiomics models were constructed via logistic regression. A clinical-radiomics model was developed through multivariate analysis. The models were evaluated using Receiver Operating Characteristic (ROC) curves, calibration curves with the Hosmer-Lemeshow test, and decision Curve Analysis (DCA) for clinical net benefit. Results Clinical model comprised of high-enhanced serosa sign, normalized CT value on venous phase, and perigastric fat infiltration showed good performance with AUCs of 0.902 (training set) and 0.878 (testing set). Among the radiomics models, the three-phase model outperformed others (AUC: 0.871 training, 0.865 testing). The clinical-radiomics combined model further improved discriminatory performance, achieving AUCs of 0.960 and 0.932 in the training and testing sets, respectively. DCA confirmed that the combined model provided the highest clinical net benefit. Conclusion Clinical-radiomics model incorporating three-phase radiomics signatures and CT findings achieved satisfactory performance for differentiating PGL from Borrmann type IV GC, serving as a reliable non-invasive tool for clinical decision-making.

RATIONALE AND OBJECTIVES:To assess the value of quantitative dual-energy CT (DECT) parameters for predicting resectable rectal cancer recurrence. MATERIALS AND METHODS:This retrospective study included 264 consecutive patients (182 in training cohort and 82 in validation cohort) with resectable rectal cancer who underwent upfront surgery without neoadjuvant therapy and preoperative contrast-enhanced CT at two centers between May 2019 and July 2022. DECT quantitative parameters, including iodine concentration (IC), normalized iodine concentration (NIC), electron density (Rho), effective atomic number (Z), spectral slope (K) and extracellular volume fraction (ECV) derived from both arterial and venous phases, were analysed. Univariate and multivariate Cox proportional hazards models were used to identify independent risk predictors of recurrence. A combined model was established and evaluated using the C-index, time-dependent ROC curves, calibration, decision curve analysis (DCA). The Kaplan-Meier survival curves were compared using the log-rank test. RESULTS:Recurrence occurred in 47 (25.8%) training cases and 21 (25.6%) in validation cases. The extracellular volume at venous phase (ECVV) (HR=1.82, 95%CI: 1.45-2.28, p<0.001), extramural venous invasion (EMVI) (HR=3.37, 95%CI: 1.83-6.20, p<0.001), carcinoembryonic antigen (CEA) (HR=1.87, 95%CI: 1.02-3.44, p=0.042), and carbohydrate antigen 19-9 (CA19-9) (HR=2.49, 95%CI: 1.31-4.74, p=0.005) were verified as significant predictors of recurrence. The combined model yielded a C-index of 0.792 (95% CI: 0.658-0.894) for predicting 3-year recurrence. Kaplan-Meier analysis showed significant differences in recurrence-free survival between the model-defined high- and low-risk groups (log-rank p values ranging from <0.001 to 0.020). CONCLUSION:Combining DECT-derived ECVV, EMVI, CEA, and CA19-9 demonstrates improved predictive discrimination for predicting rectal cancer recurrence.

Abstract Objective To evaluate the value of time-dependent diffusion MRI (td-dMRI) derived microstructural parameters for predicting lymphovascular invasion (LVI) in rectal cancer. Materials and methods Eighty-four resectable rectal cancer patients (stage T1, T2, T3a, T3b, and T4a) who underwent preoperative td-dMRI between March 2023 and June 2025 without neoadjuvant therapy were enrolled. Manual segmentation of tumors was performed by an experienced radiologist on each tumor’s largest cross-sectional area. Microstructural parameters (intracellular volume fraction (ICVF), cell diameter, extracellular diffusivity and cellularity) were fitted using the limited spectrally edited diffusion model implemented in MATLAB (MathWorks, Inc.). Apparent diffusion coefficient (ADC) values at different diffusion times, relative ADC, ADC ratio, and MRI-reported extramural vascular invasion (EMVI) were also evaluated. Mann–Whitney U test was used to evaluate parameter differences between LVI-positive and LVI-negative. Logistic regression and receiver operating characteristic (ROC) curves (with DeLong test) were used to identify predictors of LVI and diagnostic performance. Results Of 84 participants (median age, 66 years; IQR, 60–70 years; 50 male), 30 were LVI-positive and 54 LVI-negative. ICVF, cell diameter, and cellularity were significantly higher in LVI-positive cases (all p < 0.05). MRI-EMVI (OR = 3.251), ICVF (OR = 8.137), and cellularity (OR = 1.159) were independent risk factors of LVI. The combined model integrating MRI-reported EMVI, cellularity, and ICVF achieved an area under the ROC curve (AUC) of 0.860, outperforming individual parameters including MRI-reported EMVI (AUC = 0.730), ICVF (AUC = 0.815), cellularity (AUC = 0.792) and ADC measurements (AUC = 0.631–0.710) (all p < 0.05). Conclusion td-dMRI-derived parameters, especially ICVF and cellularity combined with MRI-reported EMVI, show potential as noninvasive biomarkers for LVI prediction in rectal cancer. Critical relevance statement This study develops a preoperative time-dependent diffusion MRI-based microstructure parameters model that diagnoses and predicts lymphovascular invasion of rectal cancer, improving diagnostic accuracy and advancing personalized treatment strategies in clinical radiology. Key Points The time-dependent diffusion MRI-derived microstructural parameters model and clinical data for predicting lymphovascular invasion in rectal cancer. The combined model outperforms single-modality models with 0.860 AUC and 96.3% specificity. The combined model provides a noninvasive, reliable tool for personalized lymphovascular invasion diagnosis and treatment planning. Graphical Abstract

To investigate the clinical value of Time-dependent diffusion MRI (td-dMRI) quantitative parameters in differentiating between rectal neuroendocrine neoplasms (rNEN) and rectal adenocarcinoma (RAC) and in distinguishing between high and low Ki-67 expression in RAC. This two-center study prospectively enrolled 90 patients with rectal tumors who underwent td-dMRI (January 2024 to March 2025), including 85 with RAC (development cohort, n = 61, validation cohort, n = 24) and 5 with rNEN. The td-dMRI-derived microstructural parameters were estimated using the IMPULSED model, and apparent diffusion coefficient (ADC) values were measured at different diffusion times. Firth’s logistic regression was performed for rNEN-associated parameter selection, while standard logistic regression identified independent predictors to build a combined model for high Ki-67 stratification. Receiver operating characteristic analysis evaluated diagnostic performance. The td-dMRI-derived microstructural parameters were validated against histopathologic measurements. Significant differences in several td-dMRI parameters were observed between rNENs and RACs, as well as between groups with high and low Ki-67 expression. In the small rNEN cohort, td-dMRI parameters showed promising preliminary diagnostic performance in distinguishing rNEN from RAC (AUC = 0.882). For predicting high Ki-67 expression in RAC, the combined model yielded an AUC of 0.916 in the development cohort and 0.844 in the validation cohort. Microstructural parameters showed strong correlations with histopathology: cell diameter, cellularity, and intracellular volume fraction reached Pearson correlation coefficients (r) of 0.808, 0.773, and 0.765, respectively (all p < 0.001). Quantitative td-dMRI parameters may serve as a noninvasive tool for evaluating rectal tumor histological types (rNEN vs. RAC) and predicting Ki-67 expression in RAC.

This meta-analysis aims to evaluate the diagnostic performance of magnetic resonance imaging (MRI)-based artificial intelligence (AI) in the preoperative detection of lymph node metastasis (LNM) in patients with rectal cancer and to compare it with the diagnostic performance of radiologists. A thorough literature search was conducted across PubMed, Embase, and Web of Science to identify relevant studies published up to September 2024. The selected studies focused on the diagnostic performance of MRI-based AI in detecting rectal cancer LNM. A bivariate random-effects model was employed to calculate pooled sensitivity and specificity, each reported with 95

Background:Acute pancreatitis (AP), recurrent acute pancreatitis (RAP), and chronic pancreatitis (CP) are increasingly viewed as stages in a continuous disease spectrum when the underlying etiology remains unresolved. Previous studies have investigated the effect of different etiologies on AP severity, but few have specifically examined the clinical and radiologic characteristics of RAP stratified by etiology. This study aimed to investigate the computed tomography (CT) features of RAP stratified by etiology. Methods:We retrospectively analyzed the data of 683 RAP patients who underwent contrast-enhanced computed tomography (CECT) at three tertiary hospitals between January 2015 and December 2019. The patients were categorized into five etiologic groups: alcoholic, cholelithiasis-related, hypertriglyceridemia-related, multifactorial, and idiopathic. Clinical and imaging data, including demographic data, 2012 revised Atlanta classification (RAC), Acute Physiology and Chronic Health Evaluation (APACHE) II scores, modified computed tomography severity index (MCTSI) scores, extrapancreatic inflammation on computed tomography (EPIC) scores, the presence and extent of pancreatic necrosis, and local complications, were compared across groups using Kruskal-Wallis and Chi-squared (or Fisher's exact) tests, with Bonferroni correction for multiple comparisons. Results:Among the 683 patients {449 male, 234 female; median age: 45 years [interquartile range (IQR), 39-52 years]; median hospital stay: 11 days (IQR, 7-15 days)} included in the study, hypertriglyceridemia was the most common cause of RAP (50.95%), followed by idiopathic causes (18.74%), cholelithiasis (15.37%), alcoholism (9.37%), and multiple causes (5.56%). Patients with hypertriglyceridemic or multifactorial RAP had higher triglyceride [median (IQR): 17.99 (12.12-25.77) vs. 1.33 (0.86-2.00) mmol/L] and blood glucose [median (IQR): 10.55 (7.44-14.33) vs. 7.43 (5.74-10.12) mmol/L] levels, as well as a higher prevalence of pre-existing diabetes (30.46% vs. 9.52%) (all P<0.001). Compared to the patients with biliary RAP, those with hyperlipidemic RAP exhibited milder CT features, including lower rates of necrotizing pancreatitis (NP) (33.05% vs. 48.57%), combined necrosis (CN) (25.57% vs. 40.00%), pancreatic necrosis >50% (3.74% vs. 14.29%), and severe disease (23.85% vs. 40.00%) as defined by the MCTSI (all P<0.05). These patients also had lower EPIC scores [median (IQR): 3 (2-5) vs. 4 (2-5.5)] and shorter hospital stays [median (IQR): 10 (7-15) vs. 13 (9.5-18) days] (both P<0.05). No significant differences in the severity indicators were observed among alcoholic, multifactorial, and idiopathic groups. Alcohol-related RAP occurred predominantly in males (96.9%), while biliary RAP was more frequent in older female patients (both P<0.05). Conclusions:RAP presents with distinct clinical and CT features depending on its etiology. Hypertriglyceridemic RAP is associated with milder disease severity than biliary RAP. Our findings provide new insights into the etiology-specific manifestations of RAP, and may inform future research on its underlying mechanisms and long-term outcomes.

Background Lymph node metastasis (LNM) is a poor prognostic predictor and is highly correlated with local recurrence in rectal cancer patients. Objective To investigate the value of radiomics from dual-energy CT-derived iodine maps for the preoperative prediction of LNM in rectal cancer patients. Methods A total of 176 patients were enrolled in this study (training group, n = 123; validation group, n = 53). A radiomic signature was constructed via support vector machine (SVM) modeling. Seven models, including a clinical feature model (Model 1), an arterial model (Model 2), a venous model (Model 3), an arterial-venous model (Model 4), an arterial–clinical model (Model 5), a venous-clinical model (Model 6) and an arterial–venous–clinical model (Model 7), were established via logistic regression modeling. Diagnostic performance was assessed via receiver operating characteristic (ROC) curves. Results Tumor location and carcinoembryonic antigen levels were used to construct Model 1 (training group, AUC [area under the ROC curve] = 0.721, 95% CI [confidence intervals], 0.630–0.813; validation group, AUC = 0.729, 95% CI, 0.593–0.865). Model 6 and Model 7 further improved the discriminatory performance in the training (AUC = 0.850 and 0.869, 95% CI, 0.782–0.919 and 0.807–0.932, respectively; p = 0.250) and validation groups (AUC = 0.780 and 0.716, 95% CI, 0.653–0.906 and 0.576–0.856, respectively; p = 0.115). Moreover, decision curve analysis revealed a greater net benefit with Model 6. Conclusions The combination of radiomic features based on dual-energy CT-derived iodine maps and clinical features provides better diagnostic performance for predicting LNM in rectal cancer patients.

Objective To investigate the utility values of time-dependent diffusion MRI (td-dMRI) in predicting differentiation degree and Ki-67 expression in rectal cancer. Methods Seventy-three resectable rectal cancer patients who underwent td-dMRI examination were consecutively enrolled. Intracellular volume fraction (ICVF), cell diameter, extracellular diffusivity (Dex), cellularity, apparent diffusion coefficient (ADC) values at different diffusion times, relative ADC value, and ADC ratio were investigated. Intraclass correlation coefficients and Bland-Altman plots were used to determine repeatability. Mann-Whitney U test, logistic regression analysis, receiver operating characteristic (ROC), and the Delong’s test were used for parameter differences evaluation, independent risk factor identifying, diagnostic ability assessing, and the area under the ROC curve (AUC) comparations for differentiation degree and Ki-67 expressions, respectively. Results The cellularity, ADCPGSE, and ADC40HZ in the low differentiation degree group were significantly higher than moderate to high differentiation degree group (all P < 0.05). While, ICVF was significantly lower in low differentiation group. Tumor length, ICVF, cellularity, ADCPGSE and ADC40HZ were independent risk factors for low differentiation degree. The combined model achieved the highest diagnostic performance, with an AUC of 0.831 (95 %CI: 0.73, 0.93) for differentiation. For Ki-67 expression, ICVF in the high Ki-67 expression group was lower than that of low Ki-67 expression group (P = 0.001). MR-reported lymph node stage, extramural vascular invasion (EMVI), and ICVF were independent clinical risk factors for predicting high Ki-67 expression. The diagnostic ability of combined indicators reached an AUC of 0.820 (95 %CI: 0.71, 0.93) surpassed the individual of ICVF (AUC = 0.750, 95 %CI: 0.62, 0.88, P < 0.01). Conclusion Td-dMRI-derived microstructural parameters may provide an alternative form of non-invasive imaging marker of differentiation degree and Ki-67 expression in rectal cancer and provide valuable information for treatment decisions.

To determine whether quantitative parameters derived from dual-energy CT (DECT) could predict prognosis in patients with resectable rectal cancer (RC). One hundred and thirty-four patients (recurrence/distant metastasis group, n = 36; non-metastasis/non-recurrence group, n = 98) with RC who underwent radical resection and DECT were retrospectively included. DECT quantitative parameters, including iodine concentration (IC), normalized iodine concentration (NIC), electron density (Rho), effective atomic number (Zeff), dual-energy index (DEI), the slope of the spectral Hounsfield unit curve (λHU) on arterial and venous phase images. Univariate and multivariate Cox proportional hazards models were employed to identify independent risk factors of prognosis. The area under the receiver operating characteristic curve (AUC) was used to assess the performance. Disease-free survival (DFS) curves were constructed using the Kaplan–Meier method. Patients in the metastasis/recurrence group had higher Rho in arterial phase (A-Rho), NIC in venous phase (V-NIC), Rho in venous phase (V-Rho), Zeff in venous phase (V-Zeff), λHU in venous phase (V-λHU), pT stage, pN stage, serum carcinoembryonic antigen (CEA), carbohydrate antigen-199 levels and more frequent in extramural venous invasion than those in non-metastasis/non-recurrence group (all p < 0.05). V-NIC, V-λHU, and CEA were independent risk factors of recurrence/distant metastasis (all p < 0.05). The AUC of combined indicator integrating three independent risk factors achieved the best diagnostic performance (AUC = 0.900). In stratified survival analysis, patients with high V-NIC, V-λHU, and CEA had lower 3-year DFS than those with low V-NIC, V-λHU, and CEA. Combining V-NIC, V-λHU, and CEA could be used to noninvasively predict prognosis in resectable RC. Question TNM staging fails to accurately prognosticate; can quantitative parameters derived from dual-energy CT predict prognosis in patients with resectable rectal cancer? Findings Normalized iodine concentration (V-NIC) and the slope of the spectral Hounsfield unit curve in venous phase (V-λHU), and carcinoembryonic antigen (CEA) are independent risk factors for recurrence/metastasis. Clinical relevance The combined indicator integrating V-NIC, V-λHU, and CEA could predict 3-year disease-free survival in patients with resectable rectal cancer and could aid in postoperative survival risk stratification to guide personalized treatment.

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.

BACKGROUND:Increasing remnant liver volume before major liver resection is an effective measure to reduce postoperative adverse events of hepatocellular carcinoma (HCC). We aimed to provide evidence for optimal management of HCC patients with insufficient future remnant liver volume (FRLV). METHODS:A comprehensive search of various large medical databases, research registry platforms, and gray literature was performed up to May 2023. All comparative studies grouped by preoperative hepatic augmentation (PHA) and transarterial chemoembolization (TACE) were included. A random-effects model was used for meta-analysis, and the heterogeneity of the results was quantitatively assessed by funnel plots, sensitivity analyses, and subgroup analyses. RESULTS:A total of eight comparative studies were selected for inclusion in this analysis, including 3523 patients. 5-year overall survival (hazard ratio [HR] = 1.52, 95% confidence intervals [CI] = 1.07-2.15) and disease-free survival (HR = 1.72, 95% CI = 1.40-2.10) were significantly different between the PHA and TACE groups. There was no significant difference between PHA and TACE with respect to 90-day mortality, postoperative complication rate, or serious complication rate (p > 0.05). In subgroup analysis, compared with portal vein embolization, associating liver partition and portal vein ligation was highly associated with longer survival and fewer recurrences (p < 0.05). None of the above results exhibited obvious bias or heterogeneity. CONCLUSIONS:This study demonstrates that PHA allows for radical liver resection for HCC patients with insufficient FRLV without increasing the incidence of postoperative adverse events, which can effectively improve patient outcomes and delay tumor recurrence.

This study aims to compare, select, and investigate MRI-based multiregional radiomics model to predict pathologic complete response (pCR) in Locally advanced rectal cancer (LARC) patients after neoadjuvant chemoradiotherapy (nCRT). This retrospective study included 245 patients who underwent rectal MRI examination before nCRT were recruited and split into training (hospital 1, n = 177) and external validation cohort (hospital 2, n = 68). Pretreatment T2WI and ADC images were used to manually delineate volumetric region of interest. Intratumoral, peritumoral-2 mm, peritumoral-3 mm, peritumoral-5 mm, and peritumoral-mesorectal fat (MRF) radiomics features were extracted. Clinical model was built based on clinical and MRI features. Diagnosis performance was compared among models. 3-year recurrence-free survival (RFS) was evaluated by Kaplan-Meier curve. cN stage (odds ratio = 2.62, 95

Background:Multiple magnetic resonance imaging (MRI) features suggestive of placenta accreta spectrum (PAS) disorders exist. However, the impact of placental location on clinical characteristics and MRI features in PAS has not been fully explored. The aim of this study was to explore the difference of MRI signs in different placental position in PAS disorders. Methods:We retrospectively reviewed surgically or pathologically confirmed PAS cases at Sichuan Provincial People's Hospital from 2016 to 2021. Placental location was categorized based on MRI as anterior, posterior, or anterior/posterior. MRI features were thoroughly reviewed and compared. Results:A total of 262 patients were included in the study, comprising 38 (14.50%) with placenta accreta, 120 (45.80%) with placenta increta, 21 (8.02%) with placenta percreta, and 83 (31.68%) with normal placentas. The distribution of placental location was as follows: 32.06% posterior, 48.09% anterior, and 19.85% anterior/posterior. Placental location varied significantly between patients with and without PAS disorders and among patients with different PAS subtypes (P<0.05). The prevalence of placental bulge was higher in anterior and anterior/posterior placentas than in the posterior placentas (P<0.05). Moreover, T2 dark bands, placental heterogeneity, abnormal intraplacental vascularity, focal exophytic mass, and bladder wall interruption varied among different subtypes of PAS disorders (P<0.05). Conclusions:Placental bulge emerged as the only MRI sign that exhibited differences based on placental location. Furthermore, MRI features demonstrated variations across different subtypes of PAS.

This study aimed to assess the diagnostic accuracy of radiomics for predicting osteoporosis and the quality of radiomic studies. The study protocol was prospectively registered on PROSPERO (CRD42023425058). We searched PubMed, EMBASE, Web of Science, and Cochrane Library databases from inception to June 1, 2023, for eligible articles that applied radiomic techniques to diagnosing osteoporosis or abnormal bone mass. Quality and risk of bias of the included studies were evaluated with radiomics quality score (RQS), METhodological RadiomICs Score (METRICS), and Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tools. The data analysis utilized the R program with mada, metafor, and meta packages. Ten retrospective studies with 5926 participants were included in the systematic review and meta-analysis. The overall risk of bias and applicability concerns for each domain of the studies were rated as low, except for one study which was considered to have a high risk of flow and time bias. The mean METRICS score was 70.1% (range 49.6-83.2%). There was moderate heterogeneity across studies and meta-regression identified sources of heterogeneity in the data, including imaging modality, feature selection method, and classifier. The pooled diagnostic odds ratio (DOR) under the bivariate random effects model across the studies was 57.22 (95% CI 27.62-118.52). The pooled sensitivity and specificity were 87% (95% CI 81-92%) and 87% (95% CI 77-93%), respectively. The area under the summary receiver operating characteristic curve (AUC) of the radiomic models was 0.94 (range 0.8 to 0.98). The results supported that the radiomic techniques had good accuracy in diagnosing osteoporosis or abnormal bone mass. The application of radiomics in osteoporosis diagnosis needs to be further confirmed by more prospective studies with rigorous adherence to existing guidelines and multicenter validation.

RATIONALE AND OBJECTIVES:To develop and validate multiple machine learning predictive models incorporating clinical features and pretreatment multiparametric magnetic resonance imaging (MRI) radiomic features for predicting treatment response to transarterial chemoembolization combined with molecular targeted therapy plus immunotherapy in unresectable hepatocellular carcinoma (HCC). MATERIALS AND METHODS:This retrospective study involved 276 patients with unresectable HCC who received combination therapy from 4 medical centers. Patients were divided into one training cohort and two independent external validation cohorts. 16 radiomic features from six multiparametric MRI sequences and 2 clinical features were used to build six machine learning models. The models were evaluated using the area under the curve (AUC), decision curve analysis, and incremental predictive value. RESULTS:Alpha-fetoprotein and neutrophil-to-lymphocyte ratio are clinical independent predictors of treatment response. In the training cohort and two external validation cohorts, the AUCs and 95% confidence intervals for predicting treatment response were respectively 0.782 (0.698-0.857) 0.695 (0.566-0.823), and 0.679 (0.542-0.810) for the clinical model; 0.942 (0.903-0.974), 0.869 (0.761-0.949), and 0.868 (0.769-0.942) for the radiomics model; and 0.956 (0.920-0.984), 0.895 (0.810-0.967), and 0.892 (0.804-0.957) for the combined clinical-radiomics model. In the three cohorts, the incremental predictive value of the radiomics model over the clinical model was 49.2% (P < 0.001), 28.8% (P < 0.001), and 31.5% (P < 0.001). CONCLUSION:The combined clinical-radiomics model may provide a reliable and non-invasive tool to predict individual treatment responses and guide and improve clinical decision-making in combination therapy of HCC patients.

Background MRI features may be associated with adverse maternal outcome in patients with placenta accreta spectrum (PAS) disorders even with abdominal aortic balloon occlusion (AABO). Purpose This study aimed to identify risk factors of MRI for association with adverse maternal outcome in patients with PAS disorders after AABO. Study Type Retrospective. Population Clinical and MRI features of 80 patients were retrospectively reviewed from October 2016 to August 2021. A total of 40 patients had adverse maternal outcomes including intrapartum/peripartum bleeding >1000 mL and/or emergency hysterectomy after AABO. Sequence Half‐Fourier acquisition single‐shot turbo spin echo and gradient echo imaging True fast imaging with steady‐state precession (True‐FISP) at 1.5T MR scanner. Assessment MRI features were evaluated by three radiologists and were tested for any association with adverse maternal outcome. Statistical Tests Interobserver agreement was calculated with kappa ( k ) statistics. Association between MRI features and adverse maternal outcomes were evaluated by univariate and multivariate analyses. A nomogram was constructed based on the logistic regression. Results The interobserver agreement ranged from fair to substantial ( k = 0.379–0.783). Multivariate analyses revealed that short cervical length (OR: 4.344), abnormal intraplacental vascularity (OR: 6.005), placental bulge (OR: 9.085), and myometrial interruption (OR: 9.550) were independent risk factors for adverse maternal outcomes. The combination of four risk factors together demonstrated the highest AUC of 0.851 (95% CI 0.769–0.933) with a sensitivity and specificity of 77.5% and 72.5%, respectively and then a nomogram composed of the above four risk factors was constructed to represent the probability of adverse maternal outcome. Data Conclusion The nomogram demonstrated the association between MRI features and patient's poor outcome after undergoing AABO and C‐section delivery for PAS. Evidence Level 4 Technical Efficacy Stage 3

PurposeTo establish and evaluate multiregional T2-weighted imaging (T2WI)-based clinical-radiomics model for predicting lymph node metastasis (LNM) and prognosis in patients with resectable rectal cancer. MethodsA total of 346 patients with pathologically confirmed rectal cancer from two hospitals between January 2019 and December 2021 were prospectively enrolled. Intra- and peritumoral features were extracted separately, and least absolute shrinkage and selection operator regression was applied for feature selection. Radiomics signatures were built using the selected features from different regions. The clinical-radiomic nomogram was developed by combining the intratumoral and peritumoral radiomics signatures score (radscore) and the most predictive clinical parameters. The diagnostic performances of the nomogram and clinical model were evaluated using the area under the receiver operating characteristic curve (AUC). The prognostic model for 3-year recurrence-free survival (RFS) was constructed using univariate and multivariate Cox analysis. ResultsThe intratumoral radscore (radscore 1) included four features, the peritumoral radscore (radscore 2) included five features, and the combined intratumoral and peritumoural radscore (radscore 3) included ten features. The AUCs for radscore 3 were higher than that of radscore 1 in training cohort (0.77 vs. 0.71, P=0.182) and internal validation cohort (0.76 vs. 0.64, P=0.041). The AUCs for radscore 3 were higher than that of radscore 2 in training cohort (0.77 vs. 0.74, P=0.215) and internal validation cohort (0.76 vs. 0.68, P=0.083). A clinical-radiomic nomogram showed a higher AUC compared with the clinical model in training cohort (0.84 vs. 0.67, P<0.001) and internal validation cohort (0.78 vs. 0.64, P=0.038) but not in external validation (0.72 vs. 0.76, P=0.164). Multivariate Cox analysis showed MRI-reported extramural vascular invasion (EMVI) (HR=1.099, 95%CI: 0.462-2.616; P=0.031) and clinical-radiomic nomogram-based LNM (HR=2.232, 95%CI:1.238-7.439; P=0.017) were independent risk factors for assessing 3-year RFS. Combined clinical-radiomic nomogram based LNM and MRI-reported EMVI showed good performance in training cohort (AUC=0.748), internal validation cohort (AUC=0.706) and external validation (AUC=0.688) for predicting 3-year RFS. ConclusionA clinical-radiomics nomogram exhibits good performance for predicting preoperative LNM. Combined clinical-radiomic nomogram based LNM and MRI-reported EMVI showed clinical potential for assessing 3-year RFS.

Background: Placenta accreta spectrum (PAS) disorder encompasses a spectrum of pathologies, from placenta accreta to placenta percreta, which is usually associated with postpartum hemorrhage (PPH).Methods: This cross-sectional study enrolled 109 patients suspected of having PAS disorders based on previous ultrasound results or clinical risk factors from November 2018 to March 2022 in Sichuan Provincial People's Hospital. Of the 109 patients, 34 had PPH and 75 did not have PPH. Magnetic resonance imaging (MRI) including diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), and diffusion kurtosis imaging (DKI) was performed for each patient and the apparent diffusion coefficient (ADC) from DWI, perfusion fraction (f), pure diffusion coefficient (D), and pseudo-diffusion coefficient (D*) from IVIM, and mean diffusion kurtosis (MK) and mean diffusion coefficient (MD) from DKI were measured and compared. The correlation between the DWI parameters and estimated blood loss (EBL) during surgery was identified using correlation analysis. The diagnostic performance for predicting PPH was compared between the two methods.Results: The amount of bleeding during delivery was positively correlated with D [r=0.331, P<0.001, 95% confidence interval (CI): 0.170 to 0.477], D* (r=0.389, P<0.001, 95% CI: 0.207 to 0.527), f (r=0.222, P=0.02, 95% CI: 0.036 to 0.398), and MD (r=0.277, P=0.003, 95% CI: 0.108 to 0.439), but negatively correlated with MK (r=-0.280, P=0.003, 95% CI: -0.431 to -0.098). In predicting PPH, multivariate analyses showed the independent risk factors were placenta previa and D; the area under the curve (AUC) was 0.795 (95% CI: 0.711 to 0.878) when the two risk factors were combined together. Conclusions: IVIM and DKI parameters are correlated with EBL. The combined use of placenta previa and D are helpful for predicting PPH in patients at high risk of PAS disorders.

Purpose To investigate the diagnostic value of monoexponential, biexponential, and diffusion kurtosis MR imaging (MRI) in distinguishing invasive placentas. Methods A total of 53 patients with invasive placentas and 47 patients with noninvasive placentas undergoing conventional diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), and diffusion kurtosis imaging (DKI) were retrospectively enrolled. The mean, minimum, and maximum parameters including the apparent diffusion coefficient (ADC) and exponential ADC (eADC) from standard DWI, diffusion kurtosis (MK), and diffusion coefficient (MD) from DKI and pure diffusion coefficient (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f) from IVIM were measured and compared from the volumetric analysis. Receiver operating characteristics (ROC) curve and logistic regression analyses were conducted to evaluate the diagnostic efficiency of different diffusion parameters for distinguishing invasive placentas. Results Comparisons between accreta lesions in patients with invasive placentas (AL) and lower 1/3 part of the placenta in patients with noninvasive placentas (LP) demonstrated that MD mean, D mean, and D* mean were significantly lower while ADC max and D max were significantly higher in invasive placentas (all p < 0.05). Multivariate analysis demonstrated that D mean, D max and D* mean differed significantly among all the studied parameters for invasive placentas. A combined use of these three parameters yielded an AUC of 0.86 with sensitivity, specificity, and accuracy of 84.91%, 76.60%, and 80%, respectively. Conclusion The combined use of different IVIM parameters is helpful in distinguishing invasive placentas.

To develop and validate clinical-radiomics models for predicting lymph node metastasis following neoadjuvant chemoradiation therapy in locally advanced rectal cancer .83 patients were retrospectively enrolled.pre-,post- and delta radiomics signatures of T2WI and ADC images were constructed by support vector machine model. These models were applied to predict LNM and 5-year disease-free survival. The clinical-deltaADC radiomics combined model presented good performance for predicting post-CRT LNM in the training cohort (AUC=0.895) and validation cohort (AUC=0.900). In ypT0-T2 stage, this model could predict 5-year RFS. Clinical-deltaADC radiomics combined model has good performance to predict LNM after nCRT and helped identify patients with poor prognosis.