Objective Based on MRI images, an automatic fat quantification model was developed to quantitatively assess the degree of fatty infiltration in the bilateral thigh muscles of children with Duchenne muscular dystrophy (DMD). Methods The fat ratio (FR) of 11 symmetrical muscles of the bilateral thighs was measured by the automatic fat quantification model on T₁WI image in 48 patients with DMD. The correlation between the model-FR values and the degree of fat infiltration by graded semi-quantitation, as well as the age of DMD patients, was analyzed respectively. According to the Kim scoring criterion, The Kappa value between the scores of the automatic fat quantification model and radiologists' fat scores was calculated and a consistency test was used. Results A strong positive correlation (r≥0.87, P<0.05) was observed between model-FR values and radiologists' scores. In DMD patients, the model-FR values of the bilateral thigh muscles were highest in the adductor magnus, followed by the vastus lateralis, while the adductor longus and gracilis exhibited the lowest FR values. Furthermore, FR values demonstrated a significant positive correlation with the age of DMD patients (r=0.52, P<0.05). The consistency between the scores of the automatic fat quantification model and radiologists' fat scores was relatively high (Kappa≥0.81,P<0.05) . Conclusion Compared to radiologists' fat scores, the model-FR values can be measured rapidly and accurately by the automated fat quantification model, and it can objectively assess the degree and the distribution pattern of fatty infiltration in thigh muscles of DMD patients. This model can provide objective evidence for clinical diagnosis, treatment, and subsequent monitoring of disease progression.

Objective To explore the X-ray characteristics of osteogenesis imperfecta (OI) in children, to improve the understanding and diagnostic accuracy of OI. Methods Clinical and X-ray data of 20 children with OI diagnosed clinically between January 1, 2000 and December 30, 2024 were collected from Yueqing Maternal and Child Health Hospital, Yueqing People's Hospital, and Wenzhou Medical University Affiliated Second Hospital in Zhejiang Province. The X-ray manifestations of osteogenesis imperfecta in children were retrospectively analyzed. Results Among the 20 cases, 16 cases had multiple fractures of long bones, totaling 45 fractures; Long bones were bent and deformed, with uneven thickness and thinning of 32 cortical bones; Femurs became shorter and thicker, with an "earpiece like" change in 8 cases and 13 bones; There were 22 asymmetrical transverse dense lines observed at the metaphysis of the lower femur and upper tibia. 20 cases had varying degrees of decreased bone density, with 8 cases and 13 roots showing "cystic expansion" changes in limb bones; There were 8 cases and 12 roots of bone localized 'loofah like' changes. Among the 20 cases, 14 ribs became thinner and elongated, including 8 ribs with multiple fractures and 6 ribs with high scapular and thoracic deformities. Among the 20 cases, 11 cases had scoliosis and vertebral osteoporosis, of which 35 vertebral bodies were flattened and wedge-shaped, and 12 vertebral bodies were biconcave. Among the 20 cases, 5 newborns had thin and deformed skulls, presenting as membranous skulls. Seven cases showed pelvic deformation with a triangular pelvic entrance and acetabular invagination. Among the 20 cases, 13 cases had 23 joints with enlarged bone ends and joint deformation, and 13 hip sockets and femoral heads were deformed and concave into the pelvis. Conclusion The clinical manifestations of OI include blue sclera, joint laxity, short stature, hearing loss, etc; X-rays are characterized by multiple fractures, decreased bone mass, “cystic expansion” and “loofah like” changes in bone mass, flattened vertebral bodies, biconcave or wedge-shaped changes, triangular pelvic entrance, and joint deformation.

Objective To explore the relationship between fat attenuation index (FAI), myocardial bridge (MB) parameters and coronary atherosclerosis in the proximal segment of left anterior descending artery (LAD) myocardial bridge (MB), and to evaluate their value in risk prediction. Methods A total of 243 patients with LAD-MB (confirmed by coronary computed tomography angiography, CCTA) were enrolled. The FAI and MB parameters in the proximal segment of MB were measured. The patients were divided into four groups: simple MB group (Group A, n=127), MB combined with proximal atherosclerosis group (Group B, n=116), Group B1 (coronary atherosclerotic stenosis >50%, n=30) and Group B2 (matched control for Group B1, n=30). Logistic regression analysis was used to identify risk factors, and receiver operating characteristic (ROC) curves were plotted. Results There were significant differences in age, hypertension, smoking history, MB length, myocardial bridge index (MMI), stenosis rate and FAI between Group A and Group B (all P<0.05). Age (OR=1.055), stenosis rate (OR=1.160) and FAI (OR=1.099) were independent risk factors, with their respective area under the ROC curve (AUC) values being 0.675, 0.846 and 0.666. The combined AUC of these three factors increased to 0.889. In Group B1 and Group B2, only FAI was an independent risk factor for coronary atherosclerotic stenosis >50% (OR=1.130, AUC=0.719). Conclusion The combination of age, MB stenosis rate and FAI can predict proximal atherosclerosis (AUC=0.889), and FAI is the only predictor for coronary atherosclerotic stenosis >50% (P<0.05).

Objective To evaluate the added value of clinical and radiomic features in differentiating benign and malignant non-mass enhancement (NME) lesions based on breast multi-parametric magnetic resonance imaging (MRI). Methods A retrospective analysis was performed on 147 NME patients who met the inclusion and exclusion criteria between September 2021 and September 2024. These patients were randomly divided into a training set and a validation set at a ratio of 7∶3. Radiomic features were extracted from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and apparent diffusion coefficient (ADC) images. The least absolute shrinkage and selection operator (LASSO) model was used for feature selection and to construct a radiomic model. For clinical features, a clinical model was established using multivariate Logistic regression analysis. The optimal-performing radiomic model was combined with clinical features to build a visual nomogram model. The performance of the nomogram model was evaluated using calibration curves and decision curve analysis (DCA). Results The results showed that the nomogram model exhibited excellent performance in distinguishing benign and malignant NME lesions. The area under the curve (AUC) was 0.928 in the training set and 0.882 in the validation set. This performance was superior to that of the single clinical model and the single radiomic model. Conclusion sThis study used clinical features combined with multi-parametric MRI radiomics to quantitatively analyze the information contained in images. It was confirmed that the nomogram model provides good benefits for clinical decision-making in differentiating benign and malignant NME lesions, which is conducive to disease treatment decisions.

Objective To establish an interpretable machine learning prediction model for breast cancer tumor-infiltrating lymphocyte (TIL) levels based on MRI habitat imaging. Methods MRI enhanced T₁ sequence phase 2 images of breast cancer patients who underwent enhanced MRI at Taizhou Second People's Hospital from January 2021 to December 2024 were retrospectively collected. After preprocessing the images, the region of interest (ROI) was manually segmented. Nineteen local radiomics features were extracted, and K-means clustering combined with the CH index was used to identify habitat subregions. Radiomics features were extracted from both subregions and the whole tumor region, and a multi-layer perceptron (MLP) was used to construct habitat and whole tumor prediction models. The diagnostic performance of the two models was summarized and receiver operating characteristic (ROC) curves were plotted. DeLong test was used to compare the AUC values between the two models, and decision curve analysis (DCA) was used to compare their clinical application value. The SHAP algorithm was used to interpret the habitat prediction model at both the global and individual levels. Results A total of 198 patients were included, with 119 cases of low TIL expression and 79 cases of high TIL expression. Each whole tumor ROI was divided into three habitat subregions. The AUC value of the habitat prediction model in the validation set was 0.825 (95% CI: 0.717-0.934), which was higher than that of the whole tumor prediction model (AUC=0.755, 95% CI: 0.630-0.880). DeLong test showed a statistically significant difference (P < 0.001). DCA analysis further confirmed that the habitat prediction model has higher clinical value. The SHAP algorithm can intuitively demonstrate the process of predicting TIL levels in breast cancer using the habitat prediction model. Conclusion An interpretable machine learning model based on MRI habitat imaging can comprehensively, noninvasively, and repeatably predict TIL levels in breast cancer and provide interpretable analysis to support clinical treatment decision-making and prognosis assessment.

Objective To investigate the value of diffusion kurtosis imaging(DKI) and intravoxel incoherent motion (IVIM)combined with clinical and conventional MRI features in the prediction of isocitrate dehydrogenase 1 (IDH1) and 1p/19q molecular types of glioma. Methods The data of sixty-one patients with glioma confirmed by surgical pathology who underwent conventional magnetic resonance,DKI and IVIM scans were collected. According to the pathological results,the glioma patients were divided into three subtypes:IDH1 mutation and no 1p/19q codeletion (IDH1mut-NonCodel),IDH1 mutation and 1p/19q codeletion (IDH1mut-Codel),and IDH1 wild type (IDH1wt).The differences in the clinical data,conventional MRI features,relative mean diffusion coefficient (rMD),relative mean kurtosis (rMK),relative radial kurtosis (rKr),relative axial kurtosis (rKa),relative fractional anisotropy (rFA),relative pseudo-diffusion coefficient (rD^*),relative true diffusion coefficient (rD) and relative perfusion fraction (rf) between IDH1mut and IDH1wt groups and also among the three subtypes were compared. Receiver operating characteristic (ROC)curves were drawn, and the areas under the curve(AUCs),sensitivities and specificities were calculated and compared. Results Lesion location,boundary,T₂-FLAIR mismatch sign(T₂FM),diffusion restriction,enhancement,rMD,rMK and rD^* differed significantly among the three subtypes(P＜0.05); pairwise comparisons revealed significant differences in location,boundary,enhancement,rMD,rMK and rD^* between IDH1mut-Codel and IDH1wt groups,and significant differences in boundary,T₂FM and rMK between IDH1mut-NonCodel and IDH1wt groups(P< 0.05).There were significant differences in location,boundary,rMD,rMK,rD^* and rD between IDH1mut-Codel and IDH1mut-NonCodel and IDH1wt groups,and significant differences in age,location,boundary,T₂FM,diffusion restriction,enhancement,rMD,rMK,rKr,rKa,rD^*and rD between IDH1mut and IDH1wt groups(P< 0.05).When DKI and IVIM were used for individual diagnoses of the IDH1 genotype and IDH1mut-Codel, the rMK had the highest diagnostic efficacy,with AUC values of 0.799 and 0.784,sensitivities of 63.2% and 100.0%,and specificities of 88.1% and 62.7%,respectively. When age,conventional MRI features,DKI and IVIM were combined to diagnose the IDH1 genotype,the AUC value was 0.962,and the sensitivity and specificity were 84.2% and 97.6%,respectively. When conventional MRI features,DKI and IVIM were combined to diagnose the IDH1mut-Codel,the AUC value was 0.914,and the sensitivity and specificity were 90.0% and 92.2%,respectively. Conclusion DKI and IVIM combined with clinical and conventional MRI features can help predict the molecular type of IDH1 and 1p/19q in glioma,and provide a basis for the diagnosis,individualized treatment and prognostic assessment of glioma patients.

Objective To investigate the pathologically impaired pattern of iron deposition among subcortical gray matter nuclei in patients with Parkinson's disease (PD) from the perspective of iron metabolic network. Methods Quantitative susceptibility mapping (QSM) data from 68 patients with PD and 70 healthy controls (HC) subjects were retrospectively analyzed. Iron content in 34 subcortical gray matter nuclei were obtained via data preprocessing. Iron metabolic networks were constructed at group level, following by the calculation of topological properties. Permutation test was utilized to analysis the intergroup differences in term of the topological properties (global and nodal) and connection strength from iron metabolic networks, following with multiple comparison correction (FDR correction). Results There was no significant difference in terms of age, sex, education and cognitive score between HC subjects and PD patients (P>0.05). Compared with HC group: In terms of global properties, the Gamma and Sigma values of PD group were significantly increased; In terms of node properties, abnormal nodal centralities in PD group were distributed in the globus pallidum (internal and external), substantia nigra (pars compacta, pars reticulata), habenula and thalamus (internal medullary lamina) (P<0.05); in terms of connection strength, the aberrant paired nodes in PD group included putamen and globus pallidum, substantia nigra and habenula, caudate and thalamus, substantia nigra and thalamus as well as subthalamic nucleus and thalamus (P<0.05). Conclusion This study confirmed that PD patients exhibited with topological damage and connection alteration of the subcortical iron metabolic network, which enriched the understanding of the pathogenesis of iron deposition and expanded the clinical application of QSM.

Objective To construct a transdiagnostic framework of brain-atrophy progression for schizophrenia（SCZ）and bipolar disorder（BD）using T₁-weighted（T₁WI）magnetic resonance imaging（MRI）based on the Subtype and Stage Inference（SuStaIn）algorithm. Methods 190 patients（101 SCZ,89 BD）and 158 healthy controls were included.Gray-matter volumes from 17 predefined regions of interest were extracted from T₁WI MRI,normalized,and put into the SuStaIn model to derive subtypes and stages.Between-subtype differences in clinical characteristics were assessed with analysis of variance,and Spearman correlation was used to examine associations between clinical measures and model-inferred stage. Results SuStaIn identified two distinct transdiagnostic atrophy trajectories.One was named a cerebellar-dominant subtype,in which atrophy initiates in the cerebellum and subsequently involves Broca’s area,the frontal cortex,and the insula.The other was a prefrontal-dominant subtype,in which atrophy originates in the frontal cortex and extends to the cingulate gyrus and temporal lobe.Patients with the prefrontal-dominant subtype had significantly higher positive-symptom scores than those with the cerebellar-dominant subtype（14.2 vs. 9.15;P=0.029）.Subgroup analysis showed that patients with prefrontal dominance exhibited more severe positive symptoms（BD：14.4 vs. 11.3,P=0.041;SCZ：8.6 vs. 7.2,P=0.038）.Advancing model-inferred stage was associated with reduced gray-matter volume in the cerebellum（r=-0.169,P=0.007）and prefrontal cortex（r=-0.193,P=0.002）across subtypes,and with greater symptom severity for positive（r=0.331,P=0.012）,negative（r=0.283,P=0.033）,and general symptoms（r=0.411,P=0.002）. Conclusion Using cross-sectional MRI data,this study identified two transdiagnostic brain-atrophy pathways shared by SCZ and BD,with clinical symptom severity increasing with model-inferred stage.These findings support the view that psychiatric disorders exhibit progressive spatial-temporal patterns of brain degeneration and suggest that the SuStaIn model may be a useful imaging-based tool for stratifying psychiatric disorders.

Objective To explore the predictive value of spectral CT in forecasting recurrence and distant metastasis in locally advanced nasopharyngeal carcinoma (LANPC). Methods A retrospective analysis was performed on 57 LANPC patients who underwent spectral CT scans prior to treatment, including 13 patients in the recurrence and metastasis group and 44 in the non-recurrence and metastasis group. The follow-up endpoint was progression-free survival (PFS). Spectral CT parameters of the primary nasopharyngeal lesions were obtained using a post-processing workstation, including iodine concentration (IC), normalized iodine concentration (NIC), spectral curve slope (λHU), and effective atomic number (Zeff). Univariate analysis was performed to assess differences in clinical indicators and spectral CT parameters between the two groups. Statistically significant variables were included in logistic regression analysis to identify independent risk factors for recurrence and metastasis. The predictive performance of independent risk factors was evaluated using receiver operating characteristic (ROC) curves. Survival analysis and differences in survival were assessed using the Kaplan-Meier method and log-rank test. Results Univariate analysis indicated that the recurrence and metastasis risk was increased in the induction chemotherapy (ICT) non-responder group compared to the ICT responder group (P=0.012). The IC, NIC, and Zeff values in the recurrence and metastasis group were markedly reduced than those in the non-recurrence and metastasis group (P<0.05 for all). Logistic regression analysis revealed that NIC was an independent risk factor for recurrence and metastasis in LANPC, with a smaller NIC associated with a higher risk of recurrence and metastasis (P=0.044). The area under the curve (AUC) for NIC was 0.890. Kaplan-Meier survival curves demonstrated that patients with NIC > 0.30 had a longer PFS compared to those with NIC ≤ 0.30 (P=0.001). Conclusion The efficacy of ICT and spectral CT parameters are associated with recurrence and metastasis in LANPC. NIC values may serve as an independent predictive factor for recurrence and metastasis in patients.

Objective To explore the predictive value of a radiomics model based on dual-phase enhanced computed tomography (CT) for progression-free survival (PFS) in patients with small cell lung cancer (SCLC). Methods Clinical data of 148 SCLC patients confirmed by histopathology were collected retrospectively, including 88 cases from Center 1 (served as the training set) and 60 cases from Center 2 (served as the validation set). Three-dimensional volumes of interest (VOIs) were automatically delineated on arterial-phase and venous-phase enhanced CT images respectively. Radiomics features significantly correlated with PFS were extracted and screened, and the radiomics score (Rad-score) was calculated. Cox regression analysis was used to identify independent clinical risk factors affecting PFS. Subsequently, a clinical model, a radiomics model, and a combined model were constructed based on the independent clinical risk factors and Rad-score, and the predictive efficacy of these models was evaluated. Results A total of 5 radiomics features were finally selected. Clinical stage was an independent risk factor for PFS in SCLC patients (P< 0.001, HR=5.058, 95%CI: 2.139-11.960). According to Rad-score, SCLC patients were divided into the high-risk group (Rad-score ≥ 0.17) and the low-risk group (Rad-score<0.17), with a statistically significant difference in survival between the two groups (validation set:P< 0.001,HR=3.002, 95%CI:1.580-5.706). The predictive efficacy of the radiomics model (C-index: 0.826) and the combined model (C-index: 0.828) for PFS in SCLC patients was significantly higher than that of the clinical model (C-index: 0.582). Compared with the clinical model, the radiomics model had a net reclassification index (NRI) of 0.647 (95%CI: 0.419-0.842,P<0.05) and an integrated discrimination improvement index (IDI) of 0.324 (95%CI: 0.165-0.488,P<0.05). There were no significant differences in NRI and IDI between the radiomics model and the combined model (P>0.05). In terms of clinical utility, both the radiomics model and the combined model were superior to the clinical model. Conclusion The radiomics model based on dual-phase enhanced CT shows excellent performance in predicting PFS of SCLC patients, which can provide valuable information for individualized treatment.

Objective This study utilized cardiac magnetic resonance(CMR)to quantify epicardial adipose tissue(EAT)in patients with ST-segment elevation myocardial infarction(STEMI)following primary percutaneous coronary intervention(PCI)and to explore its correlation with acute-phase myocardial injury parameters and convalescent left ventricular systolic function. Methods This retrospective study analyzed 131 STEMI patients who underwent emergency PCI.CMR examinations were performed at two time points:acute phase(1-week post-PCI)and convalescent phase(3-4 months post-PCI).Quantitative measurements included EAT volume,infarct size(IS),microvascular obstruction(MVO)and left ventricular ejection fraction(LVEF).Patients were stratified into high-EAT(EAT>30 ml/m²,n=82)and low-EAT(EAT≤30 ml/m²,n=49)groups based on the third quartile of EAT volume index. Results The mean EAT volume at the 1-week CMR was (27±9)ml/m².The high EAT group exhibited higher IS and MVO during the acute phase,higher IS during the recovery phase,and a greater reduction in MVO between the two exams compared to the low EAT group.Univariate Pearson correlation analysis revealed a positive correlation between EAT and acute phase IS(r=0.383,P<0.001)and MVO(r=0.273,P=0.002).Multivariate linear regression analysis showed that,after adjusting for relevant confounding factors,EAT remained an independent correlate of acute phase IS(β=0.004,95%CI:0.002-0.005,P<0.001)and MVO(β=0.093,95%CI:0.026-0.161,P=0.008),respectively.Furthermore,among patients with large IS(>30% of LV mass)and MVO at the 1-week CMR(n=43),univariate Pearson correlation analysis demonstrated a significant positive correlation between EAT and recovery phase LVEF(r=0.415,P=0.006).After adjusting for relevant confounding factors,EAT remained an independent correlate of recovery phase LVEF(β=0.246,95%CI:0.027-0.466,P=0.034). Conclusion In patients with STEMI,pre-existing EAT exhibits a dual role:exacerbating acute ischemic injury yet may potentially facilitate functional recovery in severely damaged myocardium.

Objective To analyze the correlation between paravertebral muscle parameters and clinical features and lumbar function in patients with degenerative spinal deformity (DSD) by high-field magnetic resonance water-lipid separation sequence (MR Dixon). Methods A total of 85 DSD patients admitted to our hospital from March 2022 to June 2024 were included in the DSD group. In the same period, 85 healthy subjects were selected and included in the control group. All subjects underwent MR Dixon paravertebral myography and lumbar function assessment. The clinical data of all subjects were collected. Pearson correlation analysis and multiple linear regression were used to analyze the relationship between MR Dixon paravertebral muscle parameters, clinical features and lumbar function. Results There was no significant difference between the DSD group and the control group in terms of gender, age, body mass index (BMI), diabetes mellitus and hypertension (P>0.05). Compared with the control group, the CSA of the multifidus muscle, the erector spine muscle and the psoas major muscle were decreased in DSD group, while the FI of the multifidus muscle and the erector spine muscle was increased (P<0.05). There was no significant difference in psoas FI between the two groups (P>0.05). ODI score in DSD group was higher than that in control group (P<0.05). Pearson correlation analysis showed that in DSD patients, the CSA of psoas major and erector spine were weakly correlated with age, the CSA of multifidus was moderately correlated with age (P<0.05), and the CSA of multifidus was moderately correlated with ODI (P<0.05). Multifidus FI and erector spinal FI were strongly correlated with age, BMI and ODI (P<0.05). There was no correlation between CSA of psoas, erector spinalis and multifidus muscle and BMI (all P>0.05), while there was no correlation between CSA of psoas, erector spinalis and ODI (all P >0.05). Multiple linear regression analysis showed that age, diabetes and ODI were the independent influencing factors of multifidus CSA. Sex, age, diabetes mellitus and ODI were the independent influencing factors of multifidus FI. Gender, age and diabetes were the independent influencing factors of psoas major muscle CSA. Age and diabetes were the independent factors of CSA in erector spine muscle. Sex, age, BMI and diabetes mellitus were the independent influencing factors of erector spinal muscle FI (P<0.05). Conclusion DSD patients with MR Dixon paravertebral muscle parameters abnormal, lumbar function impaired, paravertebral muscle parameters are closely related to age, diabetes and other clinical features and lumbar function.

Objective To explore the value of a three-phase CT-based radiomics model in distinguishing lipid-poor adrenal adenomas from adrenal metastases. Methods A retrospective collection of 39 pathologically confirmed cases of lipid-poor adrenal adenomas(including 4 bilateral cases)and 68 cases of adrenal metastases(including 8 bilateral cases)from June 2017 to June 2024 was conducted.All patients underwent plain,arterial,and venous phase CT scans.The full tumor region of interest(ROI)was manually delineated using 3D Slicer software.A total of 1316 radiomic features were extracted using PyRadiomics,and features with high repeatability(ICC>0.75)and non-redundancy were selected.Feature selection was optimized using LASSO regression combined with 10-fold cross-validation.Single-phase,dual-phase,and three-phase combined models were constructed using five algorithms:Logistic Regression(LR),Support Vector Machine(SVM),Random Forest(RF),K-Nearest Neighbor(KNN),and XG-Boost.The performance of the models was evaluated using ROC curves,and differences between models were compared using the Delong test. Results The three-phase combined RF model(Model 7)performed best,with an AUC of 0.9498 and an accuracy of 0.8333 in the test set.Among the single-phase models,the arterial phase combined with the RF model had an AUC of 0.913.In the dual-phase models, the plain+venous phase combined with the SVM model achieved an AUC of 0.9431.The Delong test showed no statistical difference between the models(P>0.05).The plain phase combined with the SVM model had an AUC of 0.9097,suggesting its potential value for patients who cannot undergo enhanced scanning. Conclusion The three-phase CT-based radiomics can effectively distinguish lipid-poor adrenal adenomas from metastases,with the three-phase combined RF model demonstrating optimal diagnostic performance.The plain phase combined SVM model provides a feasible alternative for special populations and has important clinical significance.

Objective To investigate the predictive value of preoperative adipose tissue ¹⁸F-FDG Uptake for the prognosis of patients after radical gastrectomy. Methods A retrospective analysis was conducted on 101 gastric cancer patients who underwent radical gastrectomy between November 2019 and September 2023.The mean standardized uptake values(SUV_mean)of visceral adipose tissue(VAT)and subcutaneous adipose tissue(SAT)at the L₄/L₅ level were obtained from PET/CT images.Patients were divided into high- and low-level groups based on the median values of VAT SUV_mean and SAT SUV_mean,respectively.Clinical characteristic differences between groups were analyzed through intergroup comparisons.Spearman correlation coefficients were used for correlation analysis.Cox proportional hazards regression models were employed to assess risk factors for overall survival(OS)and recurrence-free survival(RFS).Survival analysis was performed using the Kaplan-Meier method,and differences in survival between groups were evaluated with the Log-rank test. Results The median VAT SUV_mean and SAT SUV_mean were 0.40 and 0.32,respectively.Based on these cutoff values,patients were divided into a high VAT SUV_mean group(≥0.40,n=51)and a low VAT SUV_mean group(<0.40,n=50),as well as a high SAT SUV_mean group(≥0.32,n=51)and a low SAT SUV_mean group(<0.32,n=50).The high VAT SUV_mean group showed statistically significant differences compared to the low VAT SUV_mean group in terms of body mass index(BMI),preoperative nutritional status,albumin(ALB),and prealbumin(Pre-ALB)levels(P<0.05).The high SAT SUV_mean group demonstrated a statistically significant difference in Pre-ALB levels compared to the low SAT SUV_mean group(P<0.05).Spearman correlation analysis revealed significant negative correlations between VAT SUV_mean and BMI,ALB,and Pre-ALB levels(P<0.05),while significant positive correlations were observed between VAT SUV_mean and the neutrophil-to-lymphocyte ratio(NLR)and platelet-to-lymphocyte ratio(PLR)in peripheral blood(P<0.05).Multivariate Cox analysis identified high VAT SUV_mean and advanced tumor stage as independent risk factors for both recurrence-free survival(RFS)and overall survival(OS)in gastric cancer patients(P<0.05).Kaplan-Meier survival analysis demonstrated that the high VAT SUV_mean group had significantly lower OS(Log-rank χ²=14.608,P<0.001)and RFS(Log-rank χ²=15.254,P<0.001)compared to the low VAT SUV_mean group.Similarly,the high SAT SUV_mean group exhibited significantly lower OS(Log-rank χ²=5.599,P=0.018)and RFS(Log-rank χ²=7.038,P=0.008)compared to the low SAT SUV_mean group. Conclusion A high VAT SUV_mean is identified as an independent risk factor affecting both recurrence-free survival(RFS)and overall survival(OS)in gastric cancer patients.Although SAT SUV_mean showes association with survival outcomes,it demonstrates weaker clinical significance.

Objectives To validate the diagnostic performance of the minimum apparent diffusion coefficient (ADCmin) cutoff value (1.39×10^-3mm²/s) in reducing unnecessary biopsies at breast MRI. Methods The MRI data of consecutive breast magnetic resonance imaging (MRI) examinations were retrospectively analyzed, which had been classified as Breast Imaging Report and Data System 4 or 5 lesions. MRI examination included axial T₂-weighted, diffusion-weighted and dynamic contrast-enhanced sequence, and the ADCmin of the lesions was measured. Taking biopsy or surgical histopathology as the reference standard, the negative likelihood ratio (NLR), overall lesion biopsy reduction rate, benign lesion biopsy reduction rate and sensitivity were calculated after applying the the ADCmin cutoff (1.39×10^-3mm²/s) reported in the previous study, and subgroup analysis was performed. Results The final analysis included 183 lesions of 165 patients, with an average age of 47.9±11.2 years, including 107 malignant lesions and 76 benign lesions. After the application of ADCmin cutoff of 1.39×10^-3mm²/s, the NLR of all subgroups was less than 0.1 except for non-mass enhancement (NME) and the lesion group with maximum diameter≤1cm. The overall lesion biopsy reduction rate was 26.2% (95%CI:21.2-30.9), the biopsy reduction rate of benign lesions was 57.9% (95%CI:44.6-69.1), and the sensitivity was 96.3% (95%CI:90.7-99.0). Subgroup analysis showed that the sensitivity of mass was higher than that of NME (P=0.02). The overall biopsy reduction rate and benign lesion biopsy reduction rate were similar in all groups (P>0.05). Conclusion This study verified the ability of ADCmin cutoff to reduce unnecessary biopsies at breast MRI, and this ADC cutoff is helpful for the downgrade of BI-RADS 4 and 5 lesions.

Objective To develop a model to differentiate pulmonary cryptococcosis (PPC) from pulmonary mucinous adenocarcinoma (PMA) using multidimensional parameters, including traditional imaging features, clinical characteristics, and artificial intelligence-quantified parameters. Methods This study retrospectively collected cases diagnosed with PMA or PPC at our hospital from December 2019 to June 2023. Preoperative clinical data, traditional imaging parameters, and artificial intelligence quantitative parameters of pulmonary nodules were gathered. A total of 102 cases of PMA and 117 cases of PPC were included. Effective parameters were initially screened using univariate Logistic regression and then further refined using a multivariate binary Logistic regression model. The accuracy of the model was analyzed using an ROC curve. Results Univariate analysis revealed significant differences in age, personal history of tumor, diabetes mellitus, hypertension, hyperlipidemia, CEA, cytokeratin, wide base, cavitation, bronchial passage, halo sign, long strand, satellite lesions, maximum length diameter, and total volume between the two groups (P<0.05). The final multivariate binary Logistic regression model showed that cavitation sign, halo sign, long strand, satellite lesion, history of tumor, and age were statistically significant, with the area under the curve (AUC) of the model being 0.933, with a 95% confidence interval (CI) of 0.905-0.970. Conclusion Integrating the clinical and imaging features of nodules is beneficial for preoperative identification of PPC or PMA, facilitating the development of personalized treatment strategies by clinicians and minimizing unnecessary invasive procedures.

Objective To explore the MRI features distinguishing Internal adenomyosis(IAM)from non-adenomyosis-like junction zone(JZ)thickening(hereafter referred to as JZ thickening)in cases where the maximum JZ diameter(JZ_max)exceeded 12 mm without significant differences. Methods Data from 47 IAM cases and 45 JZ thickening cases confirmed by postoperative pathology were collected.Clinical and imaging features were compared between groups.The Phi coefficient（φ)was calculated for indicators showing significant intergroup differences.Binary Logistic regression was used to construct a combined model,and receiver operating characteristic curves were used to evaluate the value of individual indicators and the combined model. Results Microcystic foci within the JZ(φ=0.644)and a ratio of maximum JZ diameter to uterine anteroposterior diameter(Ratio_max)≥1/3(φ=0.348)suggested IAM,while uniform JZ signal(φ=-0.702)indicated JZ thickening(all P<0.05).The area under the curve(AUC)for diagnosing IAM using microcystic foci and Ratiomax≥1/3 alone and their combination was 0.817,0.663,and 0.883,respectively.The AUC for uniform JZ signal in diagnosing JZ thickening was 0.849(all P<0.05). Conclusion The MRI features of IAM and JZ thickening overlap, but they differ in JZ microcystic foci,Ratio_max≥1/3,and JZ signal uniformity,Notably,the combination of the former two achieved the highest AUC.

Objective To investigate the accuracy and practicality of the Efficient-Det model based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for early prediction of neoadjuvant chemotherapy (NAC) efficacy in patients with triple-negative breast cancer (TNBC). Methods A retrospective analysis was conducted using data from TNBC patients who received NAC treatment. A total of 97 cases were used for the training set, and 52 cases for the testing set. Logistic regression analysis was used to construct a clinical prediction model based on clinical and pathological data, and a deep learning prediction model was built using the Efficient-Det model. The DeLong test was employed to compare the AUC performance of the two models and two radiologists. Gradient-weighted class activation mapping (Grad-CAM) heatmaps were used to visualize the key regions in the lesions for predicting NAC efficacy. Results Logistic regression analysis showed that the area under the curve (AUC) of the Efficient-Det model for the training and testing sets was 0.901 and 0.857, respectively, which were significantly higher than the performance of the clinical prediction model constructed by combining clinical staging and Ki-67 expression (training set: Z=2.743,P=0.005;test set: Z=3.041,P<0.001）. With the assistance of the Efficient-Det model, the efficacy assessments of DCE-MRI images at two NAC treatment cycles showed good consistency between the two radiologists (ICC=0.846, ICC=0.809), and both radiologists' diagnostic performance significantly improved (all P<0.05). Conclusion Compared with the clinical prediction model, the Efficient-Det model based on DCE-MRI provides more accurate prediction of NAC efficacy, significantly improve the diagnostic performance of radiologists with varying years of experience, and exhibit strong generalizability.

Objective To investigate alterations in cerebral blood flow (CBF) and cerebral blood flow network under different Alzheimer's disease (AD)-related pathological burdens and cognitive states, and to deepen the understanding of neurobiological mechanisms in the AD disease spectrum. Methods Data were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database, including 29 Aβ-negative cognitively normal subjects (Aβ-CN), 34 Aβ-positive cognitively normal subjects (Aβ+ CN), and 39 Aβ-positive cognitively impaired subjects (Aβ+ CI). Demographic data, neuropsychological test results, and imaging data of the subjects were collected. CBF information was obtained from arterial spin labeling (ASL) imaging. Voxel-based morphometry was used to compare and identify brain regions with CBF differences among the three groups, and partial correlation analysis was conducted between CBF in these regions and scores of neuropsychological assessment scales. A cerebral blood flow network based on CBF values was constructed, and graph theory analysis was applied to explore the changing characteristics of global network properties and nodal properties. Results Significant differences in CBF among the three groups were observed in brain regions including the frontal lobe, temporal lobe, and parietal lobe (GRF correction, voxel-level P < 0.001, cluster-level P< 0.05). Compared with the Aβ-CN group, the Aβ+ CN group showed significantly decreased CBF in the bilateral anterior cingulate and paracingulate gyri, left inferior parietal angular gyrus, left middle temporal gyrus, and left inferior temporal gyrus. Compared with the Aβ+ CN group, the Aβ+ CI group had significantly decreased CBF in the right orbital part of the middle frontal gyrus and left middle temporal gyrus; compared with the Aβ-CN group, the Aβ+ CI group showed significantly decreased CBF in all the above-mentioned brain regions with differences (all P<0.05). Correlation analysis indicated that CBF in the right orbital part of the middle frontal gyrus was negatively correlated with the score of Alzheimer's Disease Assessment Scale-Cognitive Subscale (ADAS-13) (r = -0.338, P= 0.011, FDR correction); CBF in the left inferior temporal gyrus was positively correlated with the score of Montreal Cognitive Assessment (MoCA) (r = 0.329, P = 0.011, FDR correction); CBF in the right anterior cingulate and paracingulate gyri was positively correlated with the MoCA score (r = 0.280, P= 0.044, FDR correction). Graph theory analysis results showed that compared with the Aβ+ CN group, the Aβ+ CI group had a significantly decreased global efficiency (P= 0.039); compared with the Aβ- CN group, the Aβ+ CI group had significantly increased small-worldness(P= 0.034) and local efficiency (P= 0.032), global efficiency was significantly decreased(P=0.005). Conclusion With the progression of Aβ pathology and the decline of cognitive function, CBF and cerebral blood flow network patterns in regions such as the frontal lobe, temporal lobe, and parietal lobe undergo significant changes. These findings provide a certain theoretical basis for exploring the pathological and neurobiological mechanisms of the AD disease spectrum from the perspective of cerebral blood flow.

Objective To explore changes in the resting-state brain functional network of patients with diabetic retinopathy (DR) using degree centrality (DC) and functional connectivity (FC) methods. Methods A total of 26 DR patients, 30 diabetic patients without retinopathy (NDR group), and 30 healthy controls (HC group) were enrolled. The differences in DC values across various brain regions were compared among the three groups. Brain regions with intergroup DC differences were used as seed points to analyze whether there were abnormalities in whole-brain FC values. Partial correlation analysis was applied to evaluate the correlations between DC values, FC values, scores of neuropsychological assessment scales, and clinical biochemical indicators. Results There were statistically significant differences in DC values among the three groups in the right postcentral gyrus, right precentral gyrus, left lingual gyrus, and right superior temporal gyrus (P<0.05). Compared with the HC group, the DR group had significantly decreased DC values in the right postcentral gyrus, right precentral gyrus, and left lingual gyrus; the NDR group had significantly decreased DC value in the left lingual gyrus and significantly increased DC value in the right superior temporal gyrus. Compared with the NDR group, the DR group had significantly decreased DC values in the right postcentral gyrus, right precentral gyrus, left lingual gyrus, and right superior temporal gyrus. Functional connectivity analysis among the three groups showed that when the left lingual gyrus was used as the seed point, there were statistically significant differences in FC values between the left lingual gyrus and the right lingual gyrus, as well as between the left lingual gyrus and the right cuneus. Partial correlation analysis revealed that in the DR group, the DC value of the right precentral gyrus was significantly positively correlated with the high-density lipoprotein cholesterol (HDL-C) level (r=0.45, P=0.037). Conclusion DR patients have abnormal brain functional activity in multiple brain regions, including the right postcentral gyrus, right precentral gyrus, left lingual gyrus, and right superior temporal gyrus. These abnormal regions may be related to functions such as vision, somatosensation, and cognition.

Objective The risk prediction model was established based on LASSO regression and multiple factors Logistic regression, and the application value of multimodal MRI in the diagnosis of glioblastoma (GBM) was discussed. Methods A total of 89 patients with pathologically diagnosed glioma were included, including 46 patients with GBM and 43 patients with non-glioblastoma (N-GBM). They were divided into a training set (n=62) and a validation set (n=27), and routine MRI sequence, SWI, DWI, and MRS were examined. The differences in relevant imaging parameters between the two groups were compared, and the correlation coefficient heat map analysis was included. The most valuable parameters were selected by the LASSO regression model. The prediction model was constructed by parallel multi-factor logistic regression analysis, and the model efficacy was evaluated by receiver operating characteristic (ROC) curve, calibration curve, and decision curve. Results Both the training set and the validation set showed that GBM tumor cystic necrosis, edema degree, enhancement degree score, intra tumoral susceptibility signal (ITSS), choline (Cho)/creatine (Cr), Cho/N-acetylaspartate (NAA) were higher than those of N-GBM, and pairings between the two groups showed statistically significant differences (P<0.05). The rADC value of the relative apparent diffusion coefficient of the tumor area was significantly lower than that of N-GBM, and the difference between the two groups was statistically significant (P<0.001). LASSO regression was applied to screen out 4 most valuable parameters, including tumor cystodegeneration and necrosis, enhancement score, rADC, and Cho/NAA. Multivariate Logistic regression analysis showed that tumor enhancement score, rADC, and Cho/NAA were independent predictors of GBM. A nomogram model was established based on the above parameters. The calibration curve fits well with the ideal curve. The area under the ROC curve is 0.978, the sensitivity is 97.1%, and the specificity is 91.9%. Conclusion This model has high predictive value for GBM and has great clinical benefit.

Objective To investigate the application value of computed tomography (CT)-derived extracellular volume fraction (ECV) in preoperative prediction of benign and malignant gastrointestinal stromal tumors (GISTs). Methods A total of 110 GIST patients confirmed by surgery and pathology from January 2020 to January 2025 were included retrospectively, including 8 cases in the very low-risk group, 30 cases in the low-risk group, 20 cases in the intermediate-risk group, and 52 cases in the high-risk group. Among them, the very low-risk and low-risk groups were classified as the benign group (38 cases in total), and the intermediate-risk and high-risk groups were classified as the malignant group (72 cases in total). All patients underwent multi-phase dynamic enhanced CT scan before surgery. The tumor ECV (%) was measured through plain scan and equilibrium phase images, and clinical and imaging data were collected. Univariate and multivariate Logistic regression analyses were used to identify independent risk factors and construct individual and combined diagnostic models. Receiver operating characteristic (ROC) curves were used to evaluate diagnostic efficacy, and DeLong test was applied to compare differences in area under the curve (AUC). Results Multivariate Logistic regression analysis showed that ECV (%) and maximum tumor diameter were independent predictors of GIST malignancy (both P<0.05). The AUCs of ECV (%), maximum tumor diameter, and the combined model were 0.882 (95%CI: 0.819-0.945), 0.845 (95%CI: 0.765-0.925), and 0.916 (95%CI: 0.862-0.970), respectively. The AUC of the combined model was higher than that of ECV (%) alone. Conclusion The ECV parameter of lesions can effectively quantify the degree of extracellular matrix remodeling in GISTs, significantly improve the preoperative predictive value for benign and malignant gastrointestinal stromal tumors, and provide a non-invasive evaluation index for formulating individualized treatment plans.

Objective To evaluate the efficacy of quantitative parameters [iodine concentration (IC) and normalized iodine concentration (nIC)] of spectral CT in identifying lymph node metastasis (LM) in gastric cancer using evidence-based Meta-analysis. Methods A comprehensive search was conducted in six major databases [PubMed, EMBASE, Cochrane Library, Web of Science, China National Knowledge Infrastructure (CNKI), and Wanfang Database] for literature on the diagnosis of gastric cancer LM by spectral CT, published from January 2010 to January 2025. Literature meeting the inclusion criteria was subjected to quality assessment, characteristic data extraction, and heterogeneity evaluation. According to the research objects, the literature was divided into two categories: prediction of LM using spectral CT parameters of lymph nodes, and prediction of LM using spectral CT parameters of gastric tumors. For the most studied parameters in lymph node research [arterial-phase normalized iodine concentration (nICa)] and the most studied parameters in gastric tumor research [venous-phase iodine concentration (ICp) and venous-phase normalized iodine concentration (nICp)], effect sizes were pooled to obtain combined sensitivity, specificity, positive likelihood ratio, and diagnostic odds ratio. Forest plots and summary receiver operating characteristic (SROC) curves were drawn, and the area under the curve (AUC) was calculated. Begg's funnel plot asymmetry test was used to detect publication bias. Results A total of 10 studies meeting the inclusion criteria were included in the Meta-analysis, and they were classified by data type into 3 studies with lymph node data and 7 studies with gastric tumor data. There was publication bias for lymph node nICa, while no publication bias was found for tumor ICp and nICp. The combined AUCs of the three parameters were 0.85 (95%CI: 0.81-0.87), 0.79 (95%CI: 0.76-0.83), and 0.83 (95%CI: 0.80-0.86), respectively. Conclusion Spectral CT is applied in the evaluation of gastric cancer LM due to its advantages of non-invasiveness, convenience, cost-effectiveness, and efficiency. Among its parameters, those related to iodine concentration have certain application value, providing a new approach for preoperative evaluation of gastric cancer LM.

Objective To explore the feasibility of constructing a prediction model for mild renal insufficiency based on non-contrast renal computed tomography (CT) radiomics. Methods A retrospective analysis was performed on 693 hospitalized patients who had abdominal non-contrast CT images within 1 week before or after renal function testing. According to the estimated glomerular filtration rate (eGFR), patients were divided into two groups: (1) Normal Renal Function Group: 90 ≤ eGFR < 120 ml/(min·1.73m²); (2) Mild Renal Insufficiency Group: 60 ≤ eGFR < 90 ml/(min·1.73m²). A 3D U-Net deep learning technique was used to train an automatic kidney segmentation model for delineating the region of interest (ROI). LASSO regression analysis was applied to screen features and parameters associated with mild renal insufficiency, and radiomics prediction models were established based on Logistic regression, support vector machine, and decision tree. The efficacy of the models was evaluated using receiver operating characteristic (ROC) curves and area under the curve (AUC). The model with the best efficacy was presented as a radiomics nomogram, and the performance of the nomogram in predicting mild renal insufficiency was assessed using calibration curves. Results Among the radiomics prediction models, the one established by logistic regression showed the best efficacy, with AUC values of 0.849 in the training set and 0.782 in the test set. The nomogram of this prediction model exhibited good performance in predicting mild renal insufficiency, as demonstrated by waterfall plots and calibration curves. Conclusion It is feasible to construct a radiomics prediction model for mild renal insufficiency based on non-contrast CT images, which can remind clinicians to timely assess potential renal function abnormalities in patients.

Objective To explore the diagnostic efficacy of the combined model integrating CT texture parameters based on later arterial phase(LAP)and clinical information in the preoperative prediction of microsatellite instability-high(MSI-H)status in gastric cancers. Methods A total of 150 patients with gastric cancer[MSI-H,n=30;microsatellite instability-low/microsatellite stable（MSI-L/MSS）,n=120]were included in this study.Clinical information,CT texture parameters based on LAP,and postoperative pathological characteristics of all patients were retrospectively analyzed.All patients were randomly divided into training(n=105)and validation(n=45)groups at a 7∶3 ratio.The CT signature based on LAP was built with the optimum features selected by the least absolute shrinkage and selection operator method.Logistic regression analysis and support vector machine(SVM)algorithm were utilized to establish the combined model for predicting MSI-H status in gastric cancers by integrating clinical information and CT signature based on LAP.The receiver operating characteristic(ROC)curve was used to evaluate the diagnostic performance of the combined model,and the area under the ROC curve(AUC),diagnostic accuracy,sensitivity,and specificity were calculated. Results There were significant differences in gender,systemic immune inflammatory index(SII),and tumor location between MSI-H and MSI-L/MSS gastric cancers (P<0.05).The CT signatures based on LAP achieved AUCs of 0.854 and 0.846 in the training and validation groups,respectively.The combined models based on logistic regression analysis showed favorable diagnostic efficacies in the training and validation groups(AUC=0.893 and 0.906).The combined models based on SVM algorithm further improved the diagnostic performance with AUCs of 0.905 and 0.944 in the training and validation groups,respectively.Moreover,MSI-H gastric cancers showed higher proportion of tubular adenocarcinoma and lower proportions of lymph node metastasis and nerve invasion than MSI-L/MSS gastric cancers. Conclusion Combined models that integrates CT signature based on LAP,tumor location,SII,and gender utilizing Logistic regression analysis and SVM algorithm can accurately predict MSI status in gastric cancers preoperatively with favorable efficacies.

Objective To explore the value of ¹⁸F-FDG PET/CT imaging parameters combined with serum miR-345 and miR-20a in predicting the efficacy and prognosis of immune therapy in advanced non-small cell lung cancer (NSCLC). Methods A total of 158 advanced NSCLC patients who received immune therapy at our hospital from January 2021 to June 2024 were included in the study. All patients underwent ¹⁸F-FDG-PET/CT imaging and serum miR-345 and miR-20a level detection before receiving immune therapy. The imaging parameters of ¹⁸F-FDG-PET/CT included maximum standard uptake value (SUV_max),tumor metabolic volume (MTV),and total lesion glycolysis (TLG). Clinical and pathological data were collected,and treatment efficacy was evaluated using the RECIST 1.1 criteria. All patients were followed up for 10-24 months through telephone,outpatient,and inpatient visits to collect tumor assessment results,disease progression,and overall survival. The efficacy of the immune therapy was assessed using Kaplan-Meier survival curves for progression-free survival (PFS) and overall survival (OS). Spearman correlation analysis was conducted to assess the correlation between clinical efficacy and ¹⁸F-FDG-PET/CT imaging parameters and serum miR-345 and miR-20a levels. The effectiveness of using ¹⁸F-FDG-PET/CT imaging parameters and serum miR-345 and miR-20a levels to predict clinical efficacy was analyzed using receiver operating characteristic (ROC) curves. Univariate and multivariate Cox analysis were used to identify risk factors influencing patient prognosis. Results The average values for SUV_max,MTV,and TLG were (12.52 ± 5.26),(97.36 ± 13.63) cm³,and (1669.74 ± 705.46),respectively. The relative expression levels of miR-345 and miR-20a were (24.54 ± 2.11) and (25.38 ± 2.33),respectively. The objective response rate (ORR) of immune therapy in 158 patients was 44.30% (70/158),and the disease control rate (DCR) was 89.87% (142/158). Compared to the non-responding group,the SUV_max,MTV,and TLG values,as well as miR-345 levels,decreased in the responding group,while serum miR-20a levels increased,with statistically significant differences (P < 0.05). Spearman correlation analysis showed that immune therapy efficacy was negatively correlated with SUV_max,MTV,TLG,and miR-345 levels,while it was positively correlated with miR-20a levels. ROC analysis indicated that SUV_max,MTV,TLG,miR-345 2^-∆∆Ct,and miR-20a 2^-∆∆Ct had good predictive ability for NSCLC immune therapy efficacy,with combined treatment showing higher sensitivity and specificity. During the follow-up period,80 patients died,with a mortality rate of 50.63% (80/158). Kaplan-Meier survival curve analysis showed that the median PFS and mOS of all patients were 6.75 months (95% CI 6.20-8.15) and 20.81 months (95% CI 19.34-20.06),respectively. Cox multivariate analysis revealed that ECOG performance status,degree of differentiation,tumor size,distant metastasis,clinical stage,SUV_max,MTV,TLG,miR-345,and miR-20a were independent risk factors affecting the prognosis of NSCLC patients undergoing immune therapy. Conclusion ¹⁸F-FDG-PET/CT imaging parameters (SUVmax,MTV,TLG) combined with serum miR-345 and miR-20a levels can effectively predict the efficacy and prognosis of immune therapy in advanced non-small cell lung cancer patients.

Objective To quantitatively compare the diagnostic value of dynamic contrast-enhanced MRI kinetic heterogeneity and conventional diffusion-weighted imaging for the extent of breast cancer infiltration. Methods A retrospective analysis was performed of DCE-MRI data from patients with invasive breast cancer diagnosed by pathology, to obtain quantitative parameters of kinetic heterogeneity and apparent diffusion coefficients obtained by diffusion-weighted imaging, and to extract the main six parameters for lesion heterogeneity analysis from preoperative MRI data using MATLAB and SPM12 software, including Peak Volume, Persistent, Plateau, Washout, and Heterogeneity.The diagnostic efficacy of DCE-MRI kinetic heterogeneity and conventional diffusion-weighted imaging on the extent of breast cancer infiltration was compared by analyzing the subjects' operating characteristic curves, sensitivity, specificity, and correlation. Results The Peak, plateau, washout and heterogeneity values of patients in the high-grade group were significantly higher than those in the low-grade group, and the Persistent and ADC values were significantly lower than those in the low-grade group (P< 0.001).However, in terms of Volume values, the difference between the two groups was not statistically significant (P=0.361). Correlation analysis showed that ADC value and persistent value were negatively correlated with pathological grading, while Peak, Plateau, Washout and Heterogeneity were positively correlated with pathological grading. In the diagnostic efficacy assessment, the AUC of Heterogeneity was higher than that of the ADC value (0.910 (95% CI:0.857-0.948) vs. 0.808 (95% CI:0.741-0.863),Z=2.610,P=0.0091). Conclusion Dynamic contrast-enhanced MRI (DCE-MRI) kinetic heterogeneity outperforms apparent diffusion coefficient (ADC) values in the diagnosis of the extent of breast cancer infiltration.

Objective The aim was to explore the value of using multimodal MRI to evaluate renal function in transplanted renal artery stenosis(TRAS). Methods Non-contrast-enhanced magnetic resonance angiography(NCE-MRA)was used to assess the degree of TRAS,arterial spin labeling magnetic resonance imaging(ASL-MRI)to measure renal blood flow(RBF),and intravoxel incoherent motion magnetic resonance imaging(IVIM-MRI)to evaluate diffusion characteristics.According to the degree of TRAS,all patients were categorized into non-stenosis,mild stenosis,and marked stenosis groups.Differences in ASL and IVIM parameters among the three groups and their correlations with estimated glomerular filtration rate(eGFR)were analyzed.A multivariate linear regression model was developed,and its diagnostic performance for transplant kidney dysfunction was assessed using receiver operating characteristic(ROC)analysis. Results The non-stenosis group exhibited significantly higher RBF and f-values compared to the mild and marked stenosis groups(all P<0.05).The degree of stenosis showed a moderate negative correlation with eGFR(r=-0.491,P<0.001),while RBF and f-values displayed moderate positive correlations with eGFR(r=0.669,r=0.597,both P<0.001).The area under the curve(AUC)for the multimodal MRI model identifying impairment of transplanted kidney function was 0.894. Conclusion Multimodal MRI can be used for non-invasive evaluation of TRAS renal function.

Objective To investigate the predictive value of a combination model based on phase Ⅲ enhanced CT imaging and clinical features for early postoperative recurrence of intrahepatic cholangiocarcinoma(ICC). Methods A retrospective analysis was performed on 127 ICC patients who underwent surgical resection and were confirmed by postoperative pathology in the Affiliated Hospital of Yan'an University from January 2019 to January 2024,and they were randomly divided into 89 in the training set and 38 in the verification set according to a ratio of 7∶3.All patients required CT plain scan and three phase enhanced scan before surgery.3D slicer software was used to outline ROI to generate a Mask covering the entire tumor,and then pyradiomics toolkit was used for image preprocessing and image omics feature extraction.The extracted features were sorted by mRMR algorithm,and then the image omics model was established by LASSO screening features and Rad-score was calculated.The combined model was constructed by multi-factor regression,and the predictive efficacy,calibration degree and net benefit of the model were evaluated. Results Based on the three phase enhanced images combined with clinical features,the combined model had the best performance,and the AUC in the training set was 0.908(95%CI:0.828-0.959),the specificity was 0.861,the sensitivity was 0.905,and the AUC in the validation set was 0.835(95%CI:0.679-0.935),the specificity was 0.750,and the sensitivity was 0.863.Delong’s test showed that the efficiency of the combined model was significantly higher than that of the simple image model.In addition,the model had good calibration degree and net income. Conclusion Based on the construction of radiomics combined with clinical features through three-phase enhanced CT,it has a good predictive ability for the early recurrence of intrahepatic cholangiocarcinoma after surgery,and provides important value for early clinical intervention and targeted treatment.

Objective To evaluate the diagnostic efficacy of artificial intelligence (AI)-based non-gated coronary artery calcium score (CACS), and preliminarily explore whether it has the ability to accurately and stably stratify populations with different cardiovascular disease risk factors. Methods A total of 184 patients who underwent low-dose non-gated chest CT plain scan and coronary CT angiography were collected retrospectively. CACS was measured and graded according to the following criteria: CACS = 0 for extremely low risk, 0 < CACS < 100 for low risk, 100 ≤ CACS < 400 for moderate risk, and CACS ≥ 400 for high risk. For the overall population assessment, Spearman correlation coefficient (r), Bland-Altman method, and intraclass correlation coefficient (ICC) were used to evaluate the correlation and consistency between the two methods (non-gated CACS and gated CACS). Weighted Kappa analysis was applied to assess the consistency of coronary artery calcification risk stratification. With electrocardiogram (ECG)-gated calcium score as the gold standard, predicted values were calculated via ordinal Logistic regression, and a multiclass receiver operating characteristic (ROC) curve was plotted using the “one-vs-rest” strategy to evaluate the performance of non-gated CACS. Second-order clustering was used to divide the population into different subgroups based on cardiovascular disease risk factors, with log-likelihood distance metric and Bayesian Information Criterion (BIC) as the bases. In each subgroup, ICC, r, and the diagnostic efficacy for moderate- and high-risk populations (CACS > 100) were further evaluated. A P-value < 0.05 was considered statistically significant. Results For the overall population, the correlation coefficient of non-gated CACS was r = 0.965 (P< 0.001), ICC = 0.970 (P < 0.001), the area under the curve (AUC) for each risk category was > 0.9 (P< 0.001), and the weighted Kappa was 0.854 (P< 0.001). Using second-order clustering, the population was divided into three subgroups. In all subgroups, r > 0.9 (P< 0.001) and ICC > 0.9 (P < 0.001). Additionally, non-gated CACS showed good stratification ability for moderate- and high-risk populations in different subgroups, with AUC > 0.9 in all cases. Conclusion Non-gated calcium score has high reliability and stability for risk grading, and is suitable for coronary heart disease screening in different populations.

Objective To explore in depth the changes in brain atrophy and brain network connectivity in Alzheimer's disease (AD) patients, as well as their correlation, in order to deepen our understanding of the pathogenesis of Alzheimer's disease. Methods This study included multiparametric MRI data from 50 AD patients and 50 healthy controls (HC group), including Three-dimensional T₁ weighted imaging(3D-T₁WI), diffusion tensor imaging (DTI), and resting state functional magnetic resonance imaging (rs-fMRI). The data were quantitatively analyzed using FreeSurfer version 6.0.0 and DPABI software to construct brain structure and functional connectivity maps, and graph theory methods were used to quantitatively analyze the network topology properties of brain structure and function. Results The average gray matter volumes of the motor cortex network, visual network, attention network, default network, and edge network in the AD group were 57, 52, 64, 83, and 38 ml, respectively. Compared with the HC group, the gray matter volume of the motor cortex network, visual network, attention network, default network, and edge network in the AD group were significantly reduced (P<0.05), and there were significant differences in different brain regions. At different network thresholds, the correlation between node atrophy and structural neighbor atrophy, functional neighbor atrophy, and functional right neighbor atrophy in the AD group was 0.77 ± 0.05 (P<0.01), 0.67 ± 0.04 (P<0.01), and 0.79 ± 0.03 (P<0.01), respectively, all higher than their corresponding null models. The FA value of the AD group was significantly lower than that of the HC group (P<0.05). The global network small world attribute of the AD group was lower than that of the HC group, and the clustering coefficients in multiple gray matter and white matter regions are abnormal. Conclusion Multi parameter MRI helps to comprehensively understand the changes in brain atrophy and brain network connectivity in AD patients, as well as their close relationship, providing a new perspective for the study of the pathological mechanisms of AD and contributing to the development of new diagnostic and treatment strategies.

Objective To explore the feasibility of intravoxel incoherent motion-diffusion weighted imaging(IVIM-DWI) in diagnosis of multiple myeloma(MM). Methods The lumbar IVIM-DWI imaging date of 30 patients diagnosed in our hospital and 30 volunteers were collected. The Standard Apparent Diffusion Coefficient(ADCStandard),ture diffusion coefficient(D),pseudo-diffusion coefficient(D^*) and perfusion fraction(f) between the two groups were compared by Mann-Whitney U test,and the diagnostic value of them for MM were also compared by receiver operating characteristic curve（ROC). Results The median of ADC_Standard、D and D^* value in case group were higher than in control group(Z=-4.864,-5.071,-4.494,all P<0.01).The median of value in case group were lower than in control group(Z=-4.620,P<0.01).The ROC curve showed that the ADC_Standard、D and D^* value all had high diagnostic efficacy for multiple myeloma,and their areas under the curve,respectively,were 0.866,0.881,0.838,and 0.847(all P<0.01). Conclusion The IVIM-DWI parameter values of the lumbar spine can provide a suggestive diagnosis for the quantitative assessment of MM.

Objective To investigate the value of radiomics models based on different MRI sequences in predicting microsatellite instability(MSI)status in endometrial cancer. Methods This retrospective study analyzed MRI images of 171 patients with pathologically confirmed endometrial cancer,including 132 microsatellite stable(MSS)and 39 MSI cases.Lesion volumes of interest(VOIs)were delineated using ITK-SNAP software to extract radiomics features from T₂-weighted imaging(T₂WI),diffusion-weighted imaging(DWI),and apparent diffusion coefficient(ADC)maps.Mann-Whitney tests and support vector machine-recursive feature elimination(SVM-RFE)were applied to select the optimal radiomics features for each sequence.Logistic regression models were developed for single-sequence and multi-sequence radiomics predictions.The performance of these models was evaluated using receiver operating characteristic(ROC)curves. Results Among single-sequence radiomics models,the DWI model showed the best predictive performance,with area under the curve(AUC)values of 0.869 and 0.873 in the training and validation sets,respectively.These values were higher than those of the T₂WI model(AUCs of 0.854 and 0.784)and the ADC model(AUCs of 0.835 and 0.809).Compared to the single-sequence models,the combined T₂WI+DWI+ADC model exhibited superior predictive performance,achieving AUC,sensitivity,specificity,positive predictive value and negative predictive value of 0.942,91.3%,86.1%,65.6% and 97.1% in the training set,and 0.914,87.5%,83.0%,60.9% and 95.7% in the validation set. Conclusion Radiomics models based on different MRI sequences demonstrates good predictive performance for MSI status in endometrial cancer,and the combined T₂WI+DWI+ADC multi-sequence model outperform single-sequence models.

Objective To analyze the risk factors of clinically significant prostate cancer (csPCa) using LASSO and multivariate Logistic regression, construct a nomogram prediction model, and explore the diagnostic value of PI-RADSv2.1 score combined with PSA-derived parameters for csPCa. Methods A total of 230 patients with pathologically confirmed prostate cancer or benign prostatic hyperplasia were included retrospectively. They were randomly divided into a training set (n=160) and a validation set (n=70) at a ratio of 7∶3. According to pathological results, patients were classified into the csPCa group (clinically significant prostate cancer) and the no-csPCa group (non-clinically significant prostate cancer). MRI data were analyzed to conduct PI-RADSv2.1 scoring. Prostate-specific antigen density (PSAD), prostate-specific antigen transition zone density (PSAT), free PSA ratio/prostate-specific antigen density [(F/T)/PSAD], and free PSA ratio/prostate-specific antigen transition zone density [(F/T)/PSAT] were calculated. Differences in each parameter between the two groups were compared, and correlation coefficient heatmap analysis was performed. The LASSO regression model was used to screen the most valuable parameters, and multivariate Logistic regression analysis was applied to construct a prediction model and a nomogram. Receiver operating characteristic (ROC) curve, calibration curve, and decision curve were used to evaluate the predictive efficacy of the model. Results In both the training set and validation set, PSAT, PSAD, and PI-RADSv2.1 scores in the csPCa group were significantly higher than those in the no-csPCa group, while (F/T)/PSAT and (F/T)/PSAD were significantly lower than those in the no-csPCa group; the differences in all parameters between the two groups were statistically significant (all P<0.001). Three most valuable parameters were screened out by LASSO regression, including age, PI-RADSv2.1 score, and (F/T)/PSAT. Multivariate Logistic regression analysis showed that age, PI-RADSv2.1 score, and (F/T)/PSAT were independent predictors of csPCa. A nomogram model was established based on these factors. The calibration curve fitted well with the ideal curve. The area under the ROC curve (AUC) of the model was 0.926, with a sensitivity of 88.7% and a specificity of 86.0%. Conclusion This nomogram model has high predictive value for csPCa and provides great clinical benefits.

Objective To explore the change of skeletal muscle mass during neoadjuvant chemotherapy in patients with colorectal cancer and evaluate its relationship with patient prognosis. Methods 178 patients with colorectal cancer who underwent neoadjuvant chemotherapy followed by curative resection in our hospital from August 2014 to December 2022 were selected retrospectively. Skeletal muscle area (SMA) and muscle density (MD) were measured using computed tomography (CT) at the third lumbar spine (L₃) level for evaluating skeletal muscle mass. The paired sample t-test was used to assess the change of skeletal muscle mass during neoadjuvant chemotherapy. Multivariate Cox regression analysis and Kaplan-Meier survival curves were used to evaluate the correlation between the change of skeletal muscle mass and overall survival rate of patients. Results The patients' SMA is significantly reduced after neoadjuvant chemotherapy [(99.42±25.56)cm² vs. (94.07±24.13) cm², t=8.801, P<0.001], with a median decrease rate of 4.32%. Multivariate Cox regression analysis indicated a loss of SMA > 4.32% during neoadjuvant chemotherapy (HR 1.784, 95% CI: 1.014-3.140, P=0.045) was an independent prognostic factor for overall survival rate of patients with colorectal cancer. And Kaplan-Meier survival curves showed that the overall survival of patients with colorectal cancer who lost > 4.32% of their skeletal muscle area during neoadjuvant chemotherapy was significantly reduced (Log-Rank P=0.0015). Conclusion The skeletal muscle area of patients with colorectal cancer is significantly lost during neoadjuvant chemotherapy, and the degree of loss is related to the prognosis.

Objective To explore the feasibility of spectral CT virtual non-contrast (VNC) instead of true non-contrast (TNC) in patients with breast lesions. Methods A retrospective analysis was performed on 27 cases of breast lesions confirmed by pathology. All patients underwent spectral CT chest plain scan (TNC) and dual-phase enhancement. Arterial-phase VNC (VNCa) and venous-phase VNC (VNCv) images were obtained by workstation post-processing. CT values of the mass, normal breast, and pectoralis major muscle were measured in the three groups of images, and the signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and effective dose (ED) were calculated. One-way analysis of variance (ANOVA) and LSD post hoc multiple comparisons were used to analyze the differences in CT values, SNR, and CNR among the three groups of images. Bland-Altman scatter plots were used to analyze the consistency of CT values. Results The CT values and CNR of VNCa and VNCv were not statistically different from TNC (P > 0.05), and the SNR of VNCv images was higher than that of TNC (P= 0.041). Bland-Altman results showed that, except for the VNCv healthy breast CT values which were in poor agreement with TNC (P= 0.01), the CT values of VNCa and VNCv were in good agreement with TNC (P> 0.05). Conclusion Spectral CT breast arterial phase virtual plain scanning can replace true plain scanning and reduce the radiation dose by 33.3%.

Objective To investigate the diagnostic value of Liver Imaging Reporting and Data System (LI-RADS) threshold growth (TG) based on tumor growth rate for hepatocellular carcinoma (HCC) (≤3.0 cm). Methods A retrospective analysis was conducted on liver focal lesions (≤3.0 cm) with baseline and follow-up MRI examinations. Radiologists documented changes in lesion size and LR classification adjustments for each observation during the follow-up period. Tumor growth rate was defined as the percentage increase in lesion size per month, with growth rates of ≥10% per month (TG-10%), ≥20% per month (TG-20%), and ≥30% per month (TG-30%) categorized as modified TG (mTG). The association of TG with a diagnosis of HCC was determined by calculating the diagnostic odds ratio (DOR). With LR-5 as the diagnosis criteria of HCC, the diagnostic performance using LI-RADS TG(-, not as a major feature), LI-RADS TG(+, as a major feature), and LI-RADS mTG(+) criteria was calculated separately, including sensitivity, specificity, and accuracy, and compared using McNemar's test or Fisher's exact probability method. Results 217 patients with 259 observations were included, among which 159 were HCC, 23 were non-HCC malignancies, and 67 were benign lesions. Compared to LI-RADS TG(-) criteria, LI-RADS TG(+) criteria resulted in LR adjustments in 14 HCCs. Specifically, 9 cases were upgraded from LR-3 to LR-5 [≤19 mm + non-rim arterial phase hyperenhancement (APHE) and TG], and 5 cases were upgraded from LR-4 to LR-5 [(≤19 mm + non-rim APHE + enhancing capsule + TG), n=1; (≥20 mm + non-rim APHE + TG), n=5]. TG was significantly associated with HCC, with a diagnostic odds ratio (DOR) of 3.65 (95% CI: 1.63–8.18, P<0.001). LI-RADS TG(+) criteria for diagnosing HCC showed higher sensitivity (66.7% vs. 57.9%, P<0.001) and accuracy (82.2% vs. 78.4%, P=0.002) compared to LI-RADS TG(-) criteria, while maintaining similar specificity (97.8% vs. 98.9%, P=0.317). Compared to LI-RADS TG(+) criteria, LI-RADS mTG-10% criteria significantly improved accuracy for diagnosing HCC (87.9% vs. 82.2%, P<0.001). However, the accuracy with mTG-20% (81.2% vs. 82.2%, P=0.440) and mTG-30% (78.7% vs. 82.2%, P=0.003) remains comparable or slightly lower. The sensitivity for diagnosing HCC using LI-RADS mTG-10% criteria was higher than LI-RADS TG(+) (78.0% vs. 66.7%, P<0.001), whereas their specificities were not significantly different (97.8% vs. 97.8%, P>0.999). Conclusion TG as a major feature in LI-RADS v2018 was significantly associated with HCC. LI-RADS TG(+) criteria improved the sensitivity and accuracy of diagnosing HCC, particularly in lesions presenting with non-rim APHE and TG. Furthermore, LI-RADS mTG-10% criteria further enhanced the diagnostic performance of HCC.

Objective To investigate the relationship between white matter hyperintensity (WMH) score and volume and recent small subcortical infarct (RSSI). Methods Clinical and imaging data of 213 patients with cerebral small vessel disease (CSVD) were retrospectively analyzed.According to the presence or absence of RSSI, the patients were divided into RSSI group (86 cases) and non-RSSI group (127 cases). United Imaging Intelligent CSVD intelligent analysis software was used to quantify Fazekas score and WMH volume automatically.Multifactorial binary logistic regression was used to assess the association of Fazekas score and WMH volume with RSSI. Receiver operating characteristic (ROC) curve was used to evaluate the predictive value of Fazekas score and WMH volume for RSSI. Results There were significant differences in the Fazekas score and WMH volume between the two groups (all P<0.05). Fazekas score (OR=2.825, 95%CI 2.154-4.193, P=0.006) and WMH volume (OR=4.536, 95%CI 2.173-15.719, P=0.014) were independent associated with RSSI. The area under curve of Fazekas score predicting RSSI was 0.688, the optimal cut-off value was 3 points, the sensitivity was 64.1%, and the specificity was 66.7%. The area under curve of WMH volume predicting RSSI was 0.779, the optimal cut-off value was 15.28 cm3, the sensitivity was 68.4%, and the specificity was 80.0%. Conclusion Fazekas score and WMH volume are associated with RSSI, effectively predicting the occurrence of RSSI in the patients with CSVD.

Objective To investigate the predictive value of dynamic contrast-enhanced MRI (DCE-MRI) parameters combined with clinicopathological features for axillary lymph node metastasis (ALNM) in breast cancer. Methods A total of 218 breast cancer patients treated were enrolled as the study subjects. DCE-MRI findings and clinicopathological features were used as independent variables, while the presence or absence of ALNM served as the dependent variable. Univariate and multivariate analyses were performed, and receiver operating characteristic (ROC) curves were plotted to evaluate predictive performance. Results Among the 218 breast cancer patients, 62 (28.44%) had ALNM. Univariate analysis revealed significant differences in age, tumor margin, lymphovascular invasion, Ki-67 expression, B7-H4 positivity, and MTA-3 positivity (all P < 0.05). The ALNM group exhibited higher K_ep and K^trans values than the non-ALNM group (P < 0.05). Multivariate analysis identified lymphovascular invasion, B7-H4 positivity, K_ep, and K^trans as independent predictors of ALNM (P < 0.05). The AUC values for lymphovascular invasion, B7-H4 positivity, K_ep, and K^trans were 0.614, 0.624, 0.846, and 0.859, respectively, while their combined prediction achieved an AUC of 0.939. Conclusion Lymphovascular invasion, B7-H4 positivity, K_ep, and K^trans represent predictive indicators of ALNM in breast cancer, with greater predictive value when used in combination.

Objective To explore the diagnostic value of CT lymphangiography (CTL) and MR lymphangiography (MRL) in lymphatic plastic bronchitis (PB). Methods The clinical and imaging data of 27 patients with clinically confirmed lymphatic plastic bronchitis were analyzed retrospectively. All patients underwent both CTL and MRL examinations. According to the distribution of abnormal lymphatic vessels in the neck and chest on MRL, the disease was classified into four types: TypeⅠ showed tiny abnormal lymphatic vessels in the supraclavicular region and mediastinum; Type Ⅱ showed increased abnormal lymphatic vessels in the supraclavicular region without extension to the mediastinum; Type Ⅲ showed increased abnormal lymphatic vessels in the supraclavicular region with extension to the mediastinum; Type Ⅳshowed abnormal lymphatic vessels in the supraclavicular region with extension to the mediastinum, lung parenchyma, and interstitium. Patients with TypeⅠ and Type Ⅱ were classified into the mild group, while those with Type Ⅲ and Type Ⅳ were classified into the severe group. The CTL imaging findings of each group were recorded, and the CTL imaging indicators included abnormal contrast medium deposition in the lungs, mediastinum, abdominal-pelvic cavity, thoracic duct, and its tributaries. Statistical analysis was performed on the CTL imaging indicators of each group, with a P-value < 0.05 considered statistically significant. Results Based on the range of abnormal lymphatic vessels in the neck and chest on MRL, 27 cases of lymphatic plastic bronchitis were classified into the mild group (10 cases) and the severe group (17 cases). There were no statistically significant differences between the two groups in gender, age of onset, disease course, clinical symptoms, or complicated chylous effusion (all P > 0.05), while there was a statistically significant difference in complicated lymphatic malformation (P = 0.018). The differences in patchy ground-glass opacity, large grid shadow, and thickening of bronchovascular bundles between the two groups were statistically significant (all P < 0.05), with the incidence of these signs higher in the severe group than in the mild group. The differences in abnormal contrast medium deposition around the pericardium, subcarina, pulmonary hilum, and bronchovascular bundles between the two groups were also statistically significant (all P < 0.05), and the incidence was higher in the severe group than in the mild group. Conclusion MRL is helpful for displaying the dilation and range of central lymphatic vessels in the neck and chest, while CTL can show abnormal signs in the lungs as well as the location and degree of systemic lymphatic vessel abnormalities. Both are of great value for the diagnosis and classification of lymphogenic PB, and provide an important imaging basis for the diagnosis and treatment of lymphogenic PB.