Objective To compare the segmenting efficacy of automatic segmentation models for advanced gastric cancer(AGC)on CT virtual monoenergetic images(VMI),non-linear blending images(NLBI)and mixed-energy images(MEI)based on 3D-nnU-Net.Methods Totally 216 cases of AGC were retrospectively enrolled,among them 185 cases were used to construct,train and validate models and divided into training set(n=154)and test set(n=31)at the ratio of 5∶1,while the other 31 cases were used as validation set to evaluate the generalization of the models.The 70 keV energy level VMI(VMI70 keV),NLBI and MEI were reconstructed with whole-abdominal dual-energy mode venous CT,and automatic segmentation models of AGC,including VMI70 keV,NLBI and MEI models were constructed using 3D-nnU-Net,respectively.Taken manually segmented results as golden standards,the efficacy of each model for segmenting all lesions and T2 stage lesions in test set and validation set were evaluated using Dice similarity coefficient(DSC),intersection over union(IoU)and average symmetric surface distance(ASSD).Results For all lesions in test and validation sets,DSC of 3 models were all＞0.80.DSC and IoU of VMI70 keV and NLBI models were both higher,while their ASSD was lower than those of MEI model(all P＜0.05).For T2 stage AGC in both test set and validation set(each n=5),DSC of MEI model was lower than that of VMI70 keV and NLBI models(both P＜0.05),while IoU of MEI model was lower than that of VMI70 keV model(P＜0.05),and its ASSD was higher than that of NLBI model(P＜0.05).Conclusion All 3D-nnU-Net-based VMI70 keV,NLBI and MEI models could effectively segment AGC on dual-energy CT images,and the segmentation efficacy of the former two were better.

Objective To compare the segmenting efficacy of automatic segmentation models for advanced gastric cancer(AGC)on CT virtual monoenergetic images(VMI),non-linear blending images(NLBI)and mixed-energy images(MEI)based on 3D-nnU-Net.Methods Totally 216 cases of AGC were retrospectively enrolled,among them 185 cases were used to construct,train and validate models and divided into training set(n=154)and test set(n=31)at the ratio of 5∶1,while the other 31 cases were used as validation set to evaluate the generalization of the models.The 70 keV energy level VMI(VMI70 keV),NLBI and MEI were reconstructed with whole-abdominal dual-energy mode venous CT,and automatic segmentation models of AGC,including VMI70 keV,NLBI and MEI models were constructed using 3D-nnU-Net,respectively.Taken manually segmented results as golden standards,the efficacy of each model for segmenting all lesions and T2 stage lesions in test set and validation set were evaluated using Dice similarity coefficient(DSC),intersection over union(IoU)and average symmetric surface distance(ASSD).Results For all lesions in test and validation sets,DSC of 3 models were all＞0.80.DSC and IoU of VMI70 keV and NLBI models were both higher,while their ASSD was lower than those of MEI model(all P＜0.05).For T2 stage AGC in both test set and validation set(each n=5),DSC of MEI model was lower than that of VMI70 keV and NLBI models(both P＜0.05),while IoU of MEI model was lower than that of VMI70 keV model(P＜0.05),and its ASSD was higher than that of NLBI model(P＜0.05).Conclusion All 3D-nnU-Net-based VMI70 keV,NLBI and MEI models could effectively segment AGC on dual-energy CT images,and the segmentation efficacy of the former two were better.

Objective To build an automatic segmentation model of temporomandibular joint(TMJ)based on magnetic resonance im-aging(MRI)using deep learning method.Methods The MRI data of TMJ of 104 subjects were collected,with the articular disc,con-dyle and glenoid fossa marked.The adaptive U-Net framework(nnU-Net)was used to construct a segmentation model,which was sub-jected to both quantitative and qualitative assessments.Results The segmentation model demonstrated excellent accuracy in segmenta-tion.In the segmentation of different joint structures,the model achieved Dice of 0.77 for the articular disc,0.85 for the condyle,and 0.66 for the glenoid fossa.The model showed similar segmentation performance when processing MRI images in both open-mouth and closed-mouth states.Conclusion This study developed an automatic segmentation model for TMJ MRI based on deep learning,which can assist clinicians in diagnosing anterior displacement of the TMJ disc.

Objective To build an automatic segmentation model of temporomandibular joint(TMJ)based on magnetic resonance im-aging(MRI)using deep learning method.Methods The MRI data of TMJ of 104 subjects were collected,with the articular disc,con-dyle and glenoid fossa marked.The adaptive U-Net framework(nnU-Net)was used to construct a segmentation model,which was sub-jected to both quantitative and qualitative assessments.Results The segmentation model demonstrated excellent accuracy in segmenta-tion.In the segmentation of different joint structures,the model achieved Dice of 0.77 for the articular disc,0.85 for the condyle,and 0.66 for the glenoid fossa.The model showed similar segmentation performance when processing MRI images in both open-mouth and closed-mouth states.Conclusion This study developed an automatic segmentation model for TMJ MRI based on deep learning,which can assist clinicians in diagnosing anterior displacement of the TMJ disc.

Objective:To investigate the value of CT radiomic model based on analysis of primary gastric cancer and the adipose tissue outside the gastric wall beside cancer in differentiating stage T1-2 from stage T3-4 gastric cancer.Methods:This study was a case-control study. Totally 465 patients with gastric cancer treated in Affiliated People′s Hospital of Jiangsu University from December 2011 to December 2019 were retrospectively collected. According to postoperative pathology, they were divided into 2 groups, one with 150 cases of T1-2 tumors and another with 315 cases of T3-4 tumors. The cases were divided into a training set (326 cases) and a test set (139 cases) by stratified sampling method at 7∶3. There were 104 cases of T1-2 stage and 222 cases of T3-4 stage in the training set, 46 cases of T1-2 stage and 93 cases of T3-4 stage in the test set. The axial CT images in the venous phase during one week before surgery were selected to delineate the region of interest (ROI) at the primary lesion and the extramural gastric adipose tissue adjacent to the cancer areas. The radiomic features of the ROIs were extracted by Pyradiomics software. The least absolute shrinkage and selection operator was used to screen features related to T stage to establish the radiomic models of primary gastric cancer and the adipose tissue outside the gastric wall beside cancer. Independent sample t test or χ2 test were used to compare the differences in clinical features between T1-2 and T3-4 patients in the training set, and the features with statistical significance were combined to establish a clinical model. Two radiomic signatures and clinical features were combined to construct a clinical-radiomics model and generate a nomogram. The area under the receiver operating characteristic curve (AUC) was used to evaluate the efficacy of each model in differentiating stage T1-2 from stage T3-4 gastric cancer. The calibration curve was used to evaluate the consistency between the T stage predicted by the nomogram and the actual T stage of gastric cancer. And the decision curve analysis was used to evaluate the clinical net benefit of treatment guided by the nomogram and by the clinical model. Results:There were significant differences in CT-T stage and CT-N stage between T1-2 and T3-4 patients in the training set ( χ2=10.59, 15.92, P=0.014, 0.001) and the clinical model was established. After screening and dimensionality reduction, the 5 features from primary gastric cancer and the 6 features from the adipose tissue outside the gastric wall beside cancer established the radiomic models respectively. In the training set and the test set, the AUC values of the primary gastric cancer radiomic model were 0.864 (95% CI 0.820-0.908) and 0.836 (95% CI 0.762-0.910), and the adipose tissue outside the gastric wall beside cancer radiomic model were 0.782 (95% CI 0.731-0.833) and 0.784 (95% CI 0.702-0.866). The AUC values of the clinical model were 0.761 (95% CI 0.705-0.817) and 0.758 (95% CI 0.671-0.845), and the nomogram were 0.876 (95% CI 0.835-0.917) and 0.851 (95% CI 0.781-0.921). The calibration curve reflected that there was a high consistency between the T stage predicted by the nomogram and the actual T stage in the training set ( χ2=1.70, P=0.989). And the decision curve showed that at the risk threshold 0.01-0.74, a higher clinical net benefit could be obtained by using a nomogram to guide treatment. Conclusions:The CT radiomics features of primary gastric cancer lesions and the adipose tissue outside the gastric wall beside cancer can effectively distinguish T1-2 from T3-4 gastric cancer, and the combination of CT radiomic features and clinical features can further improve the prediction accuracy.

Objective:This study aims to explore the predictive value of T2-weighted imaging (T2WI), apparent diffusion coefficient (ADC), and early-delayed phases enhanced magnetic resonance imaging (DCE-MRI) radiomics prediction model in determining human epidermal growth factor receptor 2 status in breast cancer.Methods:A retrospective study was conducted, involving 187 patients with confirmed breast cancer by postsurgical pathology at Zhenjiang First People's Hospital during January 2021 and May 2023. Immunohistochemistry or fluorescence in situ hybridization was used to determine the HER-2 status of these patients, with 48 cases classified as HER-2 positive and 139 cases as HER-2 negative. The training set was used to construct the prediction models and the validation set was used to verify the prediction models. Layers of T2WI, ADC, and early-delayed phase DCE-MRI images were used to delineate the volumeof interest and 960 radiomic features were extracted from each case using Pyradiomic. After screening and dimensionality reduction by intraclass correlation coefficient, Pearson correlation analysis, least absolute shrinkage, and selection operator, the radiomics labels were established. Logistic regression analysis was used to construct the T2WI radiomics model, ADC radiomics model, DCE-2 radiomics model, DCE-6 radiomics model, and the joint sequence radiomics model to predict the HER-2 expression status of breast cancer, respectively. Based on the clinical, pathological, and MRI image characteristics of patients, univariate and multivariate logistic regression analysis wasused to construct a clinicopathological MRI feature model. The radscore of every patient and the clinicopathological MRI features which were statistically significant after screening were used to construct a nomogram model. The receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of each model and the decision curve analysis wasused to evaluate the clinical usefulness.Results:The T2WI, ADC, DCE-2, DCE-6, and joint sequence radiomics models, the clinicopathological MRI feature model, and the nomogram model were successfully constructed to predict the expression status of HER-2 in breast cancer. ROC analysis showed that in the training set and validation set, the areas under the curve (AUC) of the T2WI radiomics model were 0.797 and 0.760, of the ADC radiomics model were 0.776 and 0.634, of the DCE-2 radiomics model were 0.804 and 0.759, of the DCE-6 radiomics model were 0.869 and 0.798, of the combined sequence radiomics model were 0.908 and 0.847, of the clinicopathological MRI feature model were 0.703 and 0.693, and of the nomogram model were 0.938 and 0.859, respectively. In the training set, the combined sequence radiomics model outperformed the clinicopathological features model ( P＜0.001). In the training and validation sets, the nomogram outperformed the clinicopathological features model ( P＜0.05). In addition, the diagnostic performance of the nomogram was better than that of the four single-modality radiomics models in the training cohort ( P＜0.05) and was better than that of DCE-2 and ADC models in the validation cohort ( P＜0.05). Decision curve analysis indicated that the value of individualized prediction models was higher than clinical and pathological prediction models in clinical practice. The calibration curve showed that the multimodal radiomics model had a high consistency with the actual results in predicting HER-2 expression. Conclusions:T2WI, ADC and early-delayed phase DCE-MRI imaging histology models for HER-2 expression status in breast cancer are expected to provide a non-invasive virtual pathological basis for decision-making on preoperative neoadjuvant regimens in breast cancer.

Objective:This study aims to explore the predictive value of T2-weighted imaging (T2WI), apparent diffusion coefficient (ADC), and early-delayed phases enhanced magnetic resonance imaging (DCE-MRI) radiomics prediction model in determining human epidermal growth factor receptor 2 status in breast cancer.Methods:A retrospective study was conducted, involving 187 patients with confirmed breast cancer by postsurgical pathology at Zhenjiang First People's Hospital during January 2021 and May 2023. Immunohistochemistry or fluorescence in situ hybridization was used to determine the HER-2 status of these patients, with 48 cases classified as HER-2 positive and 139 cases as HER-2 negative. The training set was used to construct the prediction models and the validation set was used to verify the prediction models. Layers of T2WI, ADC, and early-delayed phase DCE-MRI images were used to delineate the volumeof interest and 960 radiomic features were extracted from each case using Pyradiomic. After screening and dimensionality reduction by intraclass correlation coefficient, Pearson correlation analysis, least absolute shrinkage, and selection operator, the radiomics labels were established. Logistic regression analysis was used to construct the T2WI radiomics model, ADC radiomics model, DCE-2 radiomics model, DCE-6 radiomics model, and the joint sequence radiomics model to predict the HER-2 expression status of breast cancer, respectively. Based on the clinical, pathological, and MRI image characteristics of patients, univariate and multivariate logistic regression analysis wasused to construct a clinicopathological MRI feature model. The radscore of every patient and the clinicopathological MRI features which were statistically significant after screening were used to construct a nomogram model. The receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of each model and the decision curve analysis wasused to evaluate the clinical usefulness.Results:The T2WI, ADC, DCE-2, DCE-6, and joint sequence radiomics models, the clinicopathological MRI feature model, and the nomogram model were successfully constructed to predict the expression status of HER-2 in breast cancer. ROC analysis showed that in the training set and validation set, the areas under the curve (AUC) of the T2WI radiomics model were 0.797 and 0.760, of the ADC radiomics model were 0.776 and 0.634, of the DCE-2 radiomics model were 0.804 and 0.759, of the DCE-6 radiomics model were 0.869 and 0.798, of the combined sequence radiomics model were 0.908 and 0.847, of the clinicopathological MRI feature model were 0.703 and 0.693, and of the nomogram model were 0.938 and 0.859, respectively. In the training set, the combined sequence radiomics model outperformed the clinicopathological features model ( P＜0.001). In the training and validation sets, the nomogram outperformed the clinicopathological features model ( P＜0.05). In addition, the diagnostic performance of the nomogram was better than that of the four single-modality radiomics models in the training cohort ( P＜0.05) and was better than that of DCE-2 and ADC models in the validation cohort ( P＜0.05). Decision curve analysis indicated that the value of individualized prediction models was higher than clinical and pathological prediction models in clinical practice. The calibration curve showed that the multimodal radiomics model had a high consistency with the actual results in predicting HER-2 expression. Conclusions:T2WI, ADC and early-delayed phase DCE-MRI imaging histology models for HER-2 expression status in breast cancer are expected to provide a non-invasive virtual pathological basis for decision-making on preoperative neoadjuvant regimens in breast cancer.

Objective To observe the value of automated net water uptake(CTP-aNWU)based on CT perfusion(CTP)for predicting neurological prognosis of acute ischemic stroke(AIS).Methods A total of 145 AIS patients due to anterior circulation large vessel occlusion were retrospectively enrolled,and the clinical data at admission and follow-up were collected.Alberta stroke program early CT score net water uptake(ASPECTS-NWU)was analyzed,and CTP-aNWU of the infarct core was obtained based on CTP and non-contrast CT(NCCT).According to modified Rankin scale(mRS)score 90 days after the onset of infarction,the patients were divided into favorable prognosis group(mRS score≤2,n=54)and poor prognosis group(mRS score＞2,n=91).The clinical data and imaging data were compared between groups,and the independent predictors of neurological prognosis of AIS were evaluated,and the predictive efficacies were assessed.Results There were significant differences in age,admission NIHSS score and admission ASPECTS of patients,as well as in ASPECTS-NWU,CTP-aNWU,infarct core volume,hypoperfusion factor of the lesions between groups(all P＜0.05).The infarct core volume(OR=0.977[0.963 to 0.992],P=0.002)and CTP-aNWU(OR=0.876[0.793 to 0.969],P=0.010)were independent predictors of neurological prognosis of AIS.The area under the curve(AUC)of receiver operating characteristic curve of CTP-aNWU alone for predicting neurological prognosis of AIS patients 90 days after the onset was 0.634(95％CI[0.550,0.713]),which was 0.790(95％CI[0.714,0.853])of CTP-aNWU combined with infarct core volume,and the latter was better than the former(Z=3.500,P＜0.001).Conclusion CTP-aNWU was an independent predictor of neurological prognosis 90 days after the onset of AIS.Combining CTP-aNWU with infarct core volume could improve the predicting efficacy.

Objective To observe the impact of readout-segmented echo-planar imaging(RS-EPI)based on small field of view(FOV)and saturation band on image quality of orbital diffusion weighted imaging(DWI).Methods Orbital MR scanning were prospectively performed in 33 healthy subjects.T1W,RS-RPI and optimized RS-EPI(based on small FOV and saturation band)images were acquired.Imaging quality of RS-EPI and optimized RS-EPI on displaying intraorbital structures(eyeballs,optic nerve and intraconal compartment)and periorbital structures(nasal cavity,orbital gyrus,optic chiasma,pituitary and temporal lobe)were evaluated subjectively using 5-point method.The geometric parameters of eyeball,signal-to-noise ratio(SNR)and apparent diffusion coefficient(ADC)were compared between images of RS-EPI and optimized RS-EPI.Results The display of eyeball,nasal cavity,orbital gyrus,optic chiasm,pituitary and temporal lobe on optimized RS-EPI were all better than those on RS-EPI(all P＜0.01),whereas no significant difference of optic nerve nor intraconal compartment was found between optimized RS-EPI and RS-EPI images(P＞0.05).Deformations of eyeball volume,sphericity,surface area,3D maximum diameter,axial maximum diameter and sagittal maximum diameter of bilateral eyes on optimized RS-EPI images were all slighter than those on RS-EPI(all P＜0.01),whereas no significant difference of deformations of coronal maximum diameters was found between optimized RS-EPI and RS-EPI images(P＞0.05).SNR of left temporal lobe white matter and ADC of vitreous body on optimized RS-EPI images were both lower than those on RS-EPI(both P＜0.01),whereas no significant difference of ADC of left temporal lobe white matter nor that of pons was found between optimized RS-EPI and RS-EPI images(P＞0.05).Conclusion Optimized RS-EPI with small FOV and saturation band could be used to improve imaging quality of orbital DWI.

We propose an effective algorithm for accurate 3D segmentation of CT liver images based on statistical and specific information. We present a new intensity model which combines patient-specific intensity information of boundary with the statistical information for liver segmentation. Compared to the traditional methods, our approach not only produces excellent segmentation accuracy, but also increases the robustness.
Algorithms ; Humans ; Imaging, Three-Dimensional ; methods ; Liver ; diagnostic imaging ; Liver Diseases ; diagnostic imaging ; Liver Neoplasms ; diagnostic imaging ; Models, Statistical ; Radiographic Image Interpretation, Computer-Assisted ; methods ; Tomography, X-Ray Computed ; methods