Objective:To explore the capability of deep learning image reconstruction (DLIR) compared to adaptive statistical iterative reconstruction (ASiR-V) in improving the image quality and myocardial edge sharpness of dynamic stress myocardial CT perfusion imaging (CTP).Methods:Thirty subjects who underwent dynamic stress myocardial CTP at Guangdong Provincial People′s Hospital from September 2023 to February 2024 were recruited. Image data of all enrolled patients were reconstructed using ASiR-V 50%, ASiR-V 80%, medium-intensity DLIR(DLIR-M), and high-intensity DLIR(DLIR-H), respectively. Regions of interest were selected in the left ventricular cavity, interventricular septum, and left ventricular lateral wall for measurement of CT values and standard deviations (SD), and calculation of signal to noise ratio (SNR) and contrast to noise ratio (CNR). Matlab was utilized to obtain the differences (d) and slopes (s) of CT value changes at four left ventricular myocardial edges for objective edge sharpness evaluation. Two radiologists subjectively scored the images for noise, natural appearance, and edge sharpness. In case of disagreement between the two radiologists, a third senior radiologist′s score was decisive. Left ventricular myocardial blood flow (MBF) of ASiR-V and DLIR images with lower SD, higher SNR and CNR were calculated, respectively. When the normal distribution was satisfied, the independent sample t test was used for comparison between two groups, and the random block design ANOVA was used for comparison between multiple groups. And analysis was conducted using Friedman test for non-normally distributed data, and Bonferroni correction for pairwise comparisons. Results:There were statistically significant differences in SD, SNR, and CNR among the four images in the interventricular septum and left ventricular lateral wall (all P<0.05), with ASiR-V 80% and DLIR-H demonstrating the lowest SD, highest SNR and CNR, and the subjective image noise score. Statistically significant differences were observed in d and s for the four left ventricular myocardial edges (all P<0.05), with DLIR-M and DLIR-H exhibiting the best objective edge sharpness [5 (5, 5)], and ASiR-V 80% the worst [3.5 (3, 4)]. In the subjective scores for natural appearance, DLIR-M and DLIR-H received the highest scores [5 (5, 5)], while ASiR-V 80% received the lowest scores [3 (3, 4)], with statistically significant differences (all P<0.05). There was no statistically significant difference in MBF values calculated from ASiR-V 80% and DLIR-H images (all P>0.05). Conclusions:The SD value, SNR and CNR of dynamic stress myocardial CTP images reconstructed by DLIR-H are equivalent to ASiR-V 80%, and using DLIR-H can improve the edge sharpness of left ventricular myocardium without affecting the calculation of MBF.

Purpose To analyze the value of a deep learning(DL)radiomics model based on dynamic contrast-enhanced MRI images in predicting the expression of Ki-67 status in breast cancer.Materials and Methods A retrospective analysis of 152 breast cancer patients confirmed by pathological results at Guangdong Women and Children Hospital,MRI images and clinical pathological data were reviewed,and based on postoperative immunohistochemistry results,the images of the high and low expression groups of Ki-67 were randomly sampled in a ratio of 8∶2 to form a training set of 122 cases and a validation set of 30 cases.Single-factor and multi-factor Logistic regression analyses of clinical data were performed to select independent predictors of breast cancer expressing Ki-67 status.The ResNet-18 model was used as the basic model for DL feature extraction.Hand-crafted radiomic features and DL features were extracted.Eight machine learning models were constructed based on clinical features,hand-crafted radiomic features,DL features,and their combinations.The area under the receiver operating characteristic curve was used to evaluate the predictive performance of the models,and the best model was determined as the output model.Results The progesterone receptor status(OR=0.764,P=0.040)and human epidermal growth factor receptor-2 status(OR=1.187,P=0.046)were independent clinical predictors of breast cancer expressing Ki-67 status.The combined feature models demonstrated superior performance over the individual feature models,and the support vector machine algorithm had the highest prediction performance in the validation set,with an area under the curve of 0.847.Conclusion The DL radiomics model based on dynamic contrast-enhanced MRI images can effectively predict the expression of Ki-67 status in breast cancer.The support vector machine algorithm combined with feature model is the best,which can help the clinical diagnosis and treatment of breast cancer and prognosis evaluation.

Purpose To analyze the value of a deep learning(DL)radiomics model based on dynamic contrast-enhanced MRI images in predicting the expression of Ki-67 status in breast cancer.Materials and Methods A retrospective analysis of 152 breast cancer patients confirmed by pathological results at Guangdong Women and Children Hospital,MRI images and clinical pathological data were reviewed,and based on postoperative immunohistochemistry results,the images of the high and low expression groups of Ki-67 were randomly sampled in a ratio of 8∶2 to form a training set of 122 cases and a validation set of 30 cases.Single-factor and multi-factor Logistic regression analyses of clinical data were performed to select independent predictors of breast cancer expressing Ki-67 status.The ResNet-18 model was used as the basic model for DL feature extraction.Hand-crafted radiomic features and DL features were extracted.Eight machine learning models were constructed based on clinical features,hand-crafted radiomic features,DL features,and their combinations.The area under the receiver operating characteristic curve was used to evaluate the predictive performance of the models,and the best model was determined as the output model.Results The progesterone receptor status(OR=0.764,P=0.040)and human epidermal growth factor receptor-2 status(OR=1.187,P=0.046)were independent clinical predictors of breast cancer expressing Ki-67 status.The combined feature models demonstrated superior performance over the individual feature models,and the support vector machine algorithm had the highest prediction performance in the validation set,with an area under the curve of 0.847.Conclusion The DL radiomics model based on dynamic contrast-enhanced MRI images can effectively predict the expression of Ki-67 status in breast cancer.The support vector machine algorithm combined with feature model is the best,which can help the clinical diagnosis and treatment of breast cancer and prognosis evaluation.

Objective:To explore the capability of deep learning image reconstruction (DLIR) compared to adaptive statistical iterative reconstruction (ASiR-V) in improving the image quality and myocardial edge sharpness of dynamic stress myocardial CT perfusion imaging (CTP).Methods:Thirty subjects who underwent dynamic stress myocardial CTP at Guangdong Provincial People′s Hospital from September 2023 to February 2024 were recruited. Image data of all enrolled patients were reconstructed using ASiR-V 50%, ASiR-V 80%, medium-intensity DLIR(DLIR-M), and high-intensity DLIR(DLIR-H), respectively. Regions of interest were selected in the left ventricular cavity, interventricular septum, and left ventricular lateral wall for measurement of CT values and standard deviations (SD), and calculation of signal to noise ratio (SNR) and contrast to noise ratio (CNR). Matlab was utilized to obtain the differences (d) and slopes (s) of CT value changes at four left ventricular myocardial edges for objective edge sharpness evaluation. Two radiologists subjectively scored the images for noise, natural appearance, and edge sharpness. In case of disagreement between the two radiologists, a third senior radiologist′s score was decisive. Left ventricular myocardial blood flow (MBF) of ASiR-V and DLIR images with lower SD, higher SNR and CNR were calculated, respectively. When the normal distribution was satisfied, the independent sample t test was used for comparison between two groups, and the random block design ANOVA was used for comparison between multiple groups. And analysis was conducted using Friedman test for non-normally distributed data, and Bonferroni correction for pairwise comparisons. Results:There were statistically significant differences in SD, SNR, and CNR among the four images in the interventricular septum and left ventricular lateral wall (all P<0.05), with ASiR-V 80% and DLIR-H demonstrating the lowest SD, highest SNR and CNR, and the subjective image noise score. Statistically significant differences were observed in d and s for the four left ventricular myocardial edges (all P<0.05), with DLIR-M and DLIR-H exhibiting the best objective edge sharpness [5 (5, 5)], and ASiR-V 80% the worst [3.5 (3, 4)]. In the subjective scores for natural appearance, DLIR-M and DLIR-H received the highest scores [5 (5, 5)], while ASiR-V 80% received the lowest scores [3 (3, 4)], with statistically significant differences (all P<0.05). There was no statistically significant difference in MBF values calculated from ASiR-V 80% and DLIR-H images (all P>0.05). Conclusions:The SD value, SNR and CNR of dynamic stress myocardial CTP images reconstructed by DLIR-H are equivalent to ASiR-V 80%, and using DLIR-H can improve the edge sharpness of left ventricular myocardium without affecting the calculation of MBF.

Objective To investigate the preventive effect of atorvastatin calcium tablets(ACT)combined with colchicine(COL)on in-stent restenosis(ISR)after percutaneous coronary intervention(PCI).Methods Clinical data of patients with acute coronary syndrome(ACS)after PCI at Jianyang People's Hospital from January 2020 to June 2023 were retrospectively analyzed.According to the postoperative treatment plans after PCI,they were divided into the ACT group(Aspirin enteric-coated tablets+Clopidogrel bisulfate tablets+ACT)and the combined group(Aspirin enteric-coated tablets+Clopidogrel bisulfate tablets+ACT+COL).The observation indicators include minimum lumen diameter(MLD)within the stent,ISR rate,blood lipid parameters(HDL,LDL,TG,and TC),and inflammatory markers(hs-CRP and IL-35).In addition,the incidence of major adverse cardiovascular events(MACEs)and drug-related adverse reactions were observed and recorded.Results A total of 479 patients were included in the study,with 249 cases in the ACT group and 230 cases in the combined group.The difference in MLD between the two groups in the immediate postoperative period was not statistically significant(P＞0.05),and at 12 months postoperatively,the MLD of patients in both groups decreased significantly(P＜0.05),and the MLD of the combined group was lower than that of the ACT group(P＜0.05).The ISR rate was significantly lower in the combined group than in the ACT group(P＜0.05).The differences in preoperative lipid parameters and inflammation indicators between the two groups were not statistically significant(P＞0.05).LDL,TG,TC,and hs-CRP decreased significantly at 12 months postoperatively compared with preoperative period,while HDL and IL-35 increased significantly compared with preoperative period(P＜0.05).At 12 months postoperatively,the differences in HDL,LDL,TC,and TG between the two groups were not statistically significant(P＞0.05);compared with the ACT group,the hs-CRP levels in the combined group decreased significantly,whereas the IL-35 levels were elevated(P＜0.05).With regard to MACEs,the rate of myocardial re-infarction and the incidence of any MACEs events in the combined group were lower than those in the ACT group(P＜0.05),and the rate of emergency coronary revascularization,stroke and cardiac mortality were not statistically different(P＞0.05).Regarding drug-related adverse reactions,the differences between the two groups in the incidence of gastrointestinal reactions,the incidence of bleeding,the incidence of hematopenia,transaminase elevation,muscle soreness,infection,and any related adverse events were not statistically significant(P＞0.05).Conclusion ACT combined with COL improve inflammation levels and reduce the incidence of ISR and MACEs,in ACS patients after PCI,but has a smaller impact on blood lipid parameters.and without adding additional drug-related adverse reactions.

Cardiovascular disease is the leading cause of death among Chinese residents, and non-invasive imaging technology has important value in the diagnosis and treatment of cardiovascular disease. Cardiac magnetic resonance (CMR) can characterize cardiac pathophysiological information from multiple dimensions, including cardiac structure, function, tissue characteristics, and microstructure, through multi parameter and multi sequence " one-stop" imaging. This article will focus on new technologies such as CMRT1 mapping, feature tracking, and diffusion tensor imaging, and explain their applications and progress in the diagnosis, efficacy monitoring, and prognosis prediction of various myocardial lesions such as non ischemic heart disease and ischemic heart disease.

Objective:To apply cardiovascular magnetic resonance (CMR) to quantitatively evaluate the morphology and tissue characteristics of thoracic skeletal muscle (TSM) in patients with Danon disease and hypertrophic cardiomyopathy (HCM), in order to provide auxiliary differential diagnostic information.Methods:A retrospective study was conducted on 15 Danon disease patients (Danon disease group) who completed CMR examination, and 15 HCM patients (HCM group) and 15 healthy volunteers (control group) who were matched by gender and age were compared. TSM (pectoralis major, pectoralis minor, erector spinae, and subscapularis) area index (TSMAi), T1 relaxation time of four groups, and extracellular volume (ECV) of skeletal muscles of two groups (pectoralis major and subscapularis) were measured for all subjects. Single factor analysis of variance and KruskalWallis test were used to compare three sets of values.Results:The total TSMAi of the Danon disease group was lower than that of the HCM group and the control group [(15.37±3.28)cm 2/m 2 vs (23.02±3.88)cm 2/m 2 vs (22.33±4.67)cm 2/m 2, P<0.001], and the TSMAi of each TSM was also lower than that of the HCM group and the control group (all P<0.05). There was no statistically significant difference in TSMAi between the HCM group and the control group (all P>0.05). The native T1 values of the pectoralis major muscle in the Danon disease group and HCM group were higher than those in the control group (all P<0.05). The ECV of the pectoralis major muscle and subscapularis muscle in the Danon disease group were higher than those in the HCM group and control group, and the enhanced T1 values were lower than those in the HCM group and control group (all P<0.05); There was no statistically significant difference in ECV and the enhanced T1 values between the HCM group and the control group ( P>0.05). Conclusions:The application of CMR can effectively evaluate the changes in TSM morphology and tissue characteristics in Danon disease patients. Compared with HCM patients, Danon disease patients showed significant atrophy of TSM with increased extracellular volume. CMR provides a quantitative reference for TSM in the differential diagnosis of the two.

Objective:To investigate the hemodynamic changes in the ascending aorta (AAo) before and after interventricular septal myocardial resection in obstructive hypertrophic cardiomyopathy (HOCM) using cardiac magnetic resonance four-dimensional blood flow (CMR 4D Flow) technology.Methods:HOCM patients who underwent interventricular septal myocardial resection at Guangdong Provincial People′s Hospital from May 2021 to September 2022 were prospectively included. Age and gender matched healthy volunteers (control group) were included during the same period. Both the control group and HOCM patients underwent CMR examination (including cine sequence and 4D Flow sequence) before and 6 months after surgery. CMR 4D flow technology was used to evaluate changes in AAo preoperative and postoperative blood flow patterns (eddy currents, spiral flow), maximum energy loss (EL max), and average energy loss (EL avg). HOCM patients underwent laboratory tests, including N-terminal pro-brain natriuretic peptide (N-pro BNP) and high-sensitivity troponin T (hsTnT). At the same time, the correlation between postoperative energy loss in HOCM patients and the degree of improvement in laboratory biomarkers was explored. Results:A total of 15 HOCM patients and 15 healthy volunteers were included. (1) In terms of blood flow patterns, the preoperative spiral flow degree of HOCM patients was significantly higher than that of the control group ( P=0.001), but the postoperative difference was not statistically significant ( P=0.059). The degree of eddy currents in HOCM patients before and after surgery was higher than that in the control group (all P<0.05). (2) In terms of energy loss, the preoperative EL max [21.17(14.30-28.10)mW vs 10.17(7.66-13.07)mW, P<0.001] and EL avg [4.87(3.46-5.77)mW vs 2.27(2.19-2.27)mW, P=0.023] of HOCM patients were higher than those of the control group, but there was no statistically significant difference between the postoperative and control groups (all P>0.05). Compared with preoperative, the postoperative EL max [12.33(8.70-17.41)mW] and EL avg [3.10(2.25-4.40)mW] of AAo in HOCM patients were significantly reduced (mean P=0.001). (3) Correlation analysis showed that there was a positive correlation ( r=0.587, P=0.021) between the EL max of AAo and the degree of improvement in hsTNT after interventricular septum myocardial resection, but no significant correlation ( r=0.229, P=0.413) with the degree of improvement in NT-pro BNP. Conclusions:The degree of postoperative AAo blood flow disorder in HOCM patients is reduced, and EL max and EL avg are significantly reduced. The EL max of postoperative AAo is positively correlated with the degree of improvement in hsTNT, suggesting that EL max may be applicable for prognostic evaluation of patients.

Objective: Review the independent risk factors of postoperative recurrence in surgical treatment of bladder cancer patients to construct a model of bladder cancer recurrence. Methods : A total of 240 surgically treated bladder cancer patients were followed up for at least 1 year and divided into recurrence ( n = 54) and non⁃recurrence (n = 186) . The general data of patients were comparative analyzed , and the different and statistically significant data were further analyzed by ROC curve , and the statistically significant data were included in the multivariate analysis after logistic obtaining univariate analysis results. Risk factors were included in the model construction , and the model correction curve and clinical net benefit analysis were analyzed. The model could be used to predict postoperative recurrence in bladder cancer patients. Results: The ROC curves of the statistically significant continuous variables were analyzed in the general data , and the results showed that the AUC of PNI , BLCA⁃4 , BTA , NMP22 and CEA were 0. 932 , 0. 979 , 0. 998 , 0. 677 and 0. 981 , respectively , and the optimal truncation values were ≤40. 18% , > 140. 04 ng/mg , ≤7. 22 U/mg , > 7. 68 μg/mg , and > 1. 99 ng/mg, respectively. Statistically significant data from univariate analysis were incorporated into the logistic regression model , and the results showed that PNI ≤40. 18% , BLCA⁃4 > 140. 04 ng/mg , BTA≤7. 22 U/mg , NMP22 > 7. 68 μg/mg was a risk factor for recurrence in patients with bladder cancer. Subsequently , PNI , BLCA⁃4 , BTA , and NMP22 were incorporated into the construction of the model as predictors of recurrence in patients with bladder cancer. Based on the model correction curve and clinical net benefit analysis , the internal verification results showed that the C ⁃index of the model predicting bladder cancer recurrence was 0. 296 (95% CI: 0. 078 - 1. 329) . The calibration curve showed good consistency between the observed and predicted values. The model predicted a risk threshold > 0. 128 for patients with bladder cancer, and the model provided a clinical net benefit; in addition , the model had a higher clinical net benefit than PNI ,BLCA⁃4 , BTA , and NMP22. Conclusion The model correction curve and clinical net benefit analysis , the results of internal verification show that the model can be used to predict recurrence in patients with bladder cancer.

Objective:To evaluate the diagnostic value of multiparametric cardiac magnetic resonance(CMR)or detecting the occurrence of acute rejection(AR)after heart transplantation(HT).Methods:From 2019 to 2021, 44 HT recipients are prospectively recruited from Guangdong Provincial People's Hospital.Another 51 healthy volunteers are recruited from a local community as healthy controls.CMR studies are performed for obtaining baseline parameters.According to the clinicopathological diagnostic criteria of AR by the consensus of International Society for Heart and Lung Transplantation, 81 CMR studies of 44 HT recipients are further divided into two groups of AR (18 cases)and non-AR(71 cases). CMR parameters includ global ventricular structure/function, T2, T1, extracellular volume(ECV)and late gadolinium enhancement(LGE). A combined model is established by binary Logistic regression and receiver operator characteristic curve(ROC)constructed.Results:The age range is(41.8±16.8)years in 44 HT recipients and(41.8±9.7)years in 51 healthy controls.T1 mapping indicated that myocardial global ECV of left ventricle is significantly higher in AR patients than non-AR controls(32.4%±6.0% vs 28.5%±2.4%; P<0.001 9). Global native T1 is higher in AR group than that in non-AR group(49.8±3.1 vs 47.5±2.8 ms, P=0.009)and the difference is statistically significant.The cutoff value of global ECV is 30.62% with a sensitivity of 61% and a specificity of 86% for detecting AR.And T2 mapping reveale that T2 value of global left ventricle is significantly higher in AR group than that in non-AR group(49.8±3.1 vs 47.5±2.8 ms, P=0.009). LGE extent is significantly higher in AR group than those in non-AR group( P=0.004). Through including global native T1 and ECV into a logistic regression model, multiparametric CMR can yield an area under curve(AUC)of 0.794.It hints at the potential of CMR for detecting AR. Conclusions:Multiparametric cardiac magnetic resonance offers an excellent predictive capacity for a noninvasive detection of AR.