ABSTRACT
Aiming at the problems of missing important features, inconspicuous details and unclear textures in the fusion of multimodal medical images, this paper proposes a method of computed tomography (CT) image and magnetic resonance imaging (MRI) image fusion using generative adversarial network (GAN) and convolutional neural network (CNN) under image enhancement. The generator aimed at high-frequency feature images and used double discriminators to target the fusion images after inverse transform; Then high-frequency feature images were fused by trained GAN model, and low-frequency feature images were fused by CNN pre-training model based on transfer learning. Experimental results showed that, compared with the current advanced fusion algorithm, the proposed method had more abundant texture details and clearer contour edge information in subjective representation. In the evaluation of objective indicators, Q AB/F, information entropy (IE), spatial frequency (SF), structural similarity (SSIM), mutual information (MI) and visual information fidelity for fusion (VIFF) were 2.0%, 6.3%, 7.0%, 5.5%, 9.0% and 3.3% higher than the best test results, respectively. The fused image can be effectively applied to medical diagnosis to further improve the diagnostic efficiency.
Subject(s)
Image Processing, Computer-Assisted/methods , Neural Networks, Computer , Tomography, X-Ray Computed , Magnetic Resonance Imaging/methods , AlgorithmsABSTRACT
Objective Early evaluate the feasibility and reproducibility of sorafenib-targeted therapy for hepatocellular carcinoma by RECIST1.1, mRECIST and three-dimensional volume measurement. Methods Seventy patients with pathology or typical imaging findings confirmed as hepatocellular carcinoma along with the sorafenib-targeted treatment for more than 2 months between October 2004 to April 2017 in the Fifth Affiliated Hospital of Sun Yat-sen University were retrospectively analyzed. Patients underwent chest, abdominal and pelvic CT scans and enhanced scans before and after 2 weeks of sorafenib treatment. Two physicians used RECIST 1.1, mRECIST, and volume measurement criteria to evaluate the efficacy of treatment. According to their averaged results, the patients were divided into two groups (control group and non-control group). Kaplan-Meier survival analysis was used to compare the prognostic values between different response evaluation criterias for early predicting the efficacy of sorafenib-targeted therapy in advanced hepatocellular carcinoma. Kappa test was used to assess the efficacy response consistency in intra-group and inter-group. Results Based on mRECIST and RECIST 1.1 measurements, the control group included 34 cases, and the non-control group included 36 cases. Based on semi-automatic volume measurement, the control group included 38 cases, and the non-control group included 32 cases. Before the treatment with sorafenib, the RECIST 1.1 and mRECIST methods were used. There was a high degree of consistency between the two doctors (Kappa values were 0.79 and 0.71, respectively), and the semi-automatic volume measurement method was extremely consistent (Kappa value was 0.90); the consistency in intra-observer by three different methods was extremely high (Kappa values were 0.91, 0.85, 0.97, respectively). After the treatment with sorafenib, the consistency between the two radiologists using RECIST 1.1 measurement was high (Kappa value was 0.65), the consistency of mRECIST measurement was moderate (Kappa value was 0.52), and the consistency of tumor volume measurement was extremely high (Kappa The value was 0.83), the consistency in intra-observer using the above three methods was high or very high (Kappa values were 0.86, 0.74, 0.90, respectively). The RECIST 1.1 and mRECIST measurements were less sensitive in early evaluation of sorafenib-targeted treatment, and there was no significant difference between the control group and the non-control group (P=0.578 and 0.613) while the semi-automatic volumetric measurement was sensitive (P=0.004). Conclusion Semi-automated three-dimensional volume measurement which has better intra-and inter-group consistency and reproducibility can reflect the efficacy of sorafenib-targeted therapy for hepatocellular carcinoma in early stage.
ABSTRACT
Objective To investigate the value of renal CT volumetric texture analysis with machine learning radiomics in assessment of pathological grade of clear cell renal cell carcinoma(ccRCC). Methods Thirty-four biopsy-confirmed ccRCC subjects who had four-phase CT scanning (NC:non-contrast, CM: Corticomedullary, N: Nephrographic, E: Excretory) were collected retrospectively from June 2013 to October 2017 for the study.Non-rigid registration was performed on multi-phase CT images in reference to CM-phase.Each lesion was segmented on CM-phase CT images using our in-house volumetric image analysis platform,"3DQI".A set of fifty-nine volumetric textures,including histogram,gradient,gray level co-occurrence matrix(GLCM),run-length(RL),moments,and shape,was calculated for each segment lesion in each phase as parameters for the training/testing of Random Forest (RF) classifier. Four groups according to pathological Fuhrman grade on a scaleⅠtoⅣ,these tumors were then divided into low(Ⅰ+Ⅱ) and high grade ( Ⅲ + Ⅳ) groups. Feature selection was performed by Boruta algorithm. A 10-fold cross-validation method was applied to validate the RF performance by receiver operating characteristic (ROC) curves analysis to determine the diagnostic accuracy of the model. Results Subjects were divided into four groups by Fuhrman grade on a scaleⅠtoⅣ:3 cases gradeⅠ,19 cases gradeⅡ,8 cases gradeⅢand 4 cases gradeⅣ.In CM-phase,kurtosis and long-run-emphasis(RLE)were selected the most important textures for ccRCC staging among 59 features. The area under curve (AUC) of ROC was 0.88 (79% sensitivity and 82% specificity)by using kurtosis and RLE textures.The mean values of kurtosis and RLE were(-20.00±22.00)×10-2and(3.00±0.40)×10-2for low group,whereas(31.00±32.00)×10-2and(5.00± 0.02)×10-2for high group.Within the mean±SD range of statistics,radiomics can distinguish between low and high grade tumors.In multi-phase analysis,three most important features were selected among 236(59× 4) textures: kurtosis (CM-phase), GLCM homogeneity I (HOMO 1) (E-phase), and GLCM homogeneity 2 (HOMO2)(E-phase).The mean values of HOMO 1(E-phase)and HOMO 2(E-phase)were(19.00±0.03)× 10-2and(11.00±0.02)×10-2for low group,whereas(22.00±0.03)×10-2and(14.00±0.02)×10-2for high group. The AUC was 0.92(93% sensitivity and 87% specificity)by using these three textures. Conclusion This study has demonstrated that renal CT volumetric texture analysis with machine learning radiomics could preoperative accurately perform cancer staging for ccRCC.
ABSTRACT
Objective To study the feasibility of applying electronic cleaning to intestinal contents tagging by diatrizoate meglumine for single-source dual-energy CT colonography with sequential acquisitions and volume scanning.Methods Twenty-four volunteers had fine effect of intestinal contents tagging by diatrizoate meglumine,good colorectal distension effect,fine image quality of dual-energy fusion colorectal images,and with informed consents were enrolled in this study.The single-source dual-energy CT colonography with sequential acquisitions and volume scanning was performed with an Acquilion ONE 320 row CT scanner,tube voltage 135 kVp/80 kVp.The intestinal contents conducted the dual-energy electronic cleaned based on decomposition of intestinal contents tagging by diatrizoate meglumine,soft tissue and air.The intestinal contents in one segment of intestinal lumen being 100% electronically cleaned served as the basic standard,the electronic cleaning effects were divided into the 5 grades:excellent,good,moderate,fair and poor;and grade 1-3 were effective fecal electronic cleaning.Results The grade 1,2,3,4,5 of electronic cleaning effect for solid as the main intestinal contents were 22.2%,53.3%,17.8%,6.7% and 0% respectively;and which of electronic cleaning effect for liquid as the main intestinal contents were 47.5%,47.5%,5.0%,0% and 0% respectively.The together total effective electronic cleaning of intestinal contents was 97.9% and the electronic cleaning effect was good.Conclusion Electronic cleaning could be used in the intestinal contents tagging by diatrizoate meglumine for single-source dual-energy CT colonography with sequential acquisitions and volume scanning.
ABSTRACT
Taking autoimmune inflammation of rheumatoid arthritis as entry point, this paper discussed the clinical effect of horsetail mixture on rheumatoid arthritis [RA] and its mechanism. A total of 60 cases of patients with RA were randomly divided into experimental group and control group using randomized controlled trial. We observed its biochemistry, TNF- alpha and IL-10 before and after treatment, and then systematically assessed the clinical effect of horsetail on RA. Results showed that the total effective rate of experimental group was 80%, while that of control group was 16.67%. After statistical treatment, the differences between two groups were significant [p>0.05]. Comparison of the difference value of TNF-alpha [p<0.05] and IL-0.05 in serum between groups before and after treatment, there were significant differences. Comparison of CRP within group before and after treatment was significant different [p<0.05], while comparison of CRP between groups was not significantly different [p<0.05]. Comparison of ESR and RF within group before and after treatment was significantly different [p<0.01], and comparison of them between groups was also significantly different [p<0.05]. Comparison of difference values within group before and after treatment were also significantly different [p<0.01]. It was concluded that horsetail mixture has remarkable curative effect on rheumatoid arthritis, and its clinical application is safe and reliable. It has obvious down regulatory effect on cell factor TNF- alpha related to RA, that is, it can down regulate the level of pre-inflammatory factor TNF- alpha as well as the level of anti-inflammatory factor IL-10. Therefore, it is considered that the regulating effect of horsetail mixture on TNF- alpha and IL-10 is one of the mechanisms of its treatment on RA
ABSTRACT
Objective To analyze the diagnosis and treatment of iliac pseudoaneurysms following renal transplantation.Method The data of two patients with pseudoaneurysm who underwent kidney transplantation were retrospective analysied.Result One case with pseudoaneurysm received transplant nephrectomy,and pathological examination diagnosed mucoraceae infection; the other patient received endovascular treatment and amphotericin B therapy,endly died of hemorrhagic shock.Conclusion The development of pseudoaneurysms at the kidney transplantation recipients with the same donor results in high rates of mucormycosis.They should take anti-fungus therapy and operation as early as possible.
ABSTRACT
Objective To observe the effect of hybrid renal replacement therapy (sustained lowefficiency hemodiafiltration) in recipients with delayed graft function (DGF) after kidney transplantation.Methods In 33 kidney transplant patients with DGF,there were 15 cases subject to sustained low-efficiency hemodiafiltration (SLEDF group),18 cases subject to hemodialysis (HD group).The renal function was tested,and the expression levels of complements C3,C4,and CRP,IL-1β,IL-6,TNF-α were detected before and after therapy.Results In SLEDF group,the renal function recovered in (29 ± 13) days after therapy,and in (47 ± 21) days in HD group.After therapy,the BUN and Scr in both two groups were lower than before (P<0.01).After therapy,the levels of IL-1β,IL-6 and TNF-α were also lower than before in two groups,and the levels were lower in SLEDF group than in HD group (P<0.05).The levels of complements C3,C4,and CRP in two groups were decreased after therapy,but there was no statistically significant difference (P>0.05).Conclusion SLEDF can decrease the level of SCr,BUN and some cytokines in recipients effectively,and It's helpful to recovery of the renal function.SLEDF can be used by DGF recipients in transit time.
ABSTRACT
BACKGROUND: Presently, acute rejection following renal transplantation remains a risk factor for chronic rejection and graft function injury, How to non-invasive, rapid and exact diagnosis and prompt treatment is important. OBJECTIVE: To investigate early diagnosis and post-treatment expression of urine monocyte chemoattractant protein-1 (MCP-1) in the acute rejection after renal transplantation, through detecting the association of the urine MCP-1 variation according to some cases of nephridial tissue biopsy. METHODS: We selected 62 chronic renal failure patients who received renal homotransplantations in the Department of Renal Transplantation of Zhengzhou People's Hospital from October 2008 to February 2009. The stable renal function group contained 42 patients with stable renal function following renal transplantation. Acute rejection group contained 20 patients with acute rejection following renal transplantation. We chose 10 patients who examined no abnormalities in the Medical Examination Center of Zhengzhou People's Hospital to detect their urine sample as control group. All patients following renal transplantation underwent conventional immunosuppression. In addition, patients in the acute rejection group were treated with antilymphocyte globulin or methylprednisolone reinforced impact therapy. MCP-1 mass concentration changes were measured by double antibodies sandwich enzyme linked immurosorbent assay. RESULTS AND CONCLUSION: Compared with control group, no significant change was determined in urine MCP-1 mass concentration in the stable renal function group (P > 0.05). The urine MCP-1 mass concentration was significantly increased in the acute rejection group (P< 0.01). Compared with pretreatment, urine MCP-1 mass concentration was significantly decreased following treatment in 20 patients from the acute rejection group (P < 0.01). Of them, 17 cases had relieved clinical symptom, and normal auxiliary examination, and their urine MCP-1 mass concentration was close to the control group; 3 cases were inefficient, whose urine MCP-1 mass concentration was greater than the control group. Eight cases received nephridial tissue biopsy, and kidney pathology demonstrated acute rejection of transplanted kidney, which was similar to urine MCP-1 mass concentration in the acute rejection group prior to treatment (P > 0.05). These indicated that the level of MCP-1 in urine can non-invasively diagnose acute rejection following renal transplantation in an early phase, and monitor therapeutic efficacy. This may be associated with renal pathological injury during acute rejection following renal transplantation.
ABSTRACT
Objective To observe the effects of turicamycin-inhibitor of N-glycosylation of proteins on expression of (?1 integrin,FAK and cyclin D1 in high glucose-induced glomerular mesangial cells (CMC), and to explore the mechanism concerned. Methods Cultured HBZY-1 rat mesangial cells were divided into 5 groups:control group;high glucose group;mannitol group;high glucose plus TM group; TM group.The expression of ?1 integrin was measured by flow cytometry, expression of FAK and cyclin D1 was measured by immunohistochemistry,and proliferation of GMCs was measured by MTT. Results There was a little expression of ?1 integrin, FAK and cyclin D1 on normal mesangial cells. High glucose induced the proliferation and increased the expression of ?1 integrin and cyclin D1.There was no significant difference in mannitol group as compared to control group.The expression of (?1 integrin FAK and cyclin D1 decreased notably by TM.TM could also decrease proliferative abilitiy of cells. All the effects of TM were dose-dependent. Conclusion Through blocking glycosylation of glycoprotein, TM can suppress cellular proliferation and expression of pi integrin induced by high glucose in a dose-dependent manners,then pi integrin affects the expression of FAK and cyclin D1.