This paper proposed a new approach to attach power devices on nonmetallized direct bonding copper (DBC) substrates by pressureless sintering of nanosilver paste in poor-oxygen sintering atmosphere and formic acid reduction atmosphere. The average shear strength of die-attach joints could reach 25 MPa, and the copper oxidation issue of the nonmetallized DBC substrates was avoided. Based on the pressureless sintering approach of nanosilver paste, insulated-gate bipolar transistor (IGBT) modules were prototyped to verify the feasibility of this approach for mass production of power modules. The electrical and thermal characteristics of the IGBT modules bonded with sintered nanosilver were then compared with those of the commercialized ones using Pb92.5Sn5Ag2.5 solder alloy. This approach could extend the applications to bonding power devices on nonmetallized DBC substrates by nanosilver paste, and guide fabricating power modules in a mass productive, low facility costs, and high-yield way.
Objectives The aim of this study was to compare the characteristics derived from fluorine-18 fluorodeoxyglucose (18F-FDG) and fluorine-18 fluorothymidine (18F-FLT) PET quantitatively and to assess their capacities during staging of esophageal squamous cell carcinoma (ESCC). Materials and methods Thirty-six patients with a diagnosis of ESCC who underwent both 18F-FDG and 18F-FLT PET were included in this study. Different image-derived indices including the standardized uptake value (SUV), gross tumor length, and texture features were determined. Considering histopathologic examination as the gold standard, the performance of the extracted indices during staging of ESCC was assessed using the Kruskal–Wallis test and the Mann–Whitney test. Results Considering the 18F-FDG PET images, the SUVmax, SUVmean, length (LEN), and eccentricity (EC) were significant during staging of American Joint Committee on Cancer (AJCC) and TNM (P<0.01), whereas for the 18F-FLT image, the SUVmax, LEN, and EC were significant during staging of AJCC and TNM (P<0.01). The characteristics of 18F-FDG and 18F-FLT PET for the classification of ESCC stage were significantly different. Conclusion 18F-FDG image-derived characteristics including image textural features, SUV, and shape feature allow for better stratification of AJCC and TNM than 18F-FLT PET in ESCC patients.
As a representative collaborative filtering method, matrix factorization has been widely used in personalized recommendation. Recently, deep matrix factorization model, which utilizes deep neural networks to project users and items into a latent structured space, has received increased attention. In this paper, inspired by the idea of BiasedSVD that introduces bias to both users and items, we propose a novel matrix factorization model with neural network architecture, named BDMF, short for Biased Deep Matrix Factorization. Specifically, we first construct a user-item interaction matrix with explicit ratings and implicit feedback, and randomly sample users and items as the input. Next, we feed this input to the proposed BDMF model to learn latent factors of both users and items, and then use them to predict the ratings for personalized ranking. We also formally show that BDMF works on the same principle as BiasedSVD, which means that BDMF can be viewed as a deep neural network implementation of BiasedSVD. Finally, extensive experiments on real-world datasets are conducted and the results verify the superiority of our model over other state-of-the-art.
Purpose: To compare the dosimetric difference of the target volume and organs at risk(OARs) between conventional intensity‐modulated radiotherapy(C‐IMRT) and knowledge‐based radiation therapy (KBRT) plans for cervix cancer. Methods: 39 patients with cervical cancer after surgery were randomly selected, 20 patient plans were used to create the model, the other 19 cases used for comparative evaluation. All plans were designed in Eclipse system. The prescription dose was 30.6Gy, 17 fractions, OARs dose satisfied to the clinical requirement. A paired t test was used to evaluate the differences of dose‐volume histograms (DVH). Results: Comparaed to C‐IMRT plan, the KBRT plan target can achieve the similar target dose coverage, D98,D95,D2,HI and CI had no difference (P≥0.05). The dose of rectum, bladder and femoral heads had no significant differences(P≥0.05). The time was used to design treatment plan was significant reduced. Conclusion: This study shows that postoperative radiotherapy of cervical KBRT plans can achieve the similar target and OARs dose, but the shorter designing time.
Purpose: To evaluate patients Dose Volume Histogram (DVH) parameters in cummulative healthy liver dose using four-dimesional computed tomography (4DCT) image and active breathing control (ABC) manner. Methods: Ten liver cancer patients were analysed retrospectively. The static plan was designed on reference CT image from the free breathing status scanning and the organ at risk(OAR) dose were evaluated.The tracking accumulative dose were calculated on 10 different breathing phases of 4DCT based on relative time weight. The mean healthy liver dose were calculated in different exhale and inhale breath hold using ABC technology. Three motion management strategies plans were compared and analysed. Results: The maxinum difference of mean healthy liver dose in 4DCT image was 9.5% between phase 10 and phase 60. The largest absolute dose in mean healthy liver was 5.05Gy between the tracking dose and the deep inhale breath hold(P﹤0.05). The difference using ABC between the deep inhale and exhale breah hold was maximum 1.45Gy and no significance was obsered between the calm inhale and exhale breah hold. Also there was no signifcance between the target tracking and any breath phase in 4DCT image. Conclusions: The target tracking dose was the actual dilevered dose to the patient. The difference in breath hold and the target tracking dose was significantly. Therefore, We suggest the deep inhale breath hold was used in the liver radiotherapy treatment using ABC technology. However, we needed more patients to further study to get the more accurate result.
Purpose: A method using four‐dimensional(4D) PET/CT in design of radiation treatment planning was proposed and the target volume and radiation dose distribution changes relative to standard three‐dimensional (3D) PET/CT were examined. Methods: A target deformable registration method was used by which the whole patient's respiration process was considered and the effect of respiration motion was minimized when designing radiotherapy planning. The gross tumor volume of a non‐small‐cell lung cancer was contoured on the 4D FDG‐PET/CT and 3D PET/CT scans by use of two different techniques: manual contouring by an experienced radiation oncologist using a predetermined protocol; another technique using a constant threshold of standardized uptake value (SUV) greater than 2.5. The target volume and radiotherapy dose distribution between VOL3D and VOL4D were analyzed. Results: For all phases, the average automatic and manually GTV volume was 18.61 cm3 (range, 16.39–22.03 cm3) and 31.29 cm3 (range, 30.11–35.55 cm3), respectively. The automatic and manually volume of merged IGTV were 27.82 cm3 and 49.37 cm3, respectively. For the manual contour, compared to 3D plan the mean dose for the left, right, and total lung of 4D plan have an average decrease 21.55%, 15.17% and 15.86%, respectively. The maximum dose of spinal cord has an average decrease 2.35%. For the automatic contour, the mean dose for the left, right, and total lung have an average decrease 23.48%, 16.84% and 17.44%, respectively. The maximum dose of spinal cord has an average decrease 1.68%. Conclusion: In comparison to 3D PET/CT, 4D PET/CT may better define the extent of moving tumors and reduce the contouring tumor volume thereby optimize radiation treatment planning for lung tumors.
Objective
To investigate the feasibility of defining the radiotherapy target of primary liver cancer using four-dimensional computed tomography (4DCT) and T2-weighted magnetic resonance (MR-T2) deformable image registration.
Methods
Ten patients with hepatocellular carcinoma (HCC) who first received radiotherapy were included in this study. The 4DCT in free breathing and MR-T2 in deep breathing were acquired sequentially. 4DCT were sorted into ten series of CT images according to the respiratory phase. MIM software was used for deformable image registration. The accuracy of deformable image registration was assessed by the maximal displacements in three-dimensional directions of the portal vein and the celiac trunk and the degree of liver overlapping (P-LIVER). Gross tumor volume (GTV) was delineated on different series of CT images and the internal GTV (IGTV) was merged by ten GTVs on 4DCT images in each phase. The MR-T2 image was deformably registered to 4DCT images in each phase to acquire ten GTVDR. The IGTVDR was obtained by merging the ten GTVDR. The differences between different target volumes were compared by paired t-test.
Results
The maximal displacements in three-dimensional directions of the portal vein were 0.3±0.8 mm along the x-axis, 0.8±1.8 mm along the y-axis, and 0.5±1.5 mm along the z-axis. The maximal displacements in three-dimensional directions of the celiac trunk were 0.1±1.0 mm along the x-axis, 0.7±1.2 mm along the y-axis, and 0.6±2.0 mm along the z-axis. Overlapping degree was 115.4±13.8%. The volumes of GTVs obtained from 4DCT images in each phase after DR increased by an average of 8.18%(P<0.05), and were consistent with those delineated on MR-T2 images. The IGTV after DR increased by an average of 9.67%(P<0.05).
Conclusions
MRI image can show more information and have a higher contrast than CT image. MRI images should be combined with 4DCT images for delineating the GTV. It can better determine the scope and trajectory of the target and improve the delineation accuracy of HCC target.
Key words:
Magnetic resonance imaging; Tomography, X-ray computed, four-dimensional; Deformable registration; Liver neoplasms/radiotherapy
Abstract Er-doped ZnO thin films on a SiO 2 /Si substrate were fabricated by radio frequency magnetron sputtering, in which embedded Si nanoparticles (NPs) were formed by ion implantation and subsequent thermal annealing. The effects of Si NPs on the Er photoluminescence (PL) at 1.54 μ m were investigated. In addition to the typical emission at 1.54 μ m from Er 3+ , a new 1.16- μ m emission peak was also observed after a thermal treatment. Further annealing resulted in shift of emission intensity between the 1.16- and 1.54- μ m luminescence features. The observed Si nanoparticles (NPs) were ∼4 nm in diameter. The formation of new components Zn 2 SiO 4 and Er 2 Si 2 O 7 was also presented in this study. The 1.16- μ m luminescence is attributed to the Si NPs, and the suppression of Si NPs related emission is caused by consumption of Si in the formation of Er silicate and zinc silicide and the energy transfer between Si NPs and Er 3+ . The intensity of Er 3+ related 1.54- μ m PL can be modulated by the Si NPs fabricated by implantation and optimizing the annealing condition.
Abstract Background: This study is to distinguish peripheral lung cancer and pulmonary inflammatory pseudotumor using CT-radiomics features extracted from PET/CT images. Methods: In this study, the standard 18F-fluorodeoxyglucose positron emission tomography/ computed tomography (18 F-FDG PET/CT) images of 21 patients with pulmonary inflammatory pseudotumor (PIPT) and 21 patients with peripheral lung cancer were retrospectively collected. The dataset was used to extract CT-radiomics features from regions of interest (ROI), The intra-class correlation coefficient (ICC) was used to screen the robust feature from all the radiomic features. Using, then, statistical methods to screen CT-radiomics features, which could distinguish peripheral lung cancer and PIPT. And the ability of radiomics features distinguished peripheral lung cancer and PIPT was estimated by receiver operating characteristic (ROC) curve and compared by the Delong test. Results: A total of 435 radiomics features were extracted, of which 361 features showed relatively good repeatability (ICC³0.6). 20 features showed the ability to distinguish peripheral lung cancer from PIPT. these features were seen in 14 of 330 Gray-Level Co-occurrence Matrix features, 1 of 49 Intensity Histogram features, 5 of 18 Shape features. The area under the curves(AUC) of these features were 0.731 0.075, 0.717, 0.748 0.038, respectively. The P values of statistical differences among ROC were 0.0499 (F9, F20), 0.0472 (F10, F11) and 0.0145 (F11, Mean4). The discrimination ability of forming new features (Parent Features) after averaging the features extracted at different angles and distances was moderate compared to the previous features(Child features). Conclusion: Radiomics features extracted from non-contrast CT based on PET/CT images can help distinguish peripheral lung cancer and PIPT.
Background: To study the feasibility of defining the individual internal gross tumor volume (IGTV) of hepatocellular carcinoma (HCC) using four-dimensional computed tomography (4DCT) imaging and T2-weighted magnetic resonance imaging (T2-weighted MRI) by deformable registration (DR).
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)