MSA-GCN: A Multi-information Selection Aggregation Graph Convolutional Network for Breast Tumor Grading.

Physicians typically combine multi-modal data to make a graded diagnosis of breast tumors. However, most existing breast tumor grading methods rely solely on image information, resulting in limited accuracy in grading. This paper proposes a Multi-information Selection Aggregation Graph Convolutional Networks (MSA-GCN) for breast tumor grading. Firstly, to fully utilize phenotypic data reflecting the clinical and pathological characteristics of tumors, an automatic combination screening and weight encoder is proposed for phenotypic data, which can construct a population graph with improved structural information. Then, a graph structure is designed through similarity learning to reflect the correlation between patient image features. Finally, a multi-information selection aggregation mechanism is employed in the graph convolution model to extract the effective features of multi-modal data and enhance the classification performance of the model. The proposed method is evaluated on different clinical datasets from the Digital Database for Screening Mammography (DDSM) and INbreast. The average classification accuracies are 90.74% and 85.35%, respectively, surpassing the performance of existing methods. In conclusion, our method effectively fuses image and non-image information, leading to a significant improvement in the accuracy of breast tumor grading.

Electrocardiographic localization of peripherally inserted central catheter tip position in critically ill patients with advanced cancer: An application study.

BACKGROUND: We compared the methods of electrocardiogram (ECG) and X-ray localization of the peripherally inserted central catheter (PICC) tip position, in order to find a more convenient, practical, and safe method. OBJECTIVE: To investigate the value of applying electrocardiographic localization of the PICC tip position in critically ill patients with advanced cancer in Hebei Province, China. METHOD: Enrolled 137 advanced cancers requiring PICC placement. The position of the catheter tip was localized with the bedside electrocardiogram in real time. Then, the localization was performed using a chest X-ray (the gold standard). The accuracy of electrocardiographic location was checked. RESULTS: Specific P waves were observed in 130 patients. No change in the P waves was observed for the remaining seven patients. The age of the latter group of patients was more advanced (87.29 [5.15] years), a significant difference to that of the 130 patients with specific P waves (71.58 [14.84] years) (t = -6.704, p < .001). Specific P waves not only involve ascendance in P waves but also ascendance in QRS waves. CONCLUSIONS: The use of an ECG to localize the PICC tip in critically ill patients with advanced cancer may replace the unnecessary use of chest X-rays. Specific P waves not only involve an increase in P waves but also an increase in QRS waves. If there is no change in the P wave, a chest X-ray film must be obtained. In elderly patients, because there is a possibility of catheter tip malposition, a comprehensive evaluation should be performed before surgery.

Adaptive Ultrasound Tissue Harmonic Imaging Based on an Improved Ensemble Empirical Mode Decomposition Algorithm.

Complete and accurate separation of harmonic components from the ultrasonic radio frequency (RF) echo signals is essential to improve the quality of harmonic imaging. There are limitations in the existing two commonly used separation methods, that is, the subjectivity for the high-pass filtering (S_HPF) method and motion artifacts for the pulse inversion (S_PI) method. A novel separation method called S_CEEMDAN, based on the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm, is proposed to adaptively separate the second harmonic components for ultrasound tissue harmonic imaging. First, the ensemble size of the CEEMDAN algorithm is calculated adaptively according to the standard deviation of the added white noise. A set of intrinsic mode functions (IMFs) is then obtained by the CEEMDAN algorithm from the ultrasonic RF echo signals. According to the IMF spectra, the IMFs that contain both fundamental and harmonic components are further decomposed. The separation process is performed until all the obtained IMFs have been divided into either fundamental or harmonic categories. Finally, the fundamental and harmonic RF echo signals are obtained from the accumulations of signals from these two categories, respectively. In simulation experiments based on CREANUIS, the S_CEEMDAN-based results are similar to the S_HPF-based results, but better than the S_PI-based results. For the dynamic carotid artery measurements, the contrasts, contrast-to-noise ratios (CNRs), and tissue-to-clutter ratios (TCRs) of the harmonic images based on the S_CEEMDAN are averagely increased by 31.43% and 50.82%, 18.96% and 10.83%, as well as 34.23% and 44.18%, respectively, compared with those based on the S_HPF and S_PI methods. In conclusion, the S_CEEMDAN method provides improved harmonic images owing to its good adaptivity and lower motion artifacts, and is thus a potential alternative to the current methods for ultrasonic harmonic imaging.

A dynamic ultrasound simulation of a pulsating three-layered CCA for validation of two-dimensional wall motion and blood velocity estimation algorithms.

PURPOSE: A dynamic ultrasound simulation model for the common carotid artery (CCA) with three arterial layers for validation of two-dimensional wall motion and blood velocity estimation algorithms is proposed in the present study. This model describes layers with not only characteristics of echo distributions conforming to clinical ones but also varying thicknesses, axial, and radial displacements with pulsatile blood pressure during a cardiac cycle. METHODS: The modeling process is as follows: first, a geometrical model according with the clinical structure size of a CCA is built based on the preset layer thicknesses and the diameter of lumen. Second, a three-dimensional scatterer model is constructed by a mapping with a Hilbert space-filling curve from the one-dimensional scatterer distribution with the position and amplitude following Gamma and Gaussian distributions, respectively. The characteristics of three layers and blood are depicted by smoothly adjusting the scatterer density, the scale, and shape parameters of the Gamma distribution as well as the mean and standard deviation of the Gaussian distribution. To obtain the values of parameters of scatterer distributions, including the shape parameter, density, and intensity, for arterial layers and blood, the envelope signals simulated from different configurations of scatterer distribution are compared with those from different kinds of tissue of CCAs in vivo through a statistic analysis. Finally, the dynamic scatterer model is realized based on the blood pressure, elasticity modulus of intima-media (IM) and adventitia, varying IM thickness, axial displacement of IM as well as blood flow velocity at central axis during a cardiac cycle. Then, the corresponding radiofrequency (RF) signals, envelope signals, and B-mode images of the pulsatile CCA are generated in a dynamic scanning mode using Field II platform. RESULTS: The three arterial layers, blood, and surrounding tissue in simulated B-mode ultrasound images are clearly legible. The results based on a statistical analysis for the envelope signals from 30 simulations indicate that the echo characteristics of blood, intima, media, and adventitia are in accordant with clinical ones. The maximum relative errors between the preset geometrical sizes and the measured ones from the simulated images for the diameter of the lumen and the thicknesses of the intima, media, and adventitia are 0.13%, 3.89%, 1.35%, and 0.06%, respectively. For the dynamic parameters, the variation in IM thickness, the radial displacements of lumen and adventitia as well as the axial displacement of IM and blood flow velocity are measured with the mean relative errors of 68.03%, 9.27%, 2.10%, 4.93%, and 17.34%, respectively. CONCLUSION: The simulated results present static sizes and dynamical variations according with preset values; echo distributions conforming to clinical versions. Therefore, the presented simulation model could be useful as a data source to evaluate the performance of studies on measurements of ultrasound-based tissue structures and dynamic parameters for the CCA layers.

Preparation and application of micro/nanoparticles based on natural polysaccharides.

Polysaccharides have attracted more and more attentions and been recognized to be the most promising materials in recent years because of their outstanding merits such as easily available, non-toxic, biocompatible, biodegradable, and easily modified. Considerable research efforts have been directed toward developing polysaccharides-based micro/nanoparticles (PM/NPs). The new major studies of PM/NPs over the past few years are outlined in this review. Methods of preparation, including self-assembly, ionic-gelation, complex coacervation, emulsification, and desolvation method and some others, are summarized. Different applications of PM/NPs in the field of drug-delivery system are highlighted. Besides, another novel application of PM/NPs that are used as emulsifiers to stabilize Pickering emulsion is also introduced. These environmental-friendly particle emulsifiers have received reasonable attention due to their novel applications, especially in food, cosmetics, and pharmaceutics. From literature surveys, we realized that studies on PM/NP systems for different applications have increased rapidly. Hence, the present review is timely.