Notes

Interactions in between alcohol consumption along with coronary disease between long-term children of digestive tract cancers: a new population-based, retrospective cohort examine.
The results showed that DualRank outperformed competing methods and could identify biomarkers with the small quantity, great prediction performance (average AUC=0.818) and biological interpretability.Precise cancer subtype and/or stage prediction is instrumental for cancer diagnosis, treatment and management. However, most of the existing methods based on genomic profiles suffer from issues such as overfitting, high computational complexity and selected features (i.e., genes) not directly related to forecast precision. These deficiencies are largely due to the nature of "high-dimensionality-small-sample (HDSS)" inherent in molecular data, and such a nature is often deemed as an obstacle to the application of deep learning to biomedical research. In this paper, we propose a DNN-based algorithm coupled with a new embedded feature selection technique, named Dropfeature-DNNs, to address these issues. We formulate Dropfeature-DNNs as an iterative AUC optimization problem when training DNNs. As such, an "optimal" feature subset that contains meaningful genes for patient stratification can be obtained when the AUC optimization converges. Since the feature subset and AUC optimizations are synchronous with the training of DNNs, model complexity and computational cost are simultaneously reduced. Rigorous feature subset convergence analysis and error bound inference provide a solid theoretical foundation for the proposed method. Extensive empirical comparisons to benchmark methods further demonstrate the efficacy of Dropfeature-DNNs in cancer subtype and/or stage prediction using HDSS gene expression data from multiple cancer types.DNA strand displacement is introduced in this study and used to construct an analog DNA strand displacement chaotic system based on six reaction modules in nanoscale size. The DNA strand displacement circuit is employed in encryption as a chaotic generator to produce chaotic sequences. In the encryption algorithm, we convert chaotic sequences to binary ones by comparing the concentration of signal DNA strand. Simulation results show that the encryption scheme is sensitive to the keys, and key space is large enough to resist the brute-force attacks, furthermore algorithm has a high capacity to resist statistic attack. Based on robustness analysis, our proposed encryption scheme is robust to the DNA strand displacement reaction rate control, noise and concentration detection to a certain extent.Customized static orthoses in rehabilitation clinics often cause side effects, such as discomfort and skin damage due to excessive local contact pressure. Currently, clinicians adjust orthoses to reduce high contact pressure based on subjective feedback from patients. However, the adjustment is inefficient and prone to variability due to the unknown contact pressure distribution as well as differences in discomfort due to pressure across patients. This paper proposed a new method to predict a threshold of contact pressure (pressure limit) associated with moderate discomfort at each critical spot under hand orthoses. A new pressure sensor skin with 13 sensing units was configured from FEA results of pressure distribution simulated with hand geometry data of six healthy participants. It was used to measure contact pressure under two types of customized orthoses for 40 patients with bone fractures. Their subjective perception of discomfort was also measured using a 6 scores discomfort scale. Based on these data, five critical spots were identified that correspond to high discomfort scores (>1) or high pressure magnitudes (>0.024 MPa). An artificial neural network was trained to predict contact pressure at each critical spot with orthosis type, gender, height, weight, discomfort scores and pressure measurements as input variables. The neural networks show satisfactory prediction accuracy with R2 values over 0.81 of regression between network outputs and measurements. This new method predicts a set of pressure limits at critical locations under the orthosis that the clinicians can use to make orthosis adjustment decisions.Multi-contrast magnetic resonance (MR) image registration is useful in the clinic to achieve fast and accurate imaging-based disease diagnosis and treatment planning. selleck inhibitor Nevertheless, the efficiency and performance of the existing registration algorithms can still be improved. In this paper, we propose a novel unsupervised learning-based framework to achieve accurate and efficient multi-contrast MR image registrations. Specifically, an end-to-end coarse-to-fine network architecture consisting of affine and deformable transformations is designed to improve the robustness and achieve end-to-end registration. Furthermore, a dual consistency constraint and a new prior knowledge-based loss function are developed to enhance the registration performances. The proposed method has been evaluated on a clinical dataset containing 555 cases, and encouraging performances have been achieved. Compared to the commonly utilized registration methods, including VoxelMorph, SyN, and LT-Net, the proposed method achieves better registration performance with a Dice score of 0.8397±0.0756 in identifying stroke lesions. With regards to the registration speed, our method is about 10 times faster than the most competitive method of SyN (Affine) when testing on a CPU. Moreover, we prove that our method can still perform well on more challenging tasks with lacking scanning information data, showing the high robustness for the clinical application.Despite the successes of deep neural networks on many challenging vision tasks, they often fail to generalize to new test domains that are not distributed identically to the training data. The domain adaptation becomes more challenging for cross-modality medical data with a notable domain shift. Given that specific annotated imaging modalities may not be accessible nor complete. Our proposed solution is based on the cross-modality synthesis of medical images to reduce the costly annotation burden by radiologists and bridge the domain gap in radiological images. We present a novel approach for image-to-image translation in medical images, capable of supervised or unsupervised (unpaired image data) setups. Built upon adversarial training, we propose a learnable self-attentive spatial normalization of the deep convolutional generator network's intermediate activations. Unlike previous attention-based image-to-image translation approaches, which are either domain-specific or require distortion of the source domain's structures, we unearth the importance of the auxiliary semantic information to handle the geometric changes and preserve anatomical structures during image translation.
My Website: https://www.selleckchem.com/products/fht-1015.html