The aberrant level expression of miRNAs is closely related to endometrial cancer progression as well as prognosis, and the survival prediction of endometrial cancer patients based on miRNAs can be used to evaluate the disease progression of patients. In this study, we used differential analysis method to screen the differentially expressed miRNAs in endometrial cancer tissues using the miRNAs expression profile data of endometrial cancer, sample related clinical information and follow-up information from TCGA database. Univariate Cox regression analysis was used to initially screen the miRNAs that had an association with the prognosis of endometrial cancer. The screened miRNAs were subjected to multivariate Cox regression analysis to construct a 9-miRNA prognostic risk score model for endometrial cancer. Based on the K-M survival curve and ROC curve, the validity of the model was assessed. The results showed that patients with a high-risk score had a significantly worse prognosis than those with a low-risk score. The area under the ROC curve of model 5-year overall survival was 0.81, which illustrated that this model could effectively predict the prognostic risk of patients with endometrial cancer. Prediction of mRNAs potentially bound by model miRNAs was performed by online databases, and their functions were predicted by Gene Ontology (GO), Kyoto Encyclopedia of genes and genomes (KEGG). The results indicated that the present model miRNAs molecules were involved in several relevant metabolic pathways in endometrial cancer. Finally, to demonstrate the potential of the model application, we give two demonstrations of applying the model. We constructed a nomogram and developed a simple clinically assisted GUI software. Overall, we screened miRNAs molecular markers with associations with endometrial cancer prognosis, constructed a miRNAs prognostic risk score model which had good predictive efficacy for the survival status of endometrial cancer patients. It can provide strong theoretical support for clinicians in making judgment and developing treatment plans.
With the popularization of 5G(5th Generation Mobile Networks), telemedicine has ushered in new opportunities. At present, the application directions of 5G medical devices mainly include 5G remote monitoring, 5G remote diagnosis, 5G remote treatment, 5G wearable devices, etc. 5G medical devices have the dual attributes of medical devices and 5G wireless communication terminals, which brings challenges to the application and supervision. This paper conducted detailed research and analysis on marketing authorization, key performance, risk analysis and control, and whole-life cycle oversight, and gave suggestions on how to ensure the safety and effectiveness of 5G medical devices, so as to better facilitate the research and development and application of 5G medical devices.
Healthcare services are one of the major public issues facing the world and have become a research hotspot in recent years. This paper provides a review and quantitative analysis of the literature related to healthcare services to sort out the research trends and hotspots in this field. This paper used bibliometric analysis to search keywords in the Web of Science database in TITLE-ABS-KEY format. After screening, the final 4188 articles from 2011 to 2022 were collated, and co-citation analysis was performed using VOSviewer literature visualization analysis software. The global healthcare services literature has grown steadily year after year. Health Care Sciences & Services (9.5%) and General & Internal Medicine (8.7%) were the most common areas of research, with the United States (N=1016) accounting for the largest share of research. “BMC Health Services Research” is the most active journal (N=149). Healthcare services are one of the current research focuses in the field of healthcare institutions. In the future, further research into assistive technologies and tools for healthcare services, such as the Internet of Things, cloud services, and big data analytics will be investigated. In addition, research on healthcare services focusing on special needs populations, especially the elderly, is a hotspot for future research, broadening the boundaries of research and having practical value.
Purpose: BrainAge is a biomarker used to evaluate whether the brain is developing or aging normally, and accurate prediction of brain age in children can be used to determine whether children’s brain development is in a normal state. The aim of this study is to propose an accurate prediction of brain age in children aged 0–5 years based on T1 magnetic resonance images and multiple machine learning methods.Methods: Using two feature extraction methods based on radiomics first order grey scale feature extraction and Freesurfer feature extraction with five machine learning modelling methods (support vector machine regression SVR, Gaussian process regression GPR, correlation vector regression RVR, K-neighbourhood regression KNR, random forest regression RFR).Results: The best mean absolute error (MAE/days) and correlation coefficient R-squared values obtained for each method are shown below: SVR: MAE=182.27, R=0.76; GPR:MAE=85.26, R=0.96; RVR: MAE=123.97, R=0.93; KNR: MAE=162.86, R=0.81; RFR: MAE=81.83, R=0.96; where Freesurfer feature extraction method and RFR machine learning modelling method can achieve the best accuracy. This study explored the effects of different feature extraction methods and brain age prediction modelling methods on the prediction of brain age prediction accuracy, and the best accuracy of MAE=81.83 was achieved for the brain age prediction of children in the age group of 0–5 years.
Dental implants are widely used to replace lost natural teeth. Initial stability of the implant system is provided through an abutment connection which has been manufactured and marketed in various designs. The different designs may differently affect the longevity of the dental implant. This study aimed to compare the stress distribution of the implant components with four different abutment connection designs, namely hex, non-hex, penta-spline, and three-spline. Four three-dimensional finite element models of the maxillary left central incisor restored with a single-unit implant were constructed. Each model corresponds to the four designs of abutment connection. An angular load of 120 N was applied to the restored tooth and the two adjacent teeth, representing the occlusal force exerted on the anterior teeth. Von Mises stresses were measured at the abutment, fixture, and retaining screw. Results showed that the maximum stress concentration occurred at the abutment regardless of the abutment connection design. The abutment connection with three-spline generated the largest range of stress, while non-hex caused the smallest stress distribution.
