Technology and Health Care最新文献

Background: Early pregnancy thyroid function assessment in mothers is covered. The benefits of using load-specific reference ranges are well-established.

Objective: We pondered whether the categorization of maternal thyroid function would change if multiple blood samples obtained early in pregnancy were used. Even though binary classification is a common goal of current disease diagnosis techniques, the data sets are small, and the outcomes are not validated. Most current approaches concentrate on model optimization, focusing less on feature engineering.

Methods: The suggested method can predict increased protein binding, non-thyroid syndrome (NTIS) (simultaneous non-thyroid disease), autoimmune thyroiditis (compensated hypothyroidism), and Hashimoto's thyroiditis (primary hypothyroidism). In this paper, we develop an automatic thyroid nodule classification system using a multi-scale vision transformer and image enhancement. Graph equalization is the chosen technique for image enhancement, and in our experiments, we used neural networks with four-layer network nodes. This work presents an enhanced linguistic coverage neuro-fuzzy classifier with chosen features for thyroid disease feature selection diagnosis. The training procedure is optimized, and a multi-scale vision transformer network is employed. Each hop connection in Dense Net now has trainable weight parameters, altering the architecture. Images of thyroid nodules from 508 patients make up the data set for this article. Sets of 80% training and 20% validation and 70% training and 30% validation are created from the data. Simultaneously, we take into account how the number of training iterations, network structure, activation function of network nodes, and other factors affect the classification outcomes.

Results: According to the experimental results, the best number of training iterations is 500, the logistic function is the best activation function, and the ideal network structure is 2500-40-2-1.

Conclusion: K-fold validation and performance comparison with previous research validate the suggested methodology's enhanced effectiveness.

Background: The global implementation of Electronic Health Records has significantly enhanced the quality of medical care and the overall delivery of public health services. The incorporation of Evidence-Based Medicine offers numerous benefits and enhances the efficacy of decision-making in areas such as prevention, prognosis, diagnosis, and therapeutic approaches.

Objective: The objective of this paper is to propose an architectural design of an Evidence-Based Medicine information system based on the Electronic Health Record, taking into account the existing and future level of interoperability of health information systems in Greece.

Methods: A study of the suggested evidence-based medicine architectures found in the existing literature was conducted. Moreover, the interoperability architecture of health information systems in Greece was analyzed. The architecture design reviewed by specialized personnel and their recommendations were incorporated into the final design of the proposed architecture.

Results: The proposed integrated architecture of an Evidence-Based Medicine system based on the Electronic Health Record integrates and utilizes citizens' health data while leveraging the existing knowledge available in the literature.

Conclusions: Taking into consideration the recently established National Interoperability Framework, which aligns with the European Interoperability Framework, the proposed realistic architectural approach contributes to improving the quality of healthcare provided through the ability to make safe, timely and accurate decisions by physicians.

Background: To avoid risks of mucosal infection from contact lenses removal, a contact lens plunger is often used.

Objective: Given various types of contact lens plungers available on the market, no study has yet been done on mechanical effects of the contact lens plunger on contact lens removal. Here, this study used finite element analysis to investigate the effects of plunger size and plunger position on the removal of soft and rigid gas permeable (RGP) contact lenses.

Methods: First, we established finite element analysis models for the plunger, contact lens, cornea, and aqueous humor. The plunger is made of mostly silicone rubber, and the contact lenses are mainly made of soft and hard material. The part of the plunger used for removal was located either at the central or the edged position, with pulling 1 mm distance. The main parameters observation indicators of in this study were the reaction force at the fixed end of the cornea, aqueous humor, the von Mises stress of the plunger, the contact lenses, and the cornea.

Results: Results of this study showed that when a plunger of a larger diameter was used, the reaction force of the plunger was also larger, especially when applied to RGP lenses, which required a slightly larger force (∼0.27 N). Also, when removing a RGP lens from the edge, there was a greater stress at the edge of the contact lens (2.5799 MPa), and this caused a higher stress on both the cornea (0.0165 MPa) and the aqueous humor (0.00114 MPa).

Conclusions: When using a plunger with a larger diameter to remove a RGP lens, although a larger force required, the relatively larger contact area likely reduced the stress on the cornea and aqueous humor, thereby reducing the risk of eye injury. In addition, when removing a RGP lens, the results of this study recommended it to be removed from the plunger edge, as that facilitated the removal of contact lens.

Background: Segmentation of retinal fragments like blood vessels, Optic Disc (OD), and Optic Cup (OC) enables the early detection of different retinal pathologies like Diabetic Retinopathy (DR), Glaucoma, etc.

Objective: Accurate segmentation of OD remains challenging due to blurred boundaries, vessel occlusion, and other distractions and limitations. These days, deep learning is rapidly progressing in the segmentation of image pixels, and a number of network models have been proposed for end-to-end image segmentation. However, there are still certain limitations, such as limited ability to represent context, inadequate feature processing, limited receptive field, etc., which lead to the loss of local details and blurred boundaries.

Methods: A multi-dimensional dense attention network, or MDDA-Net, is proposed for pixel-wise segmentation of OD in retinal images in order to address the aforementioned issues and produce more thorough and accurate segmentation results. In order to acquire powerful contexts when faced with limited context representation capabilities, a dense attention block is recommended. A triple-attention (TA) block is introduced in order to better extract the relationship between pixels and obtain more comprehensive information, with the goal of addressing the insufficient feature processing. In the meantime, a multi-scale context fusion (MCF) is suggested for acquiring the multi-scale contexts through context improvement.

Results: Specifically, we provide a thorough assessment of the suggested approach on three difficult datasets. In the MESSIDOR and ORIGA data sets, the suggested MDDA-NET approach obtains accuracy levels of 99.28% and 98.95%, respectively.

Conclusion: The experimental results show that the MDDA-Net can obtain better performance than state-of-the-art deep learning models under the same environmental conditions.

Background: Irrational pharmacotherapy and increasing pharmacy costs remain major concerns in healthcare systems. Pharmacists are expected to employ diagnosis-related group (DRG) data to analyse inpatient pharmacy utilization.

Objective: This project aimed to pilot an efficient pharmacist-led programme to analyse factors related to pharmacy expenses, evaluate the rational use of drugs in batch processing, and make further interventions based on DRG data.

Methods: Patients from the OB25 (caesarean section without comorbidities or complications) DRG were selected in 2018, and the most relevant factors were identified through statistical analysis. Interventions were implemented by sending monthly reports on prescribing data and drug review results for the same DRGs to the department starting in 2019. Pre-post comparisons were conducted to demonstrate changes in pharmacy costs and appropriateness at a tertiary teaching hospital with 2,300 beds in China.

Results: A total of 1,110 patients were identified from the OB25 DRG data in 2018. Multivariate linear analysis indicated that the number of items prescribed and wards substantially influenced pharmacy expenditure. Drugs labelled as vital, essential, and non-essential revealed that 46.6% of total pharmacy costs were spent on non-essential drugs, whereas 38.7% were spent on vital drugs. The use of inappropriate pharmaceuticals and drug items was substantially reduced, and the average pharmacy cost after intervention was 336.7 RMB in 2020. The benefit-cost ratio of the programme was 9.86.

Conclusion: Interventions based on DRG data are highly efficient and feasible for reducing inpatient pharmacy costs and non-essential drug use.

Background: The Stepped Care Model (SCM) is an evidence-based treatment approach that tailors treatment intensity based on patients' health status, aiming to achieve the most positive treatment outcomes with the least intensive and cost-effective interventions. Currently, the effectiveness of the Stepped Care Model in postoperative rehabilitation for TKA (Total Knee Arthroplasty) patients has not been reported.

Objective: The present study aimed to investigate whether the stepped care model could improve early-stage self-report quality of life and knee function after total knee arthroplasty via a prospective randomized controlled design.

Methods: It was a mono-center, parallel-group, open-label, prospective randomized controlled study. Patients who aging from 60-75 years old as well as underwent unilateral primary total knee arthroplasty due to end-stage knee osteoarthritis between 2020.06 to 2022.02 were enrolled. Participants were randomized and arranged into two groups in a 1:1 allocation. The control group was given traditional rehabilitation guidance, while the stepped care model group was given continued stepped care. Hospital for special surgery knee score, daily living ability (ADL), knee flexion range, and adverse events at 1, 3, and 6 months after total knee arthroplasty were recorded.

Results: 88 patients proceeded to the final analysis. There was no significant difference of age, gender, length of stay, BMI, and educational level between the two groups at the baseline. After specific stepped care model interventions, patients showed significant improvements in HHS in 1 month (85.00 (82.25, 86.00) vs. 80.00 (75.00, 83.00), p< 0.001), 3 months (88.00 (86.00, 92.00) vs. 83.00 (76.75, 85.00), p< 0.001), and 6 months (93.00 (90.25, 98.00) vs. 88.00 (84.25, 91.75), p< 0.001) when compared with the control group. Similar results were also found in both daily living ability and knee flexion angle measurements. No adverse event was observed during the follow-up.

Conclusion: The present study found that the stepped care model intervention significantly improved early-stage knee function and self-reported life quality after total knee arthroplasty due to knee osteoarthritis. Female patients and those less than 70 years old benefit more from the stepped care model intervention after total knee arthroplasty.

Background: Gastrointestinal tract (GIT) diseases impact the entire digestive system, spanning from the mouth to the anus. Wireless Capsule Endoscopy (WCE) stands out as an effective analytic instrument for Gastrointestinal tract diseases. Nevertheless, accurately identifying various lesion features, such as irregular sizes, shapes, colors, and textures, remains challenging in this field.

Objective: Several computer vision algorithms have been introduced to tackle these challenges, but many relied on handcrafted features, resulting in inaccuracies in various instances.

Methods: In this work, a novel Deep SS-Hexa model is proposed which is a combination two different deep learning structures for extracting two different features from the WCE images to detect various GIT ailment. The gathered images are denoised by weighted median filter to remove the noisy distortions and augment the images for enhancing the training data. The structural and statistical (SS) feature extraction process is sectioned into two phases for the analysis of distinct regions of gastrointestinal. In the first stage, statistical features of the image are retrieved using MobileNet with the support of SiLU activation function to retrieve the relevant features. In the second phase, the segmented intestine images are transformed into structural features to learn the local information. These SS features are parallelly fused for selecting the best relevant features with walrus optimization algorithm. Finally, Deep belief network (DBN) is used classified the GIT diseases into hexa classes namely normal, ulcer, pylorus, cecum, esophagitis and polyps on the basis of the selected features.

Results: The proposed Deep SS-Hexa model attains an overall average accuracy of 99.16% in GIT disease detection based on KVASIR and KID datasets. The proposed Deep SS-Hexa model achieves high level of accuracy with minimal computational cost in the recognition of GIT illness.

Conclusions: The proposed Deep SS-Hexa Model progresses the overall accuracy range of 0.04%, 0.80% better than GastroVision, Genetic algorithm based on KVASIR dataset and 0.60%, 1.21% better than Modified U-Net, WCENet based on KID dataset respectively.

Background: Esthetic orthodontic wires are preferred for their ease to fit in with natural tooth color, but their susceptibility to staining in the oral environment poses a concern. Various Coatings such as Teflon and Epoxy aim to enhance appearance and biocompatibility but may still result in discoloration. Understanding the color stability of these wires under different staining conditions is crucial for a better and enhanced treatment plan.

Objective: This study intended to assess the color stability of esthetic orthodontic wires under various staining solutions that are often used in daily life.

Method: Color changes of Teflon and Epoxy-coated esthetic orthodontic wires were meticulously measured at baseline, 7, 14, and 21-day intervals utilizing the precise CIE Lab* color measurement system. A total of thirty-two samples of wires from each brand were prepared (n= 8/group) and immersed in staining solutions (coffee, tea, cola, and saffron). The color change within and between the groups was statistically evaluated (p< 0.05).

Results: Significant variations in color stability were observed across different staining solutions. Saffron emerged as the most potent agent, inducing the most pronounced color changes, whereas cola demonstrated the least impact. Furthermore, Epoxy-coated wires consistently exhibited superior color stability compared to their Teflon-coated counterparts across all staining solutions and time intervals.

Conclusion: This study underlines the significance for orthodontists to consider staining agents' possible effects on orthodontic wires into account when selecting the orthodontic wires. The findings suggest that Epoxy-coated wires hold promise in mitigating discoloration issues during orthodontic therapy.

Background: Diagnostic challenges exist for CMV pneumonia in post-hematopoietic stem cell transplantation (post-HSCT) patients, despite early-phase radiographic changes.

Objective: The study aims to employ a deep learning model distinguishing CMV pneumonia from COVID-19 pneumonia, community-acquired pneumonia, and normal lungs post-HSCT.

Methods: Initially, 6 neural network models were pre-trained with COVID-19 pneumonia, community-acquired pneumonia, and normal lung CT images from Kaggle's COVID multiclass dataset (Dataset A), then Dataset A was combined with the CMV pneumonia images from our center, forming Dataset B. We use a few-shot transfer learning strategy to fine-tune the pre-trained models and evaluate model performance in Dataset B.

Results: 34 cases of CMV pneumonia were found between January 2018 and December 2022 post-HSCT. Dataset A contained 1681 images of each subgroup from Kaggle. Combined with Dataset A, Dataset B was initially formed by 98 images of CMV pneumonia and normal lung. The optimal model (Xception) achieved an accuracy of 0.9034. Precision, recall, and F1-score all reached 0.9091, with an AUC of 0.9668 in the test set of Dataset B.

Conclusions: This framework demonstrates the deep learning model's ability to distinguish rare pneumonia types utilizing a small volume of CT images, facilitating early detection of CMV pneumonia post-HSCT.