Background: Musculoskeletal disorders (MSDs) involve muscles, nerves, tendons, joints, cartilage, and spinal discs. These conditions can be triggered by both the work environment and the type of work performed, factors that, in some cases, can also exacerbate pre-existing conditions. This systematic review aims to provide an overview of the impact that different work-related activities have on the musculoskeletal system. Methods: A global search of publications was conducted using the following international bibliographic web databases: PubMed and Web of Science. The search strategies combined terms for musculoskeletal disorders and workers. In addition, a meta-analysis was conducted to estimate the prevalence of MSDs within the healthcare sector. Results: A total of 10,805 non-duplicated articles were identified, and finally, 32 studies were reviewed in this article. Once the literature search was completed, occupational figures were categorized into healthcare, farming, industrial, and computer sectors. In the healthcare sector, the prevalence estimate for degenerative diseases of the lumbar spine was 21% (497 out of 2547 physicians and dentists) (95% CI, 17-26%), while for osteoarthritis of the hand, it was 37% (382 out of 1013 dentists) (95% CI, 23-51%). Conclusions: Musculoskeletal disorders significantly impair workers\' quality of life, especially in healthcare sector. These conditions are also associated with high costs for employers, such as absenteeism, lost productivity, and increased costs for healthcare, disability, and workers\' compensation.

UNASSIGNED: To develop a natural language processing application capable of automatically identifying benign gallbladder diseases that require surgery, from radiology reports.
UNASSIGNED: We developed a text classifier to classify reports as describing benign diseases of the gallbladder that do or do not require surgery. We randomly selected 1,200 reports describing the gallbladder from our database, including different modalities. Four radiologists classified the reports as describing benign disease that should or should not be treated surgically. Two deep learning architectures were trained for classification: a convolutional neural network (CNN) and a bidirectional long short-term memory (BiLSTM) network. In order to represent words in vector form, the models included a Word2Vec representation, with dimensions of 300 or 1,000. The models were trained and evaluated by dividing the dataset into training, validation, and subsets (80/10/10).
UNASSIGNED: The CNN and BiLSTM performed well in both dimensional spaces. For the 300- and 1,000-dimensional spaces, respectively, the F1-scores were 0.95945 and 0.95302 for the CNN model, compared with 0.96732 and 0.96732 for the BiLSTM model.
UNASSIGNED: Our models achieved high performance, regardless of the architecture and dimensional space employed.
UNASSIGNED: Desenvolver uma aplicação de processamento de linguagem natural capaz de identificar automaticamente doenças cirúrgicas benignas da vesícula biliar a partir de laudos radiológicos.
UNASSIGNED: Desenvolvemos um classificador de texto para classificar laudos como contendo ou não doenças cirúrgicas benignas da vesícula biliar. Selecionamos aleatoriamente 1.200 laudos com descrição da vesícula biliar de nosso banco de dados, incluindo diferentes modalidades. Quatro radiologistas classificaram os laudos como doença benigna cirúrgica ou não. Duas arquiteturas de aprendizagem profunda foram treinadas para a classificação: a rede neural convolucional (convolutional neural network - CNN) e a memória longa de curto prazo bidirecional (bidirectional long short-term memory - BiLSTM). Para representar palavras de forma vetorial, os modelos incluíram uma representação Word2Vec, com dimensões variando de 300 a 1000. Os modelos foram treinados e avaliados por meio da divisão do conjunto de dados entre treinamento, validação e teste (80/10/10).
UNASSIGNED: CNN e BiLSTM tiveram bom desempenho em ambos os espaços dimensionais. Relatamos para 300 e 1000 dimensões, respectivamente, as pontuações F1 de 0,95945 e 0,95302 para o modelo CNN e de 0,96732 e 0,96732 para a BiLSTM.
UNASSIGNED: Nossos modelos alcançaram alto desempenho, independentemente de diferentes arquiteturas e espaços dimensionais.

UNASSIGNED: The American Optometric Association defines computer vision syndrome (CVS), also known as digital eye strain, as \"a group of eye- and vision-related problems that result from prolonged computer, tablet, e-reader and cell phone use\". We aimed to create a well-structured, valid, and reliable questionnaire to determine the prevalence of CVS, and to analyze the visual, ocular surface, and extraocular sequelae of CVS using a novel and smart self-assessment questionnaire.
UNASSIGNED: This multicenter, observational, cross-sectional, descriptive, survey-based, online study included 6853 complete online responses of medical students from 15 universities. All participants responded to the updated, online, fourth version of the CVS questionnaire (CVS-F4), which has high validity and reliability. CVS was diagnosed according to five basic diagnostic criteria (5DC) derived from the CVS-F4. Respondents who fulfilled the 5DC were considered CVS cases. The 5DC were then converted into a novel five-question self-assessment questionnaire designated as the CVS-Smart.
UNASSIGNED: Of 10 000 invited medical students, 8006 responded to the CVS-F4 survey (80% response rate), while 6853 of the 8006 respondents provided complete online responses (85.6% completion rate). The overall CVS prevalence was 58.78% (n = 4028) among the study respondents; CVS prevalence was higher among women (65.87%) than among men (48.06%). Within the CVS group, the most common visual, ocular surface, and extraocular complaints were eye strain, dry eye, and neck/shoulder/back pain in 74.50% (n = 3001), 58.27% (n = 2347), and 80.52% (n = 3244) of CVS cases, respectively. Notably, 75.92% (3058/4028) of CVS cases were involved in the Mandated Computer System Use Program. Multivariate logistic regression analysis revealed that the two most statistically significant diagnostic criteria of the 5DC were ≥2 symptoms/attacks per month over the last 12 months (odds ratio [OR] = 204177.2; P <0.0001) and symptoms/attacks associated with screen use (OR = 16047.34; P <0.0001). The CVS-Smart demonstrated a Cronbach\'s alpha reliability coefficient of 0.860, Guttman split-half coefficient of 0.805, with perfect content and construct validity. A CVS-Smart score of 7-10 points indicated the presence of CVS.
UNASSIGNED: The visual, ocular surface, and extraocular diagnostic criteria for CVS constituted the basic components of CVS-Smart. CVS-Smart is a novel, valid, reliable, subjective instrument for determining CVS diagnosis and prevalence and may provide a tool for rapid periodic assessment and prognostication. Individuals with positive CVS-Smart results should consider modifying their lifestyles and screen styles and seeking the help of ophthalmologists and/or optometrists. Higher institutional authorities should consider revising the Mandated Computer System Use Program to avoid the long-term consequences of CVS among university students. Further research must compare CVS-Smart with other available metrics for CVS, such as the CVS questionnaire, to determine its test-retest reliability and to justify its widespread use.

OBJECTIVE: To propose a convolutional neural network (EmbNet) for automatic pulmonary embolism detection on computed tomography pulmonary angiogram (CTPA) scans and to assess its diagnostic performance.
METHODS: 305 consecutive CTPA scans between January 2019 and December 2021 were enrolled in this study (142 for training, 163 for internal validation), and 250 CTPA scans from a public dataset were used for external validation. The framework comprised a preprocessing step to segment the pulmonary vessels and the EmbNet to detect emboli. Emboli were divided into three location-based subgroups for detailed evaluation: central arteries, lobar branches, and peripheral regions. Ground truth was established by three radiologists.
RESULTS: The EmbNet\'s per-scan level sensitivity, specificity, positive predictive value (PPV), and negative predictive value were 90.9%, 75.4%, 48.4%, and 97.0% (internal validation) and 88.0%, 70.5%, 42.7%, and 95.9% (external validation). At the per-embolus level, the overall sensitivity and PPV of the EmbNet were 86.0% and 61.3% (internal validation), and 83.5% and 57.5% (external validation). The sensitivity and PPV of central emboli were 89.7% and 52.0% (internal validation), and 94.4% and 43.0% (external validation); of lobar emboli were 95.2% and 76.9% (internal validation), and 93.5% and 72.5% (external validation); and of peripheral emboli were 82.6% and 61.7% (internal validation), and 80.2% and 59.4% (external validation). The average false positive rate was 0.45 false emboli per scan (internal validation) and 0.69 false emboli per scan (external validation).
CONCLUSIONS: The EmbNet provides high sensitivity across embolus locations, suggesting its potential utility for initial screening in clinical practice.

BACKGROUND: ICU readmissions and post-discharge mortality pose significant challenges. Previous studies used EHRs and machine learning models, but mostly focused on structured data. Nursing records contain crucial unstructured information, but their utilization is challenging. Natural language processing (NLP) can extract structured features from clinical text. This study proposes the Crucial Nursing Description Extractor (CNDE) to predict post-ICU discharge mortality rates and identify high-risk patients for unplanned readmission by analyzing electronic nursing records.
OBJECTIVE: Developed a deep neural network (NurnaNet) with the ability to perceive nursing records, combined with a bio-clinical medicine pre-trained language model (BioClinicalBERT) to analyze the electronic health records (EHRs) in the MIMIC III dataset to predict the death of patients within six month and two year risk.
METHODS: A cohort and system development design was used.
METHODS: Based on data extracted from MIMIC-III, a database of critically ill in the US between 2001 and 2012, the results were analyzed.
METHODS: We calculated patients\' age using admission time and date of birth information from the MIMIC dataset. Patients under 18 or over 89 years old, or who died in the hospital, were excluded. We analyzed 16,973 nursing records from patients\' ICU stays.
METHODS: We have developed a technology called the Crucial Nursing Description Extractor (CNDE), which extracts key content from text. We use the logarithmic likelihood ratio to extract keywords and combine BioClinicalBERT. We predict the survival of discharged patients after six months and two years and evaluate the performance of the model using precision, recall, the F1-score, the receiver operating characteristic curve (ROC curve), the area under the curve (AUC), and the precision-recall curve (PR curve).
RESULTS: The research findings indicate that NurnaNet achieved good F1-scores (0.67030, 0.70874) within six months and two years. Compared to using BioClinicalBERT alone, there was an improvement in performance of 2.05 % and 1.08 % for predictions within six months and two years, respectively.
CONCLUSIONS: CNDE can effectively reduce long-form records and extract key content. NurnaNet has a good F1-score in analyzing the data of nursing records, which helps to identify the risk of death of patients after leaving the hospital and adjust the regular follow-up and treatment plan of relevant medical care as soon as possible.

The purpose was to investigate relationships of cumulative load and cartilage turnover biomarkers with 2-year changes in cartilage in knee osteoarthritis. From participants with Kellgren-Lawrence (KL) grades of 1 to 3, cartilage thickness and transverse relaxation time (T2) were computed from 24-month (baseline) and 48-month magnetic resonance images. Cumulative load was the interaction term of the Physical Activity Scale for the Elderly (PASE) and body mass index (BMI). Serum cartilage oligomeric matrix protein (COMP) and the nitrated form of type II collagen (Coll2-1 NO2) were collected at baseline. Multiple regressions (adjusted for baseline age, KL grade, cartilage measures, pain, comorbidity) evaluated the relationships of cumulative load and biomarkers with 2-year changes. In 406 participants (63.7 (8.7) years), interactions of biomarkers with cumulative load weakly predicted 2-year cartilage changes: (i) COMP × cumulative load explained medial tibia thickness change (R2 increased 0.062 to 0.087, p < 0.001); (ii) Coll2-1 NO2 × cumulative load explained central medial femoral T2 change (R2 increased 0.177 to 0.210, p < 0.001); and (iii) Coll2-1 NO2 × cumulative load explained lateral tibia T2 change (R2 increased 0.166 to 0.188, p < 0.001). Moderate COMP or Coll2-1 NO2 at baseline appeared protective. High COMP or Coll2-1 NO2, particularly with high BMI and low PASE, associated with worsening cartilage. Moderate serum concentrations of cartilage turnover biomarkers, at high and low physical activity, associated with maintained cartilage outcomes over 2 years. In conclusion, high concentrations of cartilage turnover biomarkers, particularly with high BMI and low physical activity, associated with knee cartilage thinning and increasing T2 over 2 years. Key Points • Higher quality cartilage may be better able to tolerate a larger cumulative load than poor quality cartilage. • Among participants enrolled in the Osteoarthritis Initiative Biomarkers Consortium Project, a representation of cumulative load exposure and its interaction with cartilage turnover biomarkers were weakly related with 2-year change in knee cartilage. • These findings suggest that cartilage turnover is a factor that modifies the relationship between loading exposure and cartilage loss in knee OA.

Advancements in diagnostic methods and surgical techniques for keratoconus (KC) have increased non-invasive treatment options. Successful surgical planning for KC involves a combination of clinical science, empirical evidence, and surgical expertise. Assessment of disease progression is crucial, and halting the progression should be the focus if it is progressive. While surgeons used to rely on experience alone to decide the surgical method, comparing the network of primary factors, such as visual acuity, across studies can help them choose the most appropriate treatments for each patient and achieve optimal outcomes. Meticulous tabulation methods facilitate interpretation, highlighting the importance of selecting the correct surgical and rehabilitation approach based on each patient\'s condition and stage of the disease. We detail the outcomes of a comprehensive network meta-analysis comparing the effectiveness of various combined therapeutic refractive treatments for KC at identical stages of the disease, spanning 4 distinct follow-up intervals. Additionally, the comprehensive analysis suggests that for corneas with optimal best corrected visual acuity (BCVA) preoperatively (classified as regular), combining phakic intraocular lenses with intracorneal ring segments (ICRS) and corneal cross-linking (CXL) could offer the best therapeutic approach provided the disease stage does not exceed stage 3. For irregular corneas, although initial follow-ups show a significant difference in BCVA with surface ablation, longer-term follow-ups recommend combining surface ablation with ICRS and CXL, especially at higher stages.

This paper investigates the integration of augmented reality (AR) technology into English translation teaching for college students, emphasizing the pivotal role of innovative teaching methods in enhancing students\' translation skills and learning experiences. To address the issue of insufficient interest in English translation teaching, the paper initially assesses the purpose and significance of learning English translation through a questionnaire survey, elucidating challenges encountered in English language acquisition. Subsequently, adhering to AR principles, a teaching demonstration platform rooted in AR technology is conceived and developed, intricately aligned with English translation instruction. The platform serves as a solution to issues in English learning, such as inadequate course comprehension, low utilization of teaching resources, and instructors\' lack of experience. The research culminates in the analysis of survey results, wherein the quantitative disparities in translation ability between students utilizing the research platform and those subjected to traditional teaching methods are examined. The findings underscore the positive impact of the AR-based research platform on improving students\' translation proficiency. The AR platform heightens learners\' engagement in the learning process, contributes to constructing a robust knowledge framework, and enhances overall learning outcomes. The platform offers educators opportunities to optimize experimental courses and elevate teaching standards. The paper\'s outcomes present novel pedagogical scenarios for learners, propose technical solutions for other technical disciplines and furnish a theoretical foundation and application model for a new generation of experimental demonstration platforms.

Studies have shown that the digital divide affects students\' educational achievement across racial and ethnic groups. In light of this, the study investigates the effect of technology access at home on student learning hours during the COVID-19 pandemic and across racial and ethnic groups in the U.S. The Household Pulse Surveys (HPS), conducted by the United States Census Bureau and administered from August 19, 2020, to March 29, 2021, were used for the analysis. We compute a composite index of technology access using the principal component analysis (PCA). And for the empirical model, the study employed a Tobit regression model. The result shows that the estimated index of technology access based on PCA for the whole sample is about 0.92, indicating a higher level of access. However, the breakdown by race/ethnicity shows an average of about 0.93, 0.89, 0.90, 0.94, and 0.89 for students representing White, Black, Hispanic, Asia, and other races, respectively. This means the intensity at which households in the sample have access to technology is higher among the Asian and White students, followed by Hispanic, Black, and other races in that order. The estimated effect of technology access on the student learning hours during COVID-19 based on the Tobit regression model shows about a 3.1 unit points increase over the whole sample. And further analysis reveals variation at which access to the technology impacts learning hours across race and ethnicity groups. For example, we find that access to technology significantly increased learning hours by about 3.5, 1.6, 2.2, and 3.4 unit points among White, Black, Hispanic, and Asian students, respectively. The observed differing effect of access to technology on learning hours further highlights the racial disparities in American society\'s digital divide, which reveal how access to technology disproportionately impacts student learning hours during the COVID-19 pandemic across race and ethnicity.

BACKGROUND: The coronavirus disease 2019 (COVID-19) pandemic continues to pose significant challenges to the public health sector, including that of the United Arab Emirates (UAE). The objective of this study was to assess the efficiency and accuracy of various deep-learning models in forecasting COVID-19 cases within the UAE, thereby aiding the nation\'s public health authorities in informed decision-making.
METHODS: This study utilized a comprehensive dataset encompassing confirmed COVID-19 cases, demographic statistics, and socioeconomic indicators. Several advanced deep learning models, including long short-term memory (LSTM), bidirectional LSTM, convolutional neural network (CNN), CNN-LSTM, multilayer perceptron, and recurrent neural network (RNN) models, were trained and evaluated. Bayesian optimization was also implemented to fine-tune these models.
RESULTS: The evaluation framework revealed that each model exhibited different levels of predictive accuracy and precision. Specifically, the RNN model outperformed the other architectures even without optimization. Comprehensive predictive and perspective analytics were conducted to scrutinize the COVID-19 dataset.
CONCLUSIONS: This study transcends academic boundaries by offering critical insights that enable public health authorities in the UAE to deploy targeted data-driven interventions. The RNN model, which was identified as the most reliable and accurate for this specific context, can significantly influence public health decisions. Moreover, the broader implications of this research validate the capability of deep learning techniques in handling complex datasets, thus offering the transformative potential for predictive accuracy in the public health and healthcare sectors.