predictive model

Lifestyle and social health in fertility play a decisive role on mental health, and infertility is one of the unpleasant life experiences that can affect a person's mental health. In addition, marital satisfaction is also affected by infertility and plays an important role in the mental health of couples.

In order to find the relationship between these factors mentioned in the "background" section, this study aimed to determine a predictive model of depression in infertile women based on lifestyle and social health components mediated by marital satisfaction.

The present study was a descriptive and correlational study in which 360 infertile women were selected based on the depression score obtained from the Beck questionnaire (score 8 and above). Then, social health, lifestyle, and marital satisfaction questionnaires were completed by the selected infertile women. Subsequently, the raw data obtained were analyzed using SPSS-Ver.22 and partial least squares (PLS)-Version 3 software. Finally, a predictive model was designed based on the relationship and correlation between the variables.

The findings showed that there is a significant relationship between social health and depression (r = -0.55, P < 0.01), between social health and marital satisfaction (r = 0.48, P < 0.01), between social health and lifestyle (r = 0.52, P < 0.01), lifestyle and depression (r = -0.41, P < 0.01), between lifestyle and marital satisfaction (r = 0.59, P < 0.01), and between marital satisfaction and depression (r = -0.61, P < 0.01). Negative values of the above-mentioned correlations (r) indicate an indirect relationship and positive “r” values indicate a direct correlation.

Based on the results of the present research, the main hypothesis of this research was confirmed and it can be concluded that the model designed to predict the depression of infertile women based on lifestyle and social health components with the mediation of marital satisfaction is valid.

Mistreatment of elderlies is a global concern in the field of public health and human rights. The primary objective of this study was to identify and analyze risk factors associated with elder abuse through the use of a regression model. The present cross-sectional study was conducted to determine the predictors of elder abuse in Tehran, the number of samples was calculated 425.

Demographic Information Questionnaire and the Elderly Abuse Questionnaire with 49 items were used for this purpose. The Cronbach’s alpha was calculated 0.85 for the abuse in overall. Using univariate and multivariate linear regression models at a confidence interval of 0.95%.

The mean ± SD age of the participants was 67.44±8.01 years. Approximately 68.7% of the participants were male, while 31.3% were female. The results indicate that emotional abuse had the highest frequency of occurrence among the elderly, whereas physical abuse had the lowest. Among the factors studied, the living companion of the elderly (p< 0.001), education level (p< 0.001), financial independence (p< 0.001), and health status (p< 0.001) were found to be predictors of elder abuse.

As demographic variables have been identified as risk factors for elder abuse, it is recommended that social welfare organizations, care systems, and policymakers implement measures to develop plans and programs aimed at reducing the risk of elder abuse, preventing it at the community level, and addressing barriers related to the improvement of demographic variables.

The COVID-19 pandemic has rapidly spread and remained poorly understood by clinicians. The present work aimed to study the association between laboratory biomarkers, prognosis, and disease severity.

This is a single-center cohort study. We included young patients admitted at Razi Hospital, Ghaemshar City, Iran, from April 2020 to June 2020, whose diseases were confirmed with reverse transcription real time-PCR (rRT-PCR) test. Laboratory biomarkers were analyzed on the same day of inpatient service and after five days of hospitalization. The patients’ results and the outcomes were compared with those of the control group.

In the present study, 70 patients were investigated; 53 were discharged, and 17 died. A significant correlation was observed between patients and healthy subjects in some laboratory biomarkers: C-reactive protein (CRP), lactate dehydrogenase (LDH), total protein level, albumin level, and absolute lymphocyte count. Furthermore, CRP, LDH, total protein, albumin, absolute lymphocyte count, 25-OH vitamin D, interleukin (IL)-6, ferritin, and D-dimer levels in patients with different outcomes had significant correlations. High CRP, LDH, IL-6, ferritin, and D-dimer were predictive of mortality (area under the curve >0.70), as were low absolute lymphocyte count and 25-OH vitamin D. After adjusting age, CRP, albumin, WBC, D-dimer, LDH, and 25 OH-vitamin D, the final model of multiple binary logistic regressions with IL-6 and ferritin had high accuracy for the prediction of fatal outcome.

This finding would facilitate the early stratification of hospitalized patients with COVID-19 and help make clinical decisions.

It is of prime importance to manage trauma patients in the early hours and use easy trauma severity scoring systems to make decisions and evaluate patient prognosis.

The present study aimed to design a predictive model of the mortality of multi-trauma patients due to traffic accidents.

This cross-sectional analytical study was performed on 600 patients who suffered from multi-trauma caused by traffic accidents from December 2019 to September 2021. Collected data included age, sex, vital signs, trauma mechanism, involved vehicle in the accident, accident location, and hospital outcome.

In this study, 600 multi-trauma cases caused by traffic accidents were evaluated. Among the significant variables included in the regression model, age, Mean Arterial Pressure (MAP), Glasgow Coma Scale (GCS), AVPU (Alert, Verbal response, Pain response, Unresponsive), and vehicle versus fixed objects (in Vehicle 2) in the presence of other variables in the model, significantly predictedpatient outcomes. Therefore, with the other variables being constant, one unit increase in the age variable increases the probability of death by 1.04 times, one unit increase in the score of the two variables of MAP and GCS, and also the transfer of trauma mechanism from the fixed object to the vehicle reduces death by 0.92, 0.62, and 0.10 times, respectively. In the AVPU variable, the transition from Alert to Verbal, the transition from Verbal to Pain, and the transition from Pain to Unresponsive increases the probability of death by 32, 104, and 567, respectively.

In this study, AVPU, age, MAP, primary GCS, and trauma mechanism due to hitting a vehicle with a fixed object had significantly the highest predictive power of hospital mortality in patients with multiple trauma due to traffic accidents, respectively. It is suggested that further studies be performed to replace the AVPU variable with GCS in the newly designed formulas for calculating the severity of trauma to simplify these scores.

The COVID-19 pandemic affected educational institutions and caused the transfer to distance learning. The study aimed to assess medical students' satisfaction with synchronous distance learning (SDL) during the pandemic and predict their choice of it in the future.

A cross-sectional study was conducted among undergraduate medical students at the Al-Farabi Kazakh National University in July 2021. An online questionnaire was used to collect data on demographic and educational characteristics, satisfaction, and perspective on the future choice of SDL. IBM SPSS Statistics, version 26, was used to analyze the qualitative data on descriptive and inferential statistics.

Students' satisfaction and future choice of SDL were 43.2% and 20.2%, respectively. Regression analysis revealed the significance of SDL predictors with a direct relationship in the case of gender and academic performance and predictors with an inverse relationship in the case of residency, student life satisfaction, and SDL with student satisfaction. In the predictive model, student satisfaction and probability of future choice of SDL over traditional learning were 59.5% and 43.5%, respectively; over hybrid learning, it was 20.2% of students.

The research findings suggested that slightly less than half of the respondent medical students were satisfied by distance learning during the COVID-19 pandemic when their satisfaction probability was significantly higher in the predictive model. The predictive model of perspective of choice of distance learning did not show statistically significant results compared with traditional and hybrid learning.

In recent years, medicine supply chain management has become more significant, especially after the Covid-19 pandemic. The most important issue is supply chain cost control. Medicine costs include nearly 30% of hospital expenses. If the drug inventory is not properly managed, it will lead to issues such as the lack of inventory of certain drugs, provision of excess inventory, increased costs, and, finally, patient dissatisfaction.

In this study, an attempt has been made to predict and manage the pharmaceutical needs of hospitals using artificial neural networks and deep learning algorithms. The prescription and consumption information of one of the general hospitals in Hamedan city from 2013 until 2017 was extracted from the HIS databases. Since most drugs have different and specific characteristics and it is impossible to create the same prediction model for all drugs, the employed dataset is limited to the category of antibiotics. As a case study, the accuracy of the predictive model is evaluated, especially for cefazolin. We use a deep model to analyze the medical data time series efficiently. This model consists of two parts, a Convolutional Neural Network and a Long Short-Term Memory network (LSTM), which can sufficiently recognize the change history in time series prediction applications. The proposed model with many adjustable parameters in convolutional neural networks will bring good performance to overcome the complexities of the learning problem.

Using deep learning in the training process can increase robustness by reducing the effects of complexity and uncertainty in medical data. Eventually, the prediction evaluation results and analytical criteria such as forecasting error and convergence speed and some statistical tests like R^2, MAE, and RMSE were presented. The average forecasting error for the proposed method is 0.028, and the measured values for RMSE, MAE, and R^2 are 0.095, 0.081, and 0.788, respectively.

A comprehensive comparison between some other predictive methods and the implemented model shows the outperformance of the proposed approach. Additionally, the evaluation results indicate the efficiency of the proposed approach.

Omitting axillary lymph node dissection (ALND) is recommended for early-stage breast cancer patients with 1-2 sentinel lymph nodes (SLNs) macro-metastases and breast-conserving therapy. However, it is not safe for part of patients, so it is significant to find risk factors and develop a predictive model of non-SLNs metastases in breast cancer patients with 1-2 SLNs macro-metastases and breast-conserving therapy.

This retrospective study enrolled 228 breast cancer patients with 1-2 SLNs macro-metastases who underwent ALND and breast-conserving surgery between Jan 2012 and Dec 2017 at Cancer Hospital Chinese Academy of Medical Sciences. Chi-square test and backward stepwise binary logistic regression were used to find factors that influenced non-SLN metastases, then a predictive model was formulated and obtained its area under the curve.

Tumor pathologic invasion size, number of positive SLNs and ALN status on imaging was associated with non-SLNs metastases. The predictive model was also formulated based on these three factors to assess and the area under the curve of model was 0.708.

We developed a predictive model to assess the high-risk cohort of patients of non-SLNs metastases which can be an auxiliary tool for doctors.