The predictive power of data: machine learning analysis for Covid-19 mortality based on personal, clinical, preclinical, and laboratory variables in a case-control study.

BACKGROUND AND PURPOSE: The COVID-19 pandemic has presented unprecedented public health challenges worldwide. Understanding the factors contributing to COVID-19 mortality is critical for effective management and intervention strategies. This study aims to unlock the predictive power of data collected from personal, clinical, preclinical, and laboratory variables through machine learning (ML) analyses. METHODS: A retrospective study was conducted in 2022 in a large hospital in Abadan, Iran. Data were collected and categorized into demographic, clinical, comorbid, treatment, initial vital signs, symptoms, and laboratory test groups. The collected data were subjected to ML analysis to identify predictive factors associated with COVID-19 mortality. Five algorithms were used to analyze the data set and derive the latent predictive power of the variables by the shapely additive explanation values. RESULTS: Results highlight key factors associated with COVID-19 mortality, including age, comorbidities (hypertension, diabetes), specific treatments (antibiotics, remdesivir, favipiravir, vitamin zinc), and clinical indicators (heart rate, respiratory rate, temperature). Notably, specific symptoms (productive cough, dyspnea, delirium) and laboratory values (D-dimer, ESR) also play a critical role in predicting outcomes. This study highlights the importance of feature selection and the impact of data quantity and quality on model performance. CONCLUSION: This study highlights the potential of ML analysis to improve the accuracy of COVID-19 mortality prediction and emphasizes the need for a comprehensive approach that considers multiple feature categories. It highlights the critical role of data quality and quantity in improving model performance and contributes to our understanding of the multifaceted factors that influence COVID-19 outcomes.

Longitudinal machine learning model for predicting systolic blood pressure in patients with heart failure.

Objective: Systolic blood pressure (SBP) strongly indicates the prognosis of heart failure (HF) patients, as it is closely linked to the risk of death and readmission. Hence, maintaining control over blood pressure is a vital factor in the management of these patients. In order to determine significant variables associated with changes in SBP over time and assess the effectiveness of classical and machine learning models in predicting SBP, this study aimed to conduct a comparative analysis between the two. Methods: This retrospective cohort study involved the analysis of data from 483 patients with HF who were admitted to Farshchian Heart Center located in Hamadan in the west of Iran, and hospitalized at least two times between October 2015 and July 2019. To predict SBP, we utilized a linear mixed-effects model (LMM) and mixed-effects least-square support vector regression (MLS-SVR). The effectiveness of both models was evaluated based on the mean absolute error and root mean squared error. Results: The LMM analysis revealed that changes in SBP over time were significantly associated with sex, body mass index (BMI), sodium, time, and history of hypertension (P-value < 0.05). Furthermore, according to the MLS-SVR analysis, the four most important variables in predicting SBP were identified as history of hypertension, sodium, BMI, and triglyceride. In both the training and testing datasets, MLS-SVR outperformed LMM in terms of performance. Conclusions: Based on our results, it appears that MLS-SVR has the potential to serve as a viable alternative to classical longitudinal models for predicting SBP in patients with HF.

Comparison of Pulsed and Continuous Wave Diode Laser at 940 nm on the Viability and Migration of Gingival Fibroblasts.

Gingival fibroblasts have critical roles in oral wound healing. Photobiomodulation (PBM) has been shown to promote mucosal healing and is now recommended for managing oncotherapy-associated oral mucositis. This study examined the effects of the emission mode of a 940 nm diode laser on the viability and migration of human gingival fibroblasts. Cells were cultured in a routine growth media and treated with PBM (average power 0.1 W cm-2 , average fluence 3 J cm-2 , every 12 h for six sessions) in one continuous wave and two pulsing settings with 20% and 50% duty cycles. Cell viability was assessed using MTT, and digital imaging quantified cell migration. After 48 and 72 h, all treatment groups had significantly higher viability (n = 6, P < 0.05) compared with the control. The highest viability was seen in the pulsed (20% duty cycle) group at the 72-h time point. PBM improved fibroblast migration in all PBM-treated groups, but differences were not statistically significant (n = 2, P > 0.05). PBM treatments can promote cell viability in both continuous and pulsed modes. Further studies are needed to elucidate the optimal setting for PBM-evoked responses for its rationalized use in promoting specific phases of oral wound healing.

Prediction of Hepatitis disease using ensemble learning methods.

Objective: Hepatitis is one of the chronic diseases that can lead to liver cirrhosis and hepatocellular carcinoma, which cause deaths around the world. Hence, early diagnosis is needed to control, treat, and reduce the effects of this disease. This study's main goal was to compare the performance of traditional and ensemble learning methods for predicting hepatitis B virus (HBV), and hepatitis C virus (HCV). Also, important variables related to HBV and HCV were identified. Methods: This case-control study was conducted in Hamadan Province, in the west of Iran, between 2014 to 2019. It included 534 subjects (267 cases and 267 controls). The bagging, random forest, AdaBoost, and logistic regression were used for predicting HBV and HCV. These methods' performance was evaluated using accuracy. Results: According to the results, the accuracy of bagging, random forest, Adaboost, and logistic regression were 0.65 ± 0.03, 0.66 ± 0.03, 0.62 ± 0.04, and 0.64 ± 0.03, respectively, with random forest showing the best performance for predicting HBV. This method showed that ALT was the most important variable for predicting HBV. The the accuracy of random forest was 0.77±0.03 for predicting HCV. Also, the random forest showed that the order of variable importance has belonged to AST, ALT, and age for predicting HCV. Conclusion: This study showed that random forest performed better than other methods for predicting HBV and HCV.

Prediction the survival of patients with breast cancer using random survival forests for competing risks.

Objectives: Breast cancer (BC) is the most common cause of cancer death in Iranian women. Sometimes death from other causes precludes the event of interest and makes the analysis complicated. The purpose of this study was to identify important prognostic factors associated with survival duration among patients with BC using random survival forests (RSF) model in presence of competing risks. Also, its performance was compared with cause-specific hazard model. Methods: This retrospective cohort study assessed 222 patients with BC who were admitted to Ayatollah Khansari hospital in Arak, a major industrial city and the capital of Markazi province in Iran. The cause-specific Cox proportional hazards and RSF models were employed to determine the important risk factors for survival of the patients. Results: The mean and median survival duration of the patients were 90.71 (95%CI: 83.8-97.6) and 100.73 (95%CI: 89.2-121.5) months, respectively. The cause-specific model indicated that type of surgery and HER2 had statistically significant effects on the risk of death of BC. Moreover, the RSF model identified that HER2 was the most important variable for the event of interest. Conclusion: According to the results of this study, the performance of the RSF model was better than the cause-specific hazard model. Moreover, HER2 was the most important variable for death of BC in both of the models.

Predicting Hospital Readmission in Heart Failure Patients in Iran: A Comparison of Various Machine Learning Methods.

OBJECTIVES: Heart failure (HF) is a common disease with a high hospital readmission rate. This study considered class imbalance and missing data, which are two common issues in medical data. The current study's main goal was to compare the performance of six machine learning (ML) methods for predicting hospital readmission in HF patients. METHODS: In this retrospective cohort study, information of 1,856 HF patients was analyzed. These patients were hospitalized in Farshchian Heart Center in Hamadan Province in Western Iran, from October 2015 to July 2019. The support vector machine (SVM), least-square SVM (LS-SVM), bagging, random forest (RF), AdaBoost, and naïve Bayes (NB) methods were used to predict hospital readmission. These methods' performance was evaluated using sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. Two imputation methods were also used to deal with missing data. RESULTS: Of the 1,856 HF patients, 29.9% had at least one hospital readmission. Among the ML methods, LS-SVM performed the worst, with accuracy in the range of 0.57-0.60, while RF performed the best, with the highest accuracy (range, 0.90-0.91). Other ML methods showed relatively good performance, with accuracy exceeding 0.84 in the test datasets. Furthermore, the performance of the SVM and LS-SVM methods in terms of accuracy was higher with the multiple imputation method than with the median imputation method. CONCLUSIONS: This study showed that RF performed better, in terms of accuracy, than other methods for predicting hospital readmission in HF patients.

Effect of photobiomodulation with 810 and 940 nm diode lasers on human gingival fibroblasts.

BACKGROUND: The growth and proliferation of gingival fibroblasts are important in the process of oral wound healing, and photobiomodulation (PBM) might be able to modify this process. OBJECTIVES: The aim of the current study was to evaluate the biomodulatory effect of a single session of laser PBM by means of 810 nm and 940 nm diode lasers alone and their combined application with different fluencies on human gingival fibroblasts (HGFs). MATERIAL AND METHODS: Cells were provided by the Pasteur Institute, the National Cell Bank of Iran (NCBI) (C-165). Laser irradiation was carried out using 810 nm, 940 nm and 810 nm + 940 nm in the continuous wave (CW) mode, 100 mW, and energy densities of 0.5, 1.5 and 2.5 J/cm2. Cell viability was evaluated at 24 h with the MTT assay. Trypan blue staining was used to evaluate proliferation 24, 48 and 72 h after laser therapy. Propidium iodine was used to stain DNA and the cell nucleus. RESULTS: Laser irradiation (810 nm, 0.5 J/cm2) increased the viability of gingival fibroblasts, while this dose had an inhibitory effect with 940 nm. No positive effect on cell viability was found with other settings at 24 h. The viability results were not statistically different from those of the control in the dual wavelength group. At all single-laser irradiation doses, the cell proliferation results were lower as compared to the control at 48 and 72 h. The dual wavelength group results were significantly better than those of the control for the 1.5 J/cm2 and 2.5 J/cm2 energy densities (p < 0.001). Propidium iodine staining showed no negative effect of laser irradiation on the cell nucleus in any of the groups. CONCLUSIONS: Although a single irradiation dose of 810 nm, 0.5 J/cm2, resulted in a positive effect on cell viability at 24 h, no statistically significant stimulatory effect on viability and proliferation was observed for the other single wavelength group. When a combination of the 2 wavelengths was used, better results were observed as compared to the control, which needs to be further investigated in future studies.

Investigating the effect of climatic parameters on mental disorder admissions.

The main objective of this study was to evaluate the role of climatic parameters and phenomena including the monthly number of dusty/rainy/snowy/foggy days, cloudiness (Okta), horizontal visibility, and barometric pressure (millibar) on major depressive disorder, bipolar, schizophrenia, and schizoaffective admissions. The monthly data related to the number of admissions in Farshchian hospital and climatic parameters from March 2005 to March 2017 were extracted. Random forest regression and dynamic negative binomial regression were used to examine the relationship between variables; the statistical significance was considered as 0.05. The number of dusty/rainy/snowy/foggy days, cloudiness, and the number of days with vision less than 2 km had a significant positive relationship with admissions due to schizophrenia (p < 0.05). Barometric pressure had a negative effect on schizophrenia admissions (p < 0.001). The number of dusty/rainy/snowy/foggy days and cloudiness had a significant effect on schizoaffective admissions (p < 0.05). Bipolar admissions were negatively associated with rainy days and positively associated with dusty days and cloudiness (p < 0.05). The number of rainy/dusty/snowy days and cloudiness had a positive significant effect on major depressive disorder admissions. The results of the present study confirmed the importance of climatic parameter variability for major depressive disorder, bipolar, schizophrenia, and schizoaffective admissions.

Evaluation of Bipolar Disorder in Several Relapses over Time Using Generalized Estimating Equations.

Objective: Bipolar disorder is defined as a common and severe chronic disorder that causes several problems in a person's psychosocial functioning. This study aimed at modeling the development of bipolar disorder episodes using its determinant risk factors over time. Method: This retrospective cohort study was conducted in Hamadan province, the West of Iran, from April 2008 to September 2014. In this study, 124 patients with bipolar disorder (both Type I and Type II) participated. All patients had experienced 4 relapses. Generalized Estimating Equation (GEE) was used to model bipolar disorder episodes, and significance level was set at 0.05. Results: The mean (±SD) age of the patients was 33.2 (±11.55). Males were more likely to experience mania than depression compared to females (odds ratio = 2.30, 95% CI (1.37-3.86)). Patients who received psychotherapy plus medicine were less likely to experience mania than depression compared to drug receivers (odds ratio = 0.39, 95% CI (0.18-0.88)). In the spring, patients were more likely to experience mania than depression compared to the winter (OR = 2.22, 95% CI (1.18-4.19)). Conclusion: The results of the present study revealed that among bipolar disorder patients in the West of Iran, mania was much more prevalent than depression and mixed episodes. Moreover, it was found that sex, treatment, and season can determine the episodes of bipolar disorder.

Influence Factors on Birth Spacing and Childbearing Rates using Survival Recurrent Events Model and Parity Progression Ratios.

BACKGROUND: Reproduction rates are declining in Hamadan, western Iran. This study identified the influence factors associated with reproduction rate and birth spacing in Hamadan as an area of low population growth in Iran. STUDY DESIGN: A cross sectional study. METHODS: The study considered reproductive status of 812 women referred to health care centers of Hamadan, western Iran in 2015. Data were obtained through frequency and percentage. PWP-GT model was used to determine the influence factors on women's reproductive by R software (version 3.3.2). The parity progression ratios were determined based on Yadava and Kumar. RESULTS: The median time for the second, the third and the fourth birth was 4.53, 4.65 and 5.27 yr, respectively. PWP-GT model showed that women age at marriage (P=0.001), women's (P=0.005) and their husband's (P=0.039) employment had significant effect on time birth of first child. The women's education (P=0.001) was the only variable that had a significant effect on the birth time from the first to the second child as well as from the third to the fourth child. Education of women (P=0.001) and their husbands (P=0.034) had significant effect on the time interval from the second to the third child. The birth probability from marriage to the first birth (0 to 1st child) was high (0.91), while the probability had been fallen from the third birth and more (0.31). CONCLUSIONS: Birth spacing in Hamadan is more than WHO recommended. In addition, reducing of the population growth and childbearing has started in the past few decades in Hamadan the same as Iran. The chance of fertility has dramatically declined from the third child and then.

Recurrence in Patients with Bipolar Disorder and Its Risk Factors.

Objective: The aim of this study was to identify prognosis factors associated with recurrence in patients with bipolar disorder. Method: This retrospective cohort study was conducted in Hamadan Province, the west of Iran. All patients (n = 400) with bipolar disorder who were hospitalized for the second time or more during April 2008 to September 2014 were included in this study. Ordinal logistic regression analysis was employed to determine the effective factors in each recurrence, and odds ratio (OR) and 95% confidence intervals (CI) were obtained. Results: The mean (SD) age of the participants at the entrance to the study was 34.62 (11.68) years. There was an association between recurrence and type of bipolar disorder (P = 0.033). The OR of recurrence was 0.28 (95% CI: 0.09, 0.90) for bipolar disorder II; 0.35 (95% CI: 0.13, 0.92) for the patients with college education; 0.39 (95% CI: 0.25, 0.60) for employed patients; 0.55 (95% CI: 0.35, 0.87) for patients who received both drugs and electroconvulsive therapy, and 1.89 (95% CI: 1.23, 2.92) for patients who stopped using drugs. In addition, a non-significant association was found between recurrence and age, sex, marital status, place of residence, season, mood classification and family history of mood disorder. Conclusion: Type of bipolar disorder and cessation of medication were the leading causes of an increase in the relapse of the disease. Furthermore, patients who received both drugs and electroconvulsive therapy had a fewer risk of recurrence. .