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

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.

Prediction of Serum Creatinine in Hemodialysis Patients Using a Kernel Approach for Longitudinal Data / 대한의료정보학회지

ObjectivesLongitudinal data are prevalent in clinical research; due to their correlated nature, special analysis must be used for this type of data. Creatinine is an important marker in predicting end-stage renal disease, and it is recorded longitudinally. This study compared the prediction performance of linear regression (LR), linear mixed-effects model (LMM), least-squares support vector regression (LS-SVR), and mixed-effects least-squares support vector regression (MLS-SVR) methods to predict serum creatinine as a longitudinal outcome.MethodsWe used a longitudinal dataset of hemodialysis patients in Hamadan city between 2013 and 2016. To evaluate the performance of the methods in serum creatinine prediction, the data was divided into two sets of training and testing samples. Then LR, LMM, LS-SVR, and MLS-SVR were fitted. The prediction performance was assessed and compared in terms of mean squared error (MSE), mean absolute error (MAE), mean absolute prediction error (MAPE), and determination coefficient (R2). Variable importance was calculated using the best model to select the most important predictors.ResultsThe MLS-SVR outperformed the other methods in terms of the least prediction error; MSE = 1.280, MAE = 0.833, and MAPE = 0.129 for the training set and MSE = 3.275, MAE = 1.319, and MAPE = 0.159 for the testing set. Also, the MLS-SVR had the highest R2, 0.805 and 0.654 for both the training and testing samples, respectively. Blood urea nitrogen was the most important factor in the prediction of creatinine.ConclusionsThe MLS-SVR achieved the best serum creatinine prediction performance in comparison to LR, LMM, and LS-SVR.

Correlations between anatomical variations of the nasal cavity and ethmoidal sinuses on cone-beam computed tomography scans

PURPOSE: Anatomical variations of the external nasal wall are highly important, since they play a role in obstruction or drainage of the ostiomeatal complex and ventilation and can consequently elevate the risk of pathological sinus conditions. This study aimed to assess anatomical variations of the nasal cavity and ethmoidal sinuses and their correlations on cone-beam computed tomography (CBCT) scans. MATERIALS AND METHODS: This cross-sectional study evaluated CBCT scans of 250 patients, including 107 males and 143 females, to determine the prevalence of anatomical variations of the nasal cavity and ethmoidal sinuses. All images were taken using a New Tom 3G scanner. Data were analyzed using the chi-square test, Kruskal-Wallis test, and the Mann-Whitney test. RESULTS: The most common anatomical variations were found to be nasal septal deviation (90.4%), agger nasi air cell (53.6%), superior orbital cell (47.6%), pneumatized nasal septum (40%), and Onodi air cell (37.2%). Correlations were found between nasal septal deviation and the presence of a pneumatized nasal septum, nasal spur, and Haller cell. No significant associations were noted between the age or sex of patients and the presence of anatomical variations (P>0.05). CONCLUSION: Radiologists and surgeons must pay close attention to the anatomical variations of the sinonasal region in the preoperative assessment to prevent perioperative complications.

effect of cognitive behavioral therapy on marital quality among women

Background: Marital quality reflects the individual's overall evaluation of marital relationship. The aim of study was examine the effect of cognitive behavioral counseling on marital quality among women

Materials and Methods: The experimental study was a randomized clinical trial with two groups, on 198 qualified women who referred to selected health care centers in Hamadan, Iran in 2016. The intervention participants attended four 90-minute sessions of cognitive behavioral counseling. Demographic information questionnaire and marital quality scale [Revised Dyadic Adjustment Scale [RDAS] were completed by the two groups before and after the intervention. To perform the comparisons, t test, Chi-square test and Fisher's test, Logistic Regression and covariance analysis were used. Covariance analysis or change analysis were employed. Statistical analysis was done using SPSS Software, version 21.0. The significance level was set at 5% [P<0.05]

Results: According to the results of the present study, the mean age in the control group and the intervention group was 23.58 +/- 7.54 and 35.04 +/- 7.91 years old, respectively. Covariance analysis was utilized to examine the marital quality scores. In this analysis, after modification of the variables of age, marital quality score of agreement and satisfaction before the intervention, and income status, the total marital quality score experienced a significant change in all dimensions [P<0.05] and the mean scores increased remarkably. Moreover, according to the cut-off point of the dimensions, the scores of all dimensions increased remarkably and the proportion of individuals with high marital quality before and after the intervention changed significantly [P<0.05]

Conclusion: Due to the role of sexual relations in stabilizing marriage, cognitive behavioral consultation was effective in improving marital quality especially after agreement and can be used in health care centers in order to improve the relationship between couples and reduce divorce rates

Diabetic peripheral neuropathy class prediction by multicategory support vector machine model: a cross-sectional study / 한국역학회지

OBJECTIVES: Diabetes is increasing in worldwide prevalence, toward epidemic levels. Diabetic neuropathy, one of the most common complications of diabetes mellitus, is a serious condition that can lead to amputation. This study used a multicategory support vector machine (MSVM) to predict diabetic peripheral neuropathy severity classified into four categories using patients' demographic characteristics and clinical features. METHODS: In this study, the data were collected at the Diabetes Center of Hamadan in Iran. Patients were enrolled by the convenience sampling method. Six hundred patients were recruited. After obtaining informed consent, a questionnaire collecting general information and a neuropathy disability score (NDS) questionnaire were administered. The NDS was used to classify the severity of the disease. We used MSVM with both one-against-all and one-against-one methods and three kernel functions, radial basis function (RBF), linear, and polynomial, to predict the class of disease with an unbalanced dataset. The synthetic minority class oversampling technique algorithm was used to improve model performance. To compare the performance of the models, the mean of accuracy was used. RESULTS: For predicting diabetic neuropathy, a classifier built from a balanced dataset and the RBF kernel function with a one-against-one strategy predicted the class to which a patient belonged with about 76% accuracy. CONCLUSIONS: The results of this study indicate that, in terms of overall classification accuracy, the MSVM model based on a balanced dataset can be useful for predicting the severity of diabetic neuropathy, and it should be further investigated for the prediction of other diseases.

Application of censored quantile regression to determine overall survival related factors in breast cancer

Background: the occurrence and the mortality related to breast cancer [BC] in Iranian female population has increased over time. Although there are many studies on BC and related risk factors, however, the epidemiological aspects of this melanoma in Iranian females are uncertain. Therefore, the aim of this study was to determine the relationship between demographical and clinical factors on the shape of overall survival [OS] distribution in patients with BC

Methods: this historical cohort study was carried out using data from 522 participants with BC. Data were gathered from medical records of these patients admitted to Mahdieh Oncology Center of Hamadan Province, western Iran, from January 2000 to August 2011. Kaplan-Meier estimator was used to estimate the survival rates and, censored quantile regression [CQR] to provide in-depth insight in the multivariable association between prognosis factors and survival rates

Results: patients' follow-up ranged from around 3 to 197 months. One-, three-, and five-year survival rates were 90%, 73% and 62.5%, respectively. Results of CQR model showed that change in the age at diagnosis, number of involved lymph nodes and tumor size could significantly change the median and some other quantiles of OS

Conclusions: this study, confirm the importance of early detection of BC and usefulness of CQR because of possible changes in distribution family of survival time

Estimating the potential impact fraction of hypertension as the main risk factor of stroke: application of the distribution shift method

Few published studies have assessed the impact of quantitative risk factors such as high blood pressure on stroke. The aim of this study was to quantify the potential impact fraction [PIF] of hypertension on stroke in Hamadan Province, western Iran. Avoidable burden of stroke associated with high blood pressure was calculated using distribution shift at different scenarios. Data on the prevalence of high blood pressure among residents of Hamadan province older than 19 years were extracted from non-communicable diseases risk factors surveillance system in 2009. Five mmHg hypothetical reduction in systolic blood pressure above 140 mmHg, leads to 3.5% [PIF = 0.035] reduction in the total burden to stroke. This value may reach 7%, if systolic blood pressure decreases 10 mmHg. In addition, 5 mmHg hypothetical reduction in diastolic blood pressure above 82 mmHg, leads to 4.87% reduction in the total burden to stroke. PIF more than 10 mmHg modification on distribution of diastolic blood pressure was estimated as 9.38%. According to these findings, policy makers are advised to implement interventions on hypertension based on the distribution shift method rather than the proportion shift one

Muscular response of females with Kyphosis in balance recovery from Postural Perturbation

Purpose: This study aimed to determine muscular response in balance recovery after postural perturbation among females with increased thoracic kyphosis

Methods: This is a quasi-experimental study. A total of 12 female students with thoracic kyphosis [>40 degrees] and 12 matched controls were selected to participate in the study. Each subject underwent unexpected perturbation from anterior, posterior, right, and left directions with eyes open and eyes closed conditions while standing on a movable plate, triggered by a weight equivalent to 10% of the subjects' body weight. Then, the electromyography activity of some selected muscles was measured. SPSS version 19.0 and Analysis of variance [ANOVAs] with repeated measures were used for data analyses [P<0.05]

Results: Significant effects after perturbation were observed with regard to the activation of rectus femoris [P<0.001], biceps femoris [P<0.001], tibialis anterior [P<0.005], gastrocnemius medialis [P=0.032], and gluteus maximus [P<0.001]. Whereas, no impact was seen in activities of erector spinae [P=0.461] and rectus abdominal muscles [P=0.224]. Also the results showed increase in muscles activities when the eyes is closed compared to open eyes condition [P<0.05]. Activities of all muscles were significantly higher in kyphotic group in comparison with control group. However, these differences were statistically significant regarding gastrocnemius medialis [P=0.007], gluteus maximus [P=0.033], and rectus abdominus [P=0.010]

Conclusion: Proximal muscles activity was higher in kyphotic subjects than normal subjects during balance recovery after postural perturbation

Evaluation of an oral health intervention among mothers of young children: a clustered randomized trial

Oral health education for the mothers of very young children is important in reducing the risk of early childhood caries. This study aimed to evaluate the impact of an oral health intervention among mothers of 1-2 years old children. This cluster randomized controlled trial [2012] was conducted among ninety mothers of 1- 2 year old children. The setting of study was 10 child day-care centers out of 18 in Hamadan, western Iran. Day-care centers were randomly allocated into two groups: an intervention group [5 day-care centers, 45 mothers] and a control group [5 day-care centers, 45 mothers]. Intervention consisted of three sessions, a booklet, and mobile phone text-message reminders. The primary outcome was change in cleaning the children' teeth, while the secondary outcomes were changes in Theory of Planned Behavior [TPB] cognitions. Questionnaires at baseline, 10 days, and 3 months assessed intervention effects. Data were analyzed using SPSS v.16. T tests, chi- square, and logistic and linear generalized estimating equations [GEE] regression were used to test intervention impact. At 10-day assessment, mothers in intervention group reported a significant difference in knowledge [P=0.001], attitude [P=0.004], perceived behavioral control [P=0.008], and cleaning of children's teeth [P=0.011]. Also, at 3-month assessment compared to control group, the mothers in intervention group significantly improved in scores of knowledge [P=0.001], attitude [P=0.001], perceived behavioral control [P=0.001], and cleaning of children's teeth [P=0.001]. However, the effect sizes were small to medium and ranged from 0.1 to 0.4 for all cognitions except knowledge [effect size>0.70]. Generalized estimating equations [GEE] showed that score of attitude and perceived behavioral control of intervention group improved between the two post-test assessments. A brief multicomponent theory-based intervention among mothers of 1-2 years old children was effective moderately in improving cognitions and self-reported cleaning children's teeth

Position of impacted mandibular third molar in different skeletal facial types: First radiographic evaluation in a group of Iranian patients

PURPOSE: This study was performed to evaluate the position of impacted mandibular third molars in different skeletal facial types among a group of Iranian patients. MATERIALS AND METHODS: A total of 400 mandibular third molars in 200 subjects with different types of facial growth were radiographically investigated for their positions according to their types of facial growth on the basis of the beta angle. The subjects were divided into three groups (class I, II, and III) according to ANB angle, representing the anteroposterior relationship of the maxilla to the mandible. Meanwhile, the subjects were also divided into three groups (long, normal, and short face) according to the angle between the stella-nasion and mandibular plane (SNGoGn angle). ANOVA was used for statistical analysis. RESULTS: The mean beta angle showed no significant difference among class I, II, and III malocclusions (df=2, F=0.669, p=0.513). The same results were also found in short, normal, and long faces (df=1.842, F=2, p=0.160). The mesioangular position was the most frequent one in almost all of the facial growth patterns. Distoangular and horizontal positions of impaction were not found in the subjects with class III and normal faces. In the long facial growth pattern, the frequency of vertical and distoangular positions were not different. CONCLUSION: In almost all of the skeletal facial types, the mesioangular impaction of the mandibular third molar was the most prevalent position, followed by the horizontal position. In addition, beta angle showed no significant difference in different types of facial growth.

Survival analysis of breast cancer patients using cox and frailty models

Cox proportional hazard [CPH] model is the most widely used model for survival analysis. When there are unobserved/unmeasured individuals factor, then the results of the Cox proportional hazard model may not be reliable. The purpose of this study was to compare the results of CPH and frailty models in breast cancer [BC] patients. A historical cohort study was carried out using medical records gathered from the Fars Province Cancer Registry. The dataset consisted of 769 women having BC referred to Shiraz Namazi Hospital, south of Iran. These patients had been followed for 6 years. After selecting the most important prognostic risk factors on survival, CPH and gamma-frailty Cox models were used to estimate the effects of the risk factors. The results of CPH model showed that, tumor characteristics and number of involved lymph nodes increase the mortality hazard of BC [P < 0.05]. In addition, the frailty model showed that there is at least a latent factor in the model [P = 0.005]. Both of the frailty and CPH model emphasis that the early detection of BC improves survival in BC patients.

Association between physical activity and mental health among high-school adolescents in Boushehr province: a population based study

Mental health is one of the evaluating factors of community indicators, and physical activity is considered an important tool for the importance of public health. Previous studies have demonstrated associations between physical activity and mental health, but these studies did not include those populations in which training children have some traditional and religious aspects. Multiple logistic regression analyses were used to compare the prevalence of mental health among those who had inactive, minimally and HEPA activity in a representative sample of adolescents aged 15-19 in South of Iran using data from the Mental Health Survey [n = 2584]. The GHQ-28 and IPAQ-short forms were used to evaluate the mental health and physical activity, respectively. A total of 2584 adolescents [1401 male and 1178 female] participated in the study. The observed odds of psychological symptoms in boys compared to girls is 1.2 times [p=0.018]. We observed that HEPA-activity decreases odds of somatic distress and social dysfunction compared with inactivity [p=0.031 and 0.001, respectively]; minimally activity decreases odds of anxiety compared with inactivity [p=0.038]; but physical activity rate was not affected on odds of adolescents' depression [p>0.05]. Physical activity decreases mental health subscales except for depression among adolescents in Boushehr, southern city of Iran

[Comparing three adjuvant chemotherapy regimes after modified radical mastectomy in breast cancer patients using log-logistic model]

Chemotherapy after surgery is one of the adjuvant therapy components in breast cancer patients. In the current study, Taxane-based and Anthracycline-based regimes were compared to Cyclophosphamide, Methotrexate and 5-Fluorouracil [CMF] chemotherapy after modified radical mastectomy [MRM] surgery in terms of disease-free survival [DFS]. Sixty-two patients with confirmed breast cancer that underwent MRM surgery at no metastasis stage [M0] enrolled. These patients were received one of the three types of adjuvant chemotherapies that mentioned above and the relapse/metastasis in 8-year follow-up was recorded. In addition, some paraclinical and demographical factors were recorded and the efficacy of the different chemotherapy regimes was compared using log-logistic survival model. The mean age [SD] of patients was 49 [1.31]. More than 40% of patients had tumors with grade 3. The median time of follow-up was 20.0 months and the rate of 5-year DFS was 48%. Survival analysis indicated that type of chemotherapy [OR [Taxane/CMF] =3.676; OR [Anthracycline/CMF]=3.185], histological grade [OR=2.528], tumor size [OR=5.598] and number of lymph nodes [OR=1.112] affect the DFS. The results of the current study indicated that the efficacy of Taxane-based and Anthracycline-based is more than CMF. This study also showed that early detection of tumor was an important factor for a successful treatment in terms of DFS, and so, it declares the roll of screening and public education