Changes in cervical elastography, cervical length and endocervical canal width after cerclage for cervical insufficiency: an observational ultrasound study.

BACKGROUND: We previously demonstrated that pregnant women with a history of cervical insufficiency had a softer anterior cervical lip, shorter cervical length and wider endocervical canal in the first trimester. The aim of this study was to investigate changes in cervical elastography, cervical length, and endocervical canal width in the second trimester after cerclage, and further discuss whether these ultrasound parameters are predictive of preterm delivery. METHODS: This was a secondary analysis of cervical changes in singleton pregnancies after cerclage from January 2016 to June 2018. Cervical elastography, cervical length, and endocervical canal width were measured during the second trimester in the cervical insufficiency group and control group without cervical insufficiency. Strain elastography under transvaginal ultrasound was used to assess cervical stiffness and presented as percentage (strain rate). RESULTS: Among the 339 pregnant women enrolled, 24 had a history of cervical insufficiency and underwent cerclage. Both anterior and posterior cervical lips were significantly softer in the cervical insufficiency group even though they received cerclage (anterior strain rate: 0.18 ± 0.06% vs. 0.13 ± 0.04%; P = 0.001; posterior strain rate: 0.11 ± 0.03% vs. 0.09 ± 0.04%; P = 0.017). Cervical length was also shorter in the cervical insufficiency group (36.3 ± 3.6 mm vs. 38.3 ± 4.6 mm; P = 0.047). However, there was no significant difference in endocervical canal width between the two groups (5.4 ± 0.7 mm vs. 5.6 ± 0.7 mm; P = 0.159). Multivariate logistic regression analysis also revealed significant differences in anterior cervical lip strain rate (adjusted odds ratio [OR], 7.32, 95% confidence interval [CI], 1.70-31.41; P = 0.007), posterior cervical lip strain rate (adjusted OR, 5.22, 95% CI, 1.42-19.18; P = 0.013), and cervical length (adjusted OR, 3.17, 95% CI,1.08-9.29; P = 0.035). Among the four ultrasound parameters, softer anterior cervical lip (P = 0.024) and shorter cervical length (P < 0.001) were significantly related to preterm delivery. CONCLUSIONS: Cervical cerclage can prevent widening of the endocervical canal, but not improve cervical elasticity or cervical length. Measuring anterior cervical elastography and cervical length may be valuable to predict preterm delivery.

The association of heat shock protein genetic polymorphisms with age-related hearing impairment in Taiwan.

BACKGROUND: Age-related hearing impairment (ARHI) is a major disability among the elderly population. Heat shock proteins (HSPs) were found to be associated with ARHI in animal studies. The aim of this study was to analyze the associations of single nucleotide polymorphisms (SNPs) of HSP genes with ARHI in an elderly population in Taiwan. METHODS: Participants ≥65 years of age were recruited for audiometric tests and genetic analyses. The pure tone average (PTA) of the better hearing ear was calculated for ARHI evaluation. The associations of HSPA1L (rs2075800 and rs2227956), HSPA1A (rs1043618) and HSPA1B (rs2763979) with ARHI were analyzed in 146 ARHI-susceptible (cases) and 146 ARHI-resistant (controls) participants. RESULTS: The "T" allele of HSPA1B rs2763979 showed a decreased risk of ARHI. The "TT" genotype of rs2763979 also showed a decreased risk of ARHI in the dominant hereditary model. For HSPA1L (rs2075800 and rs2227956) and HSPA1A (rs1043618), the haplotype "CAG" was related to a decreased risk of ARHI. CONCLUSION: These findings suggest that HSP70 polymorphisms are associated with susceptibility to ARHI in the elderly population.

The Performance of Different Artificial Intelligence Models in Predicting Breast Cancer among Individuals Having Type 2 Diabetes Mellitus.

Objective: Early reports indicate that individuals with type 2 diabetes mellitus (T2DM) may have a greater incidence of breast malignancy than patients without T2DM. The aim of this study was to investigate the effectiveness of three different models for predicting risk of breast cancer in patients with T2DM of different characteristics. Study design and methodology: From 2000 to 2012, data on 636,111 newly diagnosed female T2DM patients were available in the Taiwan's National Health Insurance Research Database. By applying their data, a risk prediction model of breast cancer in patients with T2DM was created. We also collected data on potential predictors of breast cancer so that adjustments for their effect could be made in the analysis. Synthetic Minority Oversampling Technology (SMOTE) was utilized to increase data for small population samples. Each datum was randomly assigned based on a ratio of about 39:1 into the training and test sets. Logistic Regression (LR), Artificial Neural Network (ANN) and Random Forest (RF) models were determined using recall, accuracy, F1 score and area under the receiver operating characteristic curve (AUC). Results: The AUC of the LR (0.834), ANN (0.865), and RF (0.959) models were found. The largest AUC among the three models was seen in the RF model. Conclusions: Although the LR, ANN, and RF models all showed high accuracy predicting the risk of breast cancer in Taiwanese with T2DM, the RF model performed best.

Predicting weaning difficulty for planned extubation patients with an artificial neural network.

This study aims to construct a neural network to predict weaning difficulty among planned extubation patients in intensive care units.This observational cohort study was conducted in eight adult ICUs in a medical center about adult patients experiencing planned extubation.The data of 3602 patients with planned extubation in ICUs of Chi-Mei Medical Center (from Dec. 2009 through Dec. 2011) was used to train and test an artificial neural network (ANN) model. The input features contain 47 clinical risk factors and the outputs are classified into three categories: simple, difficult, and prolonged weaning. A deep ANN model with four hidden layers of 30 neurons each was developed. The accuracy is 0.769 and the area under receiver operating characteristic curve for simple weaning, prolonged weaning, and difficult weaning are 0.910, 0.849, and 0.942 respectively.The results revealed that the ANN model achieved a good performance in prediction the weaning difficulty in planned extubation patients. Such a model will be helpful for predicting ICU patients' successful planned extubation.

Artificial Intelligence Prediction Model for the Cost and Mortality of Renal Replacement Therapy in Aged and Super-Aged Populations in Taiwan.

BACKGROUND: Prognosis of the aged population requiring maintenance dialysis has been reportedly poor. We aimed to develop prediction models for one-year cost and one-year mortality in aged individuals requiring dialysis to assist decision-making for deciding whether aged people should receive dialysis or not. METHODS: We used data from the National Health Insurance Research Database (NHIRD). We identified patients first enrolled in the NHIRD from 2000-2011 for end-stage renal disease (ESRD) who underwent regular dialysis. A total of 48,153 Patients with ESRD aged ≥65 years with complete age and sex information were included in the ESRD cohort. The total medical cost per patient (measured in US dollars) within one year after ESRD diagnosis was our study's main outcome variable. We were also concerned with mortality as another outcome. In this study, we compared the performance of the random forest prediction model and of the artificial neural network prediction model for predicting patient cost and mortality. RESULTS: In the cost regression model, the random forest model outperforms the artificial neural network according to the mean squared error and mean absolute error. In the mortality classification model, the receiver operating characteristic (ROC) curves of both models were significantly better than the null hypothesis area of 0.5, and random forest model outperformed the artificial neural network. Random forest model outperforms the artificial neural network models achieved similar performance in the test set across all data. CONCLUSIONS: Applying artificial intelligence modeling could help to provide reliable information about one-year outcomes following dialysis in the aged and super-aged populations; those with cancer, alcohol-related disease, stroke, chronic obstructive pulmonary disease (COPD), previous hip fracture, osteoporosis, dementia, and previous respiratory failure had higher medical costs and a high mortality rate.

A novel atrial fibrillation prediction model for Chinese subjects: a nationwide cohort investigation of 682 237 study participants with random forest model.

AIMS: We aimed to construct a random forest model to predict atrial fibrillation (AF) in Chinese population. METHODS AND RESULTS: This study was comprised of 682 237 subjects with or without AF. Each subject had 19 features that included the subjects' age, gender, underlying diseases, CHA2DS2-VASc score, and follow-up period. The data were split into train and test sets at an approximate 9:1 ratio: 614 013 data points were placed into the train set and 68 224 data points were placed into the test set. In this study, weighted average F1, precision, and recall values were used to measure prediction model performance. The F1, precision, and recall values were calculated across the train set, the test set, and all data. The area under receiving operating characteristic (ROC) curve was also used to evaluate the performance of the prediction model. The prediction model achieved a k-fold cross-validation accuracy of 0.979 (k = 10). In the test set, the prediction model achieved an F1 value of 0.968, precision value of 0.958, and recall value of 0.979. The area under ROC curve of the model was 0.948 (95% confidence interval 0.947-0.949). This model was validated with a separate dataset. CONCLUSIONS: This study showed a novel AF risk prediction scheme for Chinese individuals with random forest model methodology.

A fitting machine learning prediction model for short-term mortality following percutaneous catheterization intervention: a nationwide population-based study.

BACKGROUND: A suitable multivariate predictor for predicting mortality following percutaneous coronary intervention (PCI) remains undetermined. We used a nationwide database to construct mortality prediction models to find the appropriate model. METHODS: Data were analyzed from the Taiwan National Health Insurance Research Database (NHIRD) covering the period from 2004 to 2013. The study cohort was composed of 3,421 patients with acute myocardial infarction (AMI) diagnosis undergoing PCI. The dataset of enrolled patients was used to construct multivariate prediction models. Of these, 3,079 and 342 patients were included in the training and test groups, respectively. Each patient had 22 input features and 2 output features that represented mortality. This study implemented an artificial neural network model (ANN), a decision tree (DT), a linear discriminant analysis classifier (LDA), a logistic regression model (LR), a naïve Bayes classifier (NB), and a support vector machine (SVM) to predict post-PCI patient mortality. RESULTS: The DT model was found to be the most suitable in terms of performance and real-world applicability. The DT model achieved an area under receiving operating characteristic of 0.895 (95% confidence interval: 0.865-0.925), F1 of 0.969, precision of 0.971, and recall of 0.974. CONCLUSIONS: The DT model constructed using data from the NHIRD exhibited effective 30-day mortality prediction for patients with AMI following PCI.

Development of a prediction model for pancreatic cancer in patients with type 2 diabetes using logistic regression and artificial neural network models.

OBJECTIVES: Patients with type 2 diabetes (T2DM) are suggested to have a higher risk of developing pancreatic cancer. We used two models to predict pancreatic cancer risk among patients with T2DM. METHODS: The original data used for this investigation were retrieved from the National Health Insurance Research Database of Taiwan. The prediction models included the available possible risk factors for pancreatic cancer. The data were split into training and test sets: 97.5% of the data were used as the training set and 2.5% of the data were used as the test set. Logistic regression (LR) and artificial neural network (ANN) models were implemented using Python (Version 3.7.0). The F 1, precision, and recall were compared between the LR and the ANN models. The areas under the receiver operating characteristic (ROC) curves of the prediction models were also compared. RESULTS: The metrics used in this study indicated that the LR model more accurately predicted pancreatic cancer than the ANN model. For the LR model, the area under the ROC curve in the prediction of pancreatic cancer was 0.727, indicating a good fit. CONCLUSION: Using this LR model, our results suggested that we could appropriately predict pancreatic cancer risk in patients with T2DM in Taiwan.

Comparison of machine learning models for the prediction of mortality of patients with unplanned extubation in intensive care units.

Unplanned extubation (UE) can be associated with fatal outcome; however, an accurate model for predicting the mortality of UE patients in intensive care units (ICU) is lacking. Therefore, we aim to compare the performances of various machine learning models and conventional parameters to predict the mortality of UE patients in the ICU. A total of 341 patients with UE in ICUs of Chi-Mei Medical Center between December 2008 and July 2017 were enrolled and their demographic features, clinical manifestations, and outcomes were collected for analysis. Four machine learning models including artificial neural networks, logistic regression models, random forest models, and support vector machines were constructed and their predictive performances were compared with each other and conventional parameters. Of the 341 UE patients included in the study, the ICU mortality rate is 17.6%. The random forest model is determined to be the most suitable model for this dataset with F1 0.860, precision 0.882, and recall 0.850 in the test set, and an area under receiver operating characteristic (ROC) curve of 0.910 (SE: 0.022, 95% CI: 0.867-0.954). The area under ROC curves of the random forest model was significantly greater than that of Acute Physiology and Chronic Health Evaluation (APACHE) II (0.779, 95% CI: 0.716-0.841), Therapeutic Intervention Scoring System (TISS) (0.645, 95% CI: 0.564-0.726), and Glasgow Coma scales (0.577, 95%: CI 0.497-0.657). The results revealed that the random forest model was the best model to predict the mortality of UE patients in ICUs.

Development of a Prediction Model for Colorectal Cancer among Patients with Type 2 Diabetes Mellitus Using a Deep Neural Network.

OBJECTIVES: Observational studies suggested that patients with type 2 diabetes mellitus (T2DM) presented a higher risk of developing colorectal cancer (CRC). The current study aims to create a deep neural network (DNN) to predict the onset of CRC for patients with T2DM. METHODS: We employed the national health insurance database of Taiwan to create predictive models for detecting an increased risk of subsequent CRC development in T2DM patients in Taiwan. We identified a total of 1,349,640 patients between 2000 and 2012 with newly diagnosed T2DM. All the available possible risk factors for CRC were also included in the analyses. The data were split into training and test sets with 97.5% of the patients in the training set and 2.5% of the patients in the test set. The deep neural network (DNN) model was optimized using Adam with Nesterov's accelerated gradient descent. The recall, precision, F1 values, and the area under the receiver operating characteristic (ROC) curve were used to evaluate predictor performance. RESULTS: The F1, precision, and recall values of the DNN model across all data were 0.931, 0.982, and 0.889, respectively. The area under the ROC curve of the DNN model across all data was 0.738, compared to the ideal value of 1. The metrics indicate that the DNN model appropriately predicted CRC. In contrast, a single variable predictor using adapted the Diabetes Complication Severity Index showed poorer performance compared to the DNN model. CONCLUSIONS: Our results indicated that the DNN model is an appropriate tool to predict CRC risk in patients with T2DM in Taiwan.

An Artificial Neural Network Model for Predicting Successful Extubation in Intensive Care Units.

BACKGROUND: Successful weaning from mechanical ventilation is important for patients in intensive care units (ICUs). The aim was to construct neural networks to predict successful extubation in ventilated patients in ICUs. METHODS: Data from 1/12/2009 through 31/12/2011 of 3602 patients with planned extubation in Chi-Mei Medical Center's ICUs was used to train and test an artificial neural network (ANN). The input was 37 clinical risk factors, and the output was a failed extubation prediction. RESULTS: One hundred eighty-five patients (5.1%) had a failed extubation. Multivariate analyses revealed that failure was positively associated with therapeutic intervention scoring system (TISS) scores (odds ratio [OR]: 1.814; 95% Confidence Interval [CI]: 1.283⁻2.563), chronic hemodialysis (OR: 12.264; 95% CI: 8.556⁻17.580), rapid shallow breathing (RSI) (OR: 2.003; 95% CI: 1.378⁻2.910), and pre-extubation heart rate (OR: 1.705; 95% CI: 1.173⁻2.480), but negatively associated with pre-extubation PaO2/FiO2 (OR: 0.529; 95%: 0.370⁻0.750) and maximum expiratory pressure (MEP) (OR: 0.610; 95% CI: 0.413⁻0.899). A multilayer perceptron ANN model with 19 neurons in a hidden layer was developed. The overall performance of this model was F1: 0.867, precision: 0.939, and recall: 0.822. The area under the receiver operating characteristic curve (AUC) was 0.85, which is better than any one of the following predictors: TISS: 0.58 (95% CI: 0.54⁻0.62; p < 0.001); 0.58 (95% CI: 0.53⁻0.62; p < 0.001); and RSI: 0.54 (95% CI: 0.49⁻0.58; p = 0.097). CONCLUSIONS: The ANN performed well when predicting failed extubation, and it will help predict successful planned extubation.

Comparisons of changes in the adapted diabetes complications severity index and CHA2DS2-VASc score for atrial fibrillation risk stratification in patients with type 2 diabetes mellitus: A nationwide cohort study.

PURPOSE: This study describes the risk prediction of atrial fibrillation (AF) after incident type 2 diabetes mellitus (DM) with either progression of adapted diabetes complications severity index (DCSI) or CHA2DS2-VASc score in a large registry from Taiwan. METHODS: The authors performed a retrospective nationwide cohort study by analyzing a Longitudinal Health Insurance Dataset, observing the ability of dynamic adapted DSCI and CHA2DS2-VASc score for AF risk discrimination in type 2 diabetic patients. The predictive performance of changes in the adapted DCSI and CHA2DS2-VASc score with regard to AF events was assessed using area under the curve of receiver operating characteristics (AUROC); and the difference between them was examined using the Delong test. RESULTS: A total of 81,655 new-onset type 2 DM patients were included in the study cohort. Among them, the AUROC for the adapted DCSI change in predicting AF (0.79, 95% CI = 0.78-0.80) was significantly higher than the change in CHA2DS2-VASc score (0.63, 95% CI = 0.62-0.64) with the DeLong test P < 0.001. CONCLUSIONS: Adapted DCSI change significantly outperforms the progression of CHA2DS2-VASc score with regard to AF prediction in type 2 diabetic patients.

Perception of hearing impairment and the willingness to use hearing aids in an elderly population in southern Taiwan: A community-based study.

OBJECTIVES: To estimate the prevalence and severity of hearing impairment (HI), the self-perception of HI, and the willingness to use a hearing aid (HA) in the elderly population in southern Taiwan. DESIGN: This community-based study was performed in a metropolitan hospital. A questionnaire about the perception of HI and the willingness to use a HA was used. The severity of HI in speech-frequency pure-tone average (PTA) was evaluated. The associations between sex, age, severity of HI, self-perception of HI, and the willingness to use a HA were analysed. STUDY SAMPLE: A total of 599 volunteers were recruited from the health management center; 324 (54.1%) males and 275 (45.9%) females, who were 65 years of age or older. RESULTS: The prevalence of HI >25 dBHL in the elderly was 78%. The predicted levels for elderly persons to perceive HI and hearing difficulties were 34.38 dBHL and 54.38 dBHL, respectively. Males and younger participants were more willing to use HA. The primary reasons for refusing HA use were discomfort (25.1%) and a self-perception that the HA was unnecessary (19.7%). CONCLUSIONS: The prevalence of HI was high among the elderly population in southern Taiwan. Age and sex were the determinants of HA use.