Periodontal disease is a chronic but treatable condition which often does not cause pain during the initial stages of the illness. Lack of awareness of symptoms can delay initiation of treatment and worsen health. The aim of this study was to develop and compare different risk prediction models for periodontal disease using machine learning algorithms. We obtained information on risk factors for periodontal disease from the Korea National Health and Nutrition Examination Survey (KNHANES) dataset. Principal component analysis and an auto-encoder were used to extract data on risk factors for periodontal disease. A synthetic minority oversampling technique algorithm was used to solve the problem of data imbalance. We used a combination of logistic regression analysis, support vector machine (SVM) learning, random forest, and AdaBoost to classify and compare risk prediction models for periodontal disease. In cases where we used principal component analysis (PCA) to extract risk factors, the recall was higher than the feature selection method in the logistic regression and support-vector machine learning models. AdaBoost’s recall was 0.98, showing the highest performance of both feature selection and PCA. The F1 score showed relatively high performance in Ada- Boost, logistic regression, and SVM learning models. By using the risk factors extracted from the research results and the predictive model based on machine learning, it will be able to help in the prevention and diagnosis of periodontal disease, and it will be used to study the relationship with various diseases related to periodontal disease.

The incidence of stomach cancer has been found to be gradually decreasing; however, it remains one of the most frequently occurring malignant cancers in Korea. According to statistics of 2017, stomach cancer is the top cancer in men and the fourth most important cancer in women, necessitating methods for its early detection and treatment. Considerable research in the field of bioinformatics has been conducted in cancer studies, and bioinformatics approaches might help develop methods and models for its early prediction. We aimed to develop a classification method based on deep learning and demonstrate its application to gene expression data obtained from patients with stomach cancer. Data of 60,483 genes from 334 patients with stomach cancer in The Cancer Genome Atlas were evaluated by principal component analysis, heatmaps, and the convolutional neural network (CNN) algorithm. We combined the RNA-seq gene expression data with clinical data, searched candidate genes, and analyzed them using the CNN deep learning algorithm. We performed learning using the sample type and vital status of patients with stomach cancer and verified the results. We obtained an accuracy of 95.96% for sample type and 50.51% for vital status. Despite overfitting owing to the limited number of patients, relatively accurate results for sample type were obtained. This approach can be used to predict the prognosis of stomach cancer, which has many types and underlying causes.

Recently, N-terminal pro-brain natriuretic peptide (NTproBNP) has been widely used in the areas of diagnosis, monitoring treatment efficiency, and prognosis for various heart diseases, especially heart failure (HF). In this paper, we try to estimate the prognostic significance of NT-proBNP as a risk evaluation marker in Non-ST-segment Elevation Myocardial Infarction (NSTEMI) patients. We selected NSTEMI patients who underwent percutaneous coronary intervention (PCI) primarily using a drug-eluting stent within 24 h after the onset of chest pain. We compared incidences of major adverse cardiac events (MACE) including death, myocardial infarction (MI), stent thrombosis (ST), and target vessel revascularization (TVR) in two patient groups according to a high or low serum concentration of NT-proBNP, which was measured in the emergency room (ER). We intend to minimize selection bias selecting comparing groups, considering covariate of observed variables together using propensity score matching (PSM) and propensity score weighting (PSW) based on propensity score (PS) to control the difference in baseline characteristics between high- and low NT-proBNP groups. We found that as the log NT-proBNP value increases by 1 through a hazard function of COX’s analysis, the risk of MACE increases by 1.312 times. This result indicated that the NT-proBNP level on ER admission can be used as a significant prognostic indicator to estimate 1 year of MACE in NSTEMI patients who were treated with PCI within 24 h after the onset of chest pain.