Suicide in adolescents is a major problem worldwide and previous history of suicide ideation and attempt represents the strongest predictors of future suicidal behavior. The aim of this study was to develop prediction model to identify Korean adolescents of high risk suicide (= who have history of suicide ideation/attempt in previous year) using machine learning techniques. A nationally representative dataset of Korea Youth Risk Behavior Web-based Survey (KYRBWS) was used (n = 59,984 of middle and high school students in 2017). The classification process was performed using machine learning techniques such as logistic regression (LR), random forest (RF), support vector machine (SVM), artificial neural network (ANN), and extreme gradient boosting (XGB). A total of 7,443 adolescents (12.4%) had a previous history of suicidal ideation/attempt. In the multivariable analysis, sadness (odds ratio [OR], 6.41; 95% confidence interval [95% CI], 6.08–6.87), violence (OR, 2.32; 95% CI, 2.01–2.67), substance use (OR, 1.93; 95% CI, 1.52–2.45), and stress (OR, 1.63; 95% CI, 1.40–1.86) were associated factors. Taking into account 26 variables as predictors, the accuracy of models of machine learning techniques to predict the high-risk suicidal was comparable with that of LR; the accuracy was best in XGB (79.0%), followed by SVM (78.7%), LR (77.9%), RF (77.8%), and ANN (77.5%). The machine leaning techniques showed comparable performance with LR to classify adolescents who have previous history of suicidal ideation/attempt. This model will hopefully serve as a foundation for decreasing future suicides as it enables early identification of adolescents at risk of suicide and modification of risk factors.

This study aimed to investigate the applicability of machine learning to predict obstructive sleep apnea (OSA) among individuals with suspected OSA in South Korea. A total of 92 clinical variables for OSA were collected from 279 South Koreans (OSA, n = 213; no OSA, n = 66), from which seven major clinical indices were selected. The data were randomly divided into training data (OSA, n = 149; no OSA, n = 46) and test data (OSA, n = 64; no OSA, n = 20). Using the seven clinical indices, the OSA prediction models were trained using four types of machine learning models—logistic regression, support vector machine (SVM), random forest, and XGBoost (XGB)—and each model was validated using the test data. In the validation, the SVM showed the best OSA prediction result with a sensitivity, specificity, and area under curve (AUC) of 80.33%, 86.96%, and 0.87, respectively, while the XGB showed the lowest OSA prediction performance with a sensitivity, specificity, and AUC of 78.69%, 73.91%, and 0.80, respectively. The machine learning algorithms showed high OSA prediction performance using data from South Koreans with suspected OSA. Hence, machine learning will be helpful in clinical applications for OSA prediction in the Korean population.

Back pain, especially lower back pain, is experienced in 60% to 80% of adults at some points during their lives. Various studies have found that lower back pain is a very common problem among adolescents, and the highest incidence rates are for adults in their 30s. There has been a remarkable increase in using computer-aided diagnosis to assist doctors in the interpretation of medical images. Spine segmentation in computed tomography (CT) scans using algorithmic methods allows improved diagnosis of back pain. In this study, we developed a web-based automatic spine segmentation method using deep learning and obtained the dice coefficient by comparison with the predicted image. Our method is based on convolutional neural networks for segmentation. More specifically, we train a hierarchical data format file using U-Net architecture and then insert the test data label to perform segmentation. Thus, we obtained more specific and detailed results. A total of 344 CT images were used in the experiment. Of these, 330 were used for learning, and the remaining 14 for testing. Our method achieved an average dice coefficient of 90.4%, a precision of 96.81%, and an F1-score of 91.64%. The proposed web-based deep learning approach can be very practical and accurate for spine segmentation as a diagnostic method.

The purpose of this study was to explore the effects of CT slice thickness, reconstruction algorithm, and radiation dose on quantification of CT features to characterize lung nodules using a chest phantom. Spherical lung nodule phantoms of known densities (−630 and + 100 HU) were inserted into an anthropomorphic thorax phantom. CT scan was performed ten times with relocations. CT data were reconstructed using 12 different imaging settings; three different slice thicknesses of 1.25, 2.5, and 5.0 mm, two reconstruction kernels of sharp and standard, and two radiation dose of 30 mAs and 12 mAs. Lesions were segmented using a semiautomated method. Twenty representative CT quantitative features representing CT density and texture were compared using multiple regression analysis. In 100 HU nodule phantoms, 18 and 19 among 20 computer features showed significant difference between different mAs and reconstruction algorithms, respectively (p≤0.05). 20, 19, and 19 computer features showed difference between slice thickness of 5.0 vs 1.25, 5.0 vs 2.5, and 2.5 vs 1.25 mm, respectively (p≤0.05). In −630 HU nodule phantoms, 18 and 19 showed significant difference between different mAs and reconstruction algorithms, respectively (p≤0.05). 18, 11, and 17 computer features showed difference between slice thickness of 5.0 vs 1.25, 5.0 vs 2.5, and 2.5 vs 1.25 mm, respectively (p≤0.05). When comparing the absolute value of regression coefficient, the effect of slice thickness in 100 HU nodule and reconstruction algorithm in −630 HU nodule was greater than the effect of remaining scan parameters. The slice thickness, mAs, and reconstruction algorithm had a significant impact on the quantitative image features. In clinical studies involving deep learning or radiomics, it should be noted that differences in values can occur when using computer features obtained from different CT scan parameters in combination. Therefore, when interpreting the statistical analysis results, it is necessary to reflect the difference in the computer features depending on the scan parameters.

This study was conducted to evaluate the diagnostic performance of machine learning in differentiating follicular adenoma from carcinoma using preoperative ultrasonography (US). In this retrospective study, preoperative US images of 348 nodules from 340 patients were collected from two tertiary referral hospitals. Two experienced radiologists independently reviewed each image and categorized the nodules according to the 2015 American Thyroid Association guideline. Categorization of a nodule as highly suspicious was considered a positive diagnosis for malignancy. The nodules were manually segmented, and 96 radiomic features were extracted from each region of interest. Ten significant features were selected and used as final input variables in our in-house developed classifier models based on an artificial neural network (ANN) and support vector machine (SVM). The diagnostic performance of radiologists and both classifier models was calculated and compared. In total, 252 nodules from 245 patients were confirmed as follicular adenoma and 96 nodules from 95 patients were diagnosed as follicular carcinoma. As measures of diagnostic performance, the average sensitivity, specificity, and accuracy of the two experienced radiologists in discriminating follicular adenoma from carcinoma on preoperative US images were 24.0%, 84.0%, and 64.8%, respectively. The sensitivity, specificity, and accuracy of the ANN and SVM-based models were 32.3%, 90.1%, and 74.1% and 41.7%, 79.4%, and 69.0%, respectively. The kappa value of the two radiologists was 0.076, corresponding to slight agreement. Machine learning-based classifier models may aid in discriminating follicular adenoma from carcinoma using preoperative US.