Attention Feature Fusion Network via Knowledge Propagation for Automated Respiratory Sound Classification.

Goal: In light of the COVID-19 pandemic, the early diagnosis of respiratory diseases has become increasingly crucial. Traditional diagnostic methods such as computed tomography (CT) and magnetic resonance imaging (MRI), while accurate, often face accessibility challenges. Lung auscultation, a simpler alternative, is subjective and highly dependent on the clinician's expertise. The pandemic has further exacerbated these challenges by restricting face-to-face consultations. This study aims to overcome these limitations by developing an automated respiratory sound classification system using deep learning, facilitating remote and accurate diagnoses. Methods: We developed a deep convolutional neural network (CNN) model that utilizes spectrographic representations of respiratory sounds within an image classification framework. Our model is enhanced with attention feature fusion of low-to-high-level information based on a knowledge propagation mechanism to increase classification effectiveness. This novel approach was evaluated using the ICBHI benchmark dataset and a larger, self-collected Pediatric dataset comprising outpatient children aged 1 to 6 years. Results: The proposed CNN model with knowledge propagation demonstrated superior performance compared to existing state-of-the-art models. Specifically, our model showed higher sensitivity in detecting abnormalities in the Pediatric dataset, indicating its potential for improving the accuracy of respiratory disease diagnosis. Conclusions: The integration of a knowledge propagation mechanism into a CNN model marks a significant advancement in the field of automated diagnosis of respiratory disease. This study paves the way for more accessible and precise healthcare solutions, which is especially crucial in pandemic scenarios.

Association between the Dietary Phytochemical Index and Lower Prevalence of Obesity in Korean Preschoolers.

Little is known regarding Korean preschooler dietary phytochemical index (DPIs). We used the 24 h recall data of 1196 participants aged 3-5 years from the Korea National Health and Nutrition Examination Survey to study the association between dietary food intake and obesity prevalence. The amount of dietary intake by food group was compared according to sex and DPI quartile. Multivariable-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using logistic regression models. The average total DPI and energy from phytochemical food groups were not significantly different according to sex, although boys had a higher total daily food intake. Different inclinations between DPI quartiles and amount of intake were observed in the food groups; specifically, beans showed a higher intake difference between Q1 and Q4 for boys than in the other food groups. The highest DPI quartile had a significantly lower obesity prevalence than the lowest DPI quartile in all models for boys only when obesity prevalence by weight percentile was analyzed (Model 3, OR: 0.287, 95% CI: 0.095-0.868, p for trend < 0.05). Our results suggest a high DPI could help prevent obesity in preschoolers.

Seasonal Changes in Vitamin D Levels of Healthy Children in Mid-Latitude, Asian Urban Area.

PURPOSE: This study aimed to investigate the seasonal changes in vitamin D levels in a healthy pediatric population living in mid-latitude East Asian urban areas. METHODS: A pediatric population was selected from single secondary hospital visitors. Clinical data and serum vitamin D levels were collected retrospectively. Statistical analyses were performed based on the month of the blood sampling date, subject age, and vitamin D supplementation history. The data were categorized into three subgroups based on serum vitamin D levels-adequate (≥30 ng/mL), insufficient (20-29 ng/mL), and deficient (<20 ng/mL). RESULTS: Of the 481 patients, 172 had vitamin D supplementation history. More than 70% of the total study population had inadequate vitamin D levels (<30 ng/mL). The non-supplemented group and the supplemented group showed significantly uneven monthly distribution of the adequate, insufficient, and deficient subgroups. Only the non-supplemented group showed significantly different average vitamin D levels in the summer months compared to the winter months. In the non-supplemented group, vitamin D levels were the lowest in March, the highest in August and September. Significant relevance was noted between vitamin D supplementation status and vitamin D serum level in February and March. There was no significant difference between different age groups in terms of the distribution of vitamin D levels. CONCLUSION: Currently-widespread vitamin D replacement methods seem to have some effect on increasing the overall serum vitamin D levels, specifically during late winter when natural serum vitamin D levels plunge. However, they are unable to fully compensate the seasonal fluctuation.

Single-center experience of the Korean-Developmental Screening Test for infants and children.

PURPOSE: We investigated the number of test takers of the Korean-Developmental Screening Test (K-DST) in a single children's hospital within a year, according to age, referral rate, and follow-up percentage. METHODS: For this study, 4,062 children who visited and received K-DST at Woorisoa Children's Hospital between January and December 2015 were enrolled. Seven test sets were used according to the Korean National Health Screening Program for infants and children in the following age groups: 4 to 6, 9 to 12, 18 to 24, 30 to 36, 42 to 48, 54 to 60, and 66 to 71 months. The results of the K-DST were categorized into 4 groups as follows: further evaluation (<-2 standard deviation [-2SD]), follow-up test (-2SD to -1SD), peer level (-1SD to 1SD), and high level (>1SD). RESULTS: The test participants' population and follow-up population were concentrated before the age of 24 months (2,532, 62.3%). The children most commonly referred for further evaluation were those in the 30- to 41-month age group. A mismatch was found between the results of the K-DST and the additional questions. Most of the infants and children with suspicious developmental delays showed catch-up development in their follow-up tests (43 of 55, 78.2%). CONCLUSION: The use of K-DST should be encouraged, especially among children aged over 24 months. Multiple-choice question format for the additional questions is recommended to avoid confusion. We suggest a nationwide study to evaluate and revise the K-DST.