Background: This study aimed to evaluate the applicability of deep learning technology to thyroid ultrasound images for classification of thyroid nodules. Methods: This retrospective analysis included ultrasound images of patients with thyroid nodules investigated by fine-needle aspiration at the thyroid clinic of a single center from April 2010 to September 2012. Thyroid nodules with cytopathologic results of Bethesda category V (suspicious for malignancy) or VI (malignant) were defined as thyroid cancer. Multiple deep learning algorithms based on convolutional neural networks (CNNs) —ResNet, DenseNet, and EfficientNet—were utilized, and Siamese neural networks facilitated multi-view analysis of paired transverse and longitudinal ultrasound images. Results: Among 1,048 analyzed thyroid nodules from 943 patients, 306 (29%) were identified as thyroid cancer. In a subgroup analysis of transverse and longitudinal images, longitudinal images showed superior prediction ability. Multi-view modeling, based on paired transverse and longitudinal images, significantly improved the model performance; with an accuracy of 0.82 (95% confidence intervals [CI], 0.80 to 0.86) with ResNet50, 0.83 (95% CI, 0.83 to 0.88) with DenseNet201, and 0.81 (95% CI, 0.79 to 0.84) with EfficientNetv2_ s. Training with high-resolution images obtained using the latest equipment tended to improve model performance in association with increased sensitivity. Conclusion CNN algorithms applied to ultrasound images demonstrated substantial accuracy in thyroid nodule classification, indicating their potential as valuable tools for diagnosing thyroid cancer. However, in real-world clinical settings, it is important to aware that model performance may vary depending on the quality of images acquired by different physicians and imaging devices.

Background: This study aimed to evaluate the applicability of deep learning technology to thyroid ultrasound images for classification of thyroid nodules. Methods: This retrospective analysis included ultrasound images of patients with thyroid nodules investigated by fine-needle aspiration at the thyroid clinic of a single center from April 2010 to September 2012. Thyroid nodules with cytopathologic results of Bethesda category V (suspicious for malignancy) or VI (malignant) were defined as thyroid cancer. Multiple deep learning algorithms based on convolutional neural networks (CNNs) —ResNet, DenseNet, and EfficientNet—were utilized, and Siamese neural networks facilitated multi-view analysis of paired transverse and longitudinal ultrasound images. Results: Among 1,048 analyzed thyroid nodules from 943 patients, 306 (29%) were identified as thyroid cancer. In a subgroup analysis of transverse and longitudinal images, longitudinal images showed superior prediction ability. Multi-view modeling, based on paired transverse and longitudinal images, significantly improved the model performance; with an accuracy of 0.82 (95% confidence intervals [CI], 0.80 to 0.86) with ResNet50, 0.83 (95% CI, 0.83 to 0.88) with DenseNet201, and 0.81 (95% CI, 0.79 to 0.84) with EfficientNetv2_ s. Training with high-resolution images obtained using the latest equipment tended to improve model performance in association with increased sensitivity. Conclusion CNN algorithms applied to ultrasound images demonstrated substantial accuracy in thyroid nodule classification, indicating their potential as valuable tools for diagnosing thyroid cancer. However, in real-world clinical settings, it is important to aware that model performance may vary depending on the quality of images acquired by different physicians and imaging devices.

Background: This study aimed to evaluate the applicability of deep learning technology to thyroid ultrasound images for classification of thyroid nodules. Methods: This retrospective analysis included ultrasound images of patients with thyroid nodules investigated by fine-needle aspiration at the thyroid clinic of a single center from April 2010 to September 2012. Thyroid nodules with cytopathologic results of Bethesda category V (suspicious for malignancy) or VI (malignant) were defined as thyroid cancer. Multiple deep learning algorithms based on convolutional neural networks (CNNs) —ResNet, DenseNet, and EfficientNet—were utilized, and Siamese neural networks facilitated multi-view analysis of paired transverse and longitudinal ultrasound images. Results: Among 1,048 analyzed thyroid nodules from 943 patients, 306 (29%) were identified as thyroid cancer. In a subgroup analysis of transverse and longitudinal images, longitudinal images showed superior prediction ability. Multi-view modeling, based on paired transverse and longitudinal images, significantly improved the model performance; with an accuracy of 0.82 (95% confidence intervals [CI], 0.80 to 0.86) with ResNet50, 0.83 (95% CI, 0.83 to 0.88) with DenseNet201, and 0.81 (95% CI, 0.79 to 0.84) with EfficientNetv2_ s. Training with high-resolution images obtained using the latest equipment tended to improve model performance in association with increased sensitivity. Conclusion CNN algorithms applied to ultrasound images demonstrated substantial accuracy in thyroid nodule classification, indicating their potential as valuable tools for diagnosing thyroid cancer. However, in real-world clinical settings, it is important to aware that model performance may vary depending on the quality of images acquired by different physicians and imaging devices.

Background: This study aimed to evaluate the applicability of deep learning technology to thyroid ultrasound images for classification of thyroid nodules. Methods: This retrospective analysis included ultrasound images of patients with thyroid nodules investigated by fine-needle aspiration at the thyroid clinic of a single center from April 2010 to September 2012. Thyroid nodules with cytopathologic results of Bethesda category V (suspicious for malignancy) or VI (malignant) were defined as thyroid cancer. Multiple deep learning algorithms based on convolutional neural networks (CNNs) —ResNet, DenseNet, and EfficientNet—were utilized, and Siamese neural networks facilitated multi-view analysis of paired transverse and longitudinal ultrasound images. Results: Among 1,048 analyzed thyroid nodules from 943 patients, 306 (29%) were identified as thyroid cancer. In a subgroup analysis of transverse and longitudinal images, longitudinal images showed superior prediction ability. Multi-view modeling, based on paired transverse and longitudinal images, significantly improved the model performance; with an accuracy of 0.82 (95% confidence intervals [CI], 0.80 to 0.86) with ResNet50, 0.83 (95% CI, 0.83 to 0.88) with DenseNet201, and 0.81 (95% CI, 0.79 to 0.84) with EfficientNetv2_ s. Training with high-resolution images obtained using the latest equipment tended to improve model performance in association with increased sensitivity. Conclusion CNN algorithms applied to ultrasound images demonstrated substantial accuracy in thyroid nodule classification, indicating their potential as valuable tools for diagnosing thyroid cancer. However, in real-world clinical settings, it is important to aware that model performance may vary depending on the quality of images acquired by different physicians and imaging devices.

Objective: The Scales for Outcomes in Parkinson’s Disease–Cognition (SCOPA-Cog) was developed to assess cognition in patients with Parkinson’s disease (PD). In this study, we aimed to evaluate the validity and reliability of the Korean version of the SCOPACog (K-SCOPA-Cog). Methods: We enrolled 129 PD patients with movement disorders from 31 clinics in South Korea. The original version of the SCOPA-Cog was translated into Korean using the translation-retranslation method. The test–retest method with an intraclass correlation coefficient (ICC) and Cronbach’s alpha coefficient were used to assess reliability. Spearman’s rank correlation analysis with the Montreal Cognitive Assessment-Korean version (MOCA-K) and the Korean Mini-Mental State Examination (K-MMSE) were used to assess concurrent validity. Results: The Cronbach’s alpha coefficient was 0.797, and the ICC was 0.887. Spearman’s rank correlation analysis revealed a significant correlation with the K-MMSE and MOCA-K scores (r = 0.546 and r = 0.683, respectively). Conclusion Our results demonstrate that the K-SCOPA-Cog has good reliability and validity.

Background: Hand hygiene is considered the simplest and most cost-effective method of infection prevention. Regular observation and feedback on hand hygiene compliance are key strategies for its enhancement. This study evaluated the effectiveness of hand hygiene surveillance, including direct observation and feedback, by comprehensively analyzing the reported hand hygiene compliance within the Korean National Healthcare-Associated Infections Surveillance System from 2019 to 2022. Methods: Participating medical institutions included general hospitals and hospitals with infection control departments that consented to participate. Hand hygiene surveillance was conducted using direct observation. Collected data, including healthcare workers, clinical areas, hand hygiene moments, and hand hygiene compliance, were recorded to calculate hand hygiene compliance rates. Additionally, the volume of alcohol-based hand sanitizers used per patient per day was investigated as an indirect indicator of hand hygiene compliance. The study was conducted from July 2019 to December 2022. Results: Hand hygiene compliance increased from 87.2% in Q3 2019 to 89.9% in 2022. Nurses and medical technologists showed the highest compliance rates, whereas doctors showed the lowest compliance rates. Intensive care units excelled in compliance, whereas emergency de partments lagged. Compliance was highest after patient contact and lowest when the patient’s surroundings were touched. Larger hospitals consumed more alcohol-based hand sanitizers than smaller hospitals did. Conclusion This study confirmed an improvement in hand hygiene compliance through sustained surveillance, indicating its contribution not only to preventing infection transfer within healthcare facilities but also to fostering a culture of hand hygiene in the country.

In May 2023, the Committee of Clinical Practice Guidelines of the Korean Diabetes Association published the revised clinical practice guidelines for Korean adults with diabetes and prediabetes. We incorporated the latest clinical research findings through a comprehensive systematic literature review and applied them in a manner suitable for the Korean population. These guidelines are designed for all healthcare providers nationwide, including physicians, diabetes experts, and certified diabetes educators who manage patients with diabetes or individuals at risk of developing diabetes. Based on recent changes in international guidelines and the results of a Korean epidemiological study, the recommended age for diabetes screening has been lowered. In collaboration with the relevant Korean medical societies, recently revised guidelines for managing hypertension and dyslipidemia in patients with diabetes have been incorporated into this guideline. An abridgment containing practical information on patient education and systematic management in the clinic was published separately.

Background: The purpose of this study was to evaluate the effect of vanishing twin (VT) on maternal serum marker concentrations and nuchal translucency (NT). Methods: This is a secondary analysis of a multicenter prospective cohort study in 12 institutions. Serum concentrations of pregnancy-associated plasma protein-A in the first trimester and alpha-fetoprotein (AFP), total human chorionic gonadotrophin, unconjugated estriol, and inhibin A in the second trimester were measured, and NT was measured between 10 and 14 weeks of gestation. Results: Among 6,793 pregnant women, 5,381 women were measured for serum markers in the first or second trimester, including 65 cases in the VT group and 5,316 cases in the normal singleton group. The cases in the VT group had a higher median multiple of the median value of AFP and inhibin A than the normal singleton group. The values of other serum markers and NT were not different between the two groups. After the permutation test with adjustment,AFP and inhibin A remained significant differences. The frequency of abnormally increased AFP was also higher in the VT group than in the normal singleton group. Conclusion VT can be considered as an adjustment factor for risk assessment in the secondtrimester serum screening test.

Background/Aims: We evaluated the feasibility and long-term efficacy of the combination of cytarabine, idarubicin, and all-trans retinoic acid (ATRA) for treating patients with newly diagnosed acute promyelocytic leukemia (APL). Methods: We included 87 patients with newly diagnosed acute myeloid leukemia and a t(15;17) or promyelocytic leukemia/retinoic acid receptor alpha (PML-RARα) mutation. Patients received 12 mg/m2/day idarubicin intravenously for 3 days and 100 mg/m2/day cytarabine for 7 days, plus 45 mg/m2/day ATRA. Clinical outcomes included complete remission (CR), relapse-free survival (RFS), overall survival (OS), and the secondary malignancy incidence during a 20-year follow-up. Results: The CR, 10-year RFS, and 10-year OS rates were 89.7%, 94.1%, and 73.8%, respectively, for all patients. The 10-year OS rate was 100% for patients that achieved CR. Subjects were classified according to the white blood cell (WBC) count in peripheral blood at diagnosis (low-risk, WBC < 10,000/mm3; high-risk, WBC ≥ 10,000/mm3). The low-risk group had significantly higher RFS and OS rates than the high-risk group, but the outcomes were not superior to the current standard treatment (arsenic trioxide plus ATRA). Toxicities were similar to those observed with anthracycline plus ATRA, and higher than those observed with arsenic trioxide plus ATRA. The secondary malignancy incidence after APL treatment was 2.7%, among the 75 patients that achieved CR, and 5.0% among the 40 patients that survived more than 5 years after the APL diagnosis. Conclusions Adding cytarabine to anthracycline plus ATRA was not inferior to anthracycline plus ATRA alone, but it was not comparable to arsenic trioxide plus ATRA. The probability of secondary malignancy was low.

Purpose: We assessed prenatal detection rates of congenital heart disease (CHD) and associations between maternal serum biomarkers and non-chromosomal CHD in singleton pregnancies. Materials and Methods: This study was conducted as a secondary analysis of data obtained during a multicenter prospective cohort study that investigated the cost-effectiveness of prenatal testing for fetal aneuploidy. We analyzed the prenatal detection rate and accuracy for CHD screening via ultrasound during the second trimester, as well as associations between serum biomarkers and CHDs, in singleton newborns without chromosomal abnormalities. Results: Among 6715 women, 142 (2.1%) newborns were born with CHDs, of which 67 (1.0%) newborns had major CHDs. The prenatal detection rate for all CHDs and major CHDs were 34.5% and 58.2%, respectively. After excluding isolated ventricular septal defects, the detection rate for critical CHDs was 85.9%. Women with low pregnancy-associated plasma protein A (PAPP-A) (<0.4 multiples of the median, MOM) face increased risks of non-chromosomal CHDs [adjusted odds ratio (aOR) 2.76; 95% confidence interval (CI) 1.36–5.13] and major CHDs (aOR 7.30; 95% CI 3.18–15.59), compared to those without CHDs. A higher inhibin A level (≥2.5 MOM; aOR 4.84; 95% CI 1.42–12.46) was associated with non-chromosomal major CHDs. Conclusion Ultrasonography performed during the second trimester by obstetricians detected over 85% of critical CHDs. Low maternal serum PAPP-A or high inhibin-A was associated with non-chromosomal CHDs. These results may contribute to an improvement in prenatal diagnosis of CHDs.