Dietary supplementation with Lactium and L-theanine alleviates sleep disturbance in adults: a double-blind, randomized, placebo-controlled clinical study.

Introduction: The use of natural products for the treatment of sleep disturbances is increasing owing to the side effects and limitations of traditional sleep therapy. Moreover, recent studies have shown a significant correlation between sleep quality and gut microbiota composition. This study aimed to assess the impact of LTC-022, a commercially available dietary supplement containing Lactium and L-theanine, on enhancing sleep quality. Methods: Forty participants experiencing sleep discomfort were enrolled in a double-blind randomized controlled trial, wherein they received LTC-022 or a placebo orally for 8 weeks. The effects of treatment on sleep quality were assessed using the Pittsburgh Sleep Quality Index and Insomnia Severity Index. To comprehensively evaluate changes in sleep patterns, various parameters were evaluated, including the time in bed (TIB), total sleep time (TST), sleep onset latency (SOL), sleep efficiency (SE), wake after sleep onset (WASO) counts, and bedtime. These parameters were derived from daily sleep logs recorded over the 8-week study period, categorized into weekdays and weekends. Stool samples were analyzed for microbiome composition. The V4 region of bacterial 16S rRNA genes was amplified using specific primers (515F and 806R) and targeted for analysis. Microbial diversity, including operational taxonomic units, the Shannon and Chao indices, the Firmicutes/Bacteroidetes (F/B) ratio, and the variety of bacterial taxa, was assessed. Results: No significant differences were observed in sleep quality and insomnia scale characteristics between the two groups. In-depth analysis using sleep diaries showed that WASO counts after 8 weeks and bedtime after 4 weeks showed significant differences between the LTC-022 and control groups. In the LTC-022 group, significant differences were observed in the increase in TST, decrease in SOL, increase in SE, decrease in WASO counts, and earlier bedtime. Microbiome analysis revealed that the abundance of the genera Blautia and Ruminococcus increased in fecal samples from the LTC-022 group. Conclusion: These results suggest that continuous LTC-022 intake has a beneficial effect on maintaining sleep duration and an appropriate bedtime. Additionally, changes in the gut microbiota may be linked to changes in sleep patterns resulting from the consumption of Lactium and L-theanine. Clinical trial registration: https://cris.nih.go.kr/cris/search/detailSearch.do/22841, KCT0007750.

Exploring Stroke Risk through Mendelian Randomization: A Comprehensive Study Integrating Genetics and Metabolic Traits in the Korean Population.

Numerous risk factors play a role in the causation of stroke, and the cardiometabolic condition is a one of the most important. In Korea, various treatment methods are employed based on the constitutional type, which is known to differ significantly in cardiometabolic disease. In this study, we compared the estimates obtained for different groups by applying the Mendelian randomization method to investigate the causal effects of genetic characteristics on stroke, according to constitutional type. In clinical analysis, the subtypes differ significantly in diabetes or dyslipidemia. The genetic association estimates for the stroke subtype risk were obtained from MEGASTROKE, the International Stroke Genetics Consortium (ISGC), UKbiobank, and BioBank Japan (BBJ), using group-related SNPs as instrumental variables. The TE subtypes with higher risk of metabolic disease were associated with increased risk (beta = 4.190; s.e. = 1.807; p = 0.035) of cardioembolic stroke (CES), and the SE subtypes were associated with decreased risk (beta = -9.336, s.e. = 1.753; p = 3.87 × 10-5) of CES. The findings highlight the importance of personalized medicine in assessing disease risk based on an individual's constitutional type.

Consumption of dietary fiber and APOA5 genetic variants in metabolic syndrome: baseline data from the Korean Medicine Daejeon Citizen Cohort Study.

BACKGROUND: Consumption of dietary fiber has been suggested as an important aspect of a healthy diet to reduce the risk of metabolic syndrome (MetS), including cardiovascular disease. The role of fiber intake in MetS might differ by individual genetic susceptibility. APOA5 encodes a regulator of plasma triglyceride levels, which impacts the related mechanisms of MetS. This study investigated the association between dietary fiber and the risk of MetS, assessing their associations according to APOA5 genetic variants. METHODS: A total of 1985 participants aged 30-55 years were included from a cross-sectional study based on the Korean Medicine Daejeon Citizen Cohort study at baseline (2017-2019). Dietary fiber intake was measured using a semiquantitative food frequency questionnaire. The APOA5 polymorphisms (rs2266788 A > G, rs662799 A > G, and rs651821 T > C) were genotyped using the Asia Precision Medicine Research Array. Logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (95% CIs). RESULTS: A higher consumption of dietary fiber was associated with a lower prevalence of MetS (P = 0.025). Among the components of MetS, an inverse association with dietary fiber was observed in increased waist circumference (OR, 95% CI = 0.60, 0.41-0.88, P for trend = 0.009) and elevated triglycerides (OR, 95% CI = 0.69, 0.50-0.96, P for trend = 0.012). Regarding the interaction with APOA5 genetic variants, a stronger association with dietary fiber intake was shown in G allele carriers of rs662799 than in A/A carriers (OR, 95% CI = 2.34, 1.59-3.44, P for interaction = 0.024) and in C allele carriers of rs651821 than in T/T carriers (OR, 95% CI = 2.35, 1.59-3.46, P for interaction = 0.027). CONCLUSIONS: The findings of this study suggest that the benefits of dietary fiber on the risk of MetS could be modified by genetic variants of the APOA5 gene, providing a more effective strategy for preventing MetS.

HoloSR: deep learning-based super-resolution for real-time high-resolution computer-generated holograms.

This study presents HoloSR, a novel deep learning-based super-resolution approach designed to produce high-resolution computer-generated holograms from low-resolution RGBD images, enabling the real-time production of realistic three-dimensional images. The HoloSR combines the enhanced deep super-resolution network with resize and convolution layers, facilitating the direct generation of high-resolution computer-generated holograms without requiring additional interpolation. Various upscaling scales, extending up to ×4, are evaluated to assess the performance of our method. Quantitative metrics such as structural similarity and peak signal-to-noise ratio are employed to measure the quality of the reconstructed images. Our simulation and experimental results demonstrate that HoloSR successfully achieves super-resolution by generating high-resolution holograms from low-resolution RGBD inputs with supervised and unsupervised learning.

Elevated prevalence and treatment of sleep disorders from 2011 to 2020: a nationwide population-based retrospective cohort study in Korea.

OBJECTIVES: This study used National Health Insurance claims data from Korea to report the prevalence of sleep disorders and treatment status, including traditional Korean medicine, in the last 10 years. METHODS: This is a retrospective cohort study in Korea. All diagnosis and prescription data, including herbal medicine claims, from the Health Insurance Review and Assessment Service from 2011 to 2020 were reviewed. Prevalence estimation, direct medical expenses and prescribed amounts for sleep disorders were recorded. RESULTS: The prevalence of sleep disorders increased from 3 867 975 (7.62%) in 2011 to 7 446 846 (14.41%) in 2020, nearly doubling over 10 years. Insomnia was observed in 91.44% (n=9 011 692) of the patients. The mean number of hospital visits per patient for sleep disorders was 11.5 (±26.62). Benzodiazepines are the most commonly prescribed medications for sleep disorders, and gamma-isoyosan is the most frequently prescribed herbal medicine. CONCLUSIONS: Sleep disorders are continuously increasing, as is the use of medical services-personal and social medical expenses are also increasing accordingly. Sleep disorders should be recognised as a significant health problem that needs to be actively addressed to improve quality of life.

Association of Dietary Protein Sources and Their Adequacy, Body Composition and Risk of Sarcopenic Obesity in South Korean Populations: A Cross-Sectional Study.

Dietary protein sources and protein adequacy are crucial modulators of muscle quality and body composition. We investigated the association between dietary protein sources (and their adequacy) and body composition and the risk of sarcopenic obesity (SO) in South Korean populations. The participants (n = 1967) were classified into SO, obese, sarcopenia, and normal groups. A cross-sectional survey was conducted using the KS-15 questionnaire, short-form food frequency questionnaire, and anthropometric measurements. The percentage of body fat (male: 35.36 ± 0.51%; female: 44.14 ± 0.36%) was significantly high, while appendicular skeletal muscle (ASM; male: 36.39 ± 0.30%, female: 30.32 ± 0.19%) was low in the SO group. Beef and pork consumption was negatively associated with ASM (%) but positively associated with body fat (%) in the normal group and positively associated with ASM (kg/m2: beta = 0.002, p = 0.02) and BFM (kg: beta = 0.012, p = 0.03) in the SO group, respectively. The highest quintile (Q5: 173.6 g/day) showed a decreased risk of SO prevalence (AORs: 0.46, CI: 0.22-0.94) compared with that in the lowest quintile (Q1: 21.6 g/day) among the people with inadequacy protein intake. Daily poultry and egg intake was positively linked with body composition in the participants with SO, while red meat showed a negative effect on imbalanced body composition in participants in the normal and SO groups. Furthermore, a lower intake of poultry and eggs was strongly associated with SO prevalence in people who consumed inadequate amounts of daily dietary protein.

In Vivo Reprogramming Using Yamanaka Factors in the CNS: A Scoping Review.

Central nervous system diseases, particularly neurodegenerative disorders, pose significant challenges in medicine. These conditions, characterized by progressive neuronal loss, have remained largely incurable, exacting a heavy toll on individuals and society. In recent years, in vivo reprogramming using Yamanaka factors has emerged as a promising approach for central nervous system regeneration. This technique involves introducing transcription factors, such as Oct4, Sox2, Klf4, and c-Myc, into adult cells to induce their conversion into neurons. This review summarizes the current state of in vivo reprogramming research in the central nervous system, focusing on the use of Yamanaka factors. In vivo reprogramming using Yamanaka factors has shown promising results in several animal models of central nervous system diseases. Studies have demonstrated that this approach can promote the generation of new neurons, improve functional outcomes, and reduce scar formation. However, there are still several challenges that need to be addressed before this approach can be translated into clinical practice. These challenges include optimizing the efficiency of reprogramming, understanding the cell of origin for each transcription factor, and developing methods for reprogramming in non-subventricular zone areas. Further research is needed to overcome the remaining challenges, but this approach has the potential to revolutionize the way we treat central nervous system disorders.

Antimicrobial Activity of Morphology-Controlled Cu2O Nanoparticles: Oxidation Stability under Humid and Thermal Conditions.

Metal oxides can be used as antimicrobial agents, especially since they can be fabricated into various forms such as films, masks, and filters. In particular, the durability of antimicrobial agents and the duration of their antimicrobial activity are important factors that determine their suitability for a specific purpose. These factors are related to the morphology and size of particles. The metal oxide Cu2O is often oxidized to CuO in various conditions, which reduces its antimicrobial activity. This study focused on the oxidation of nanoparticles of Cu2O with three morphologies, namely, spherical, octahedral, and cubic morphologies, in excessively humid and excessive-thermal environments for a specific duration and the antimicrobial activity of the NPs. Cu2O nanoparticles were prepared using the chemical reduction method, and their morphology could be varied by adjusting the molar ratio of OH- to Cu2+ and changing the reducing agent. It was found that cubic Cu2O was the most stable against oxidation and had the smallest reduction in antimicrobial activity. This study examined the antimicrobial activity and the oxidation stability of Cu2O NPs with different morphologies but similar particle sizes.

Machine-learning model predicting quality of life using multifaceted lifestyles in middle-aged South Korean adults: a cross-sectional study.

BACKGROUND: In the context of population aging, advances in healthcare technology, and growing interest in healthy aging and higher quality of life (QOL), have gained central focus in public health, particularly among middle-aged adults. METHODS: This study presented an optimal prediction model for QOL among middle-aged South Korean adults (N = 4,048; aged 30-55 years) using a machine-learning technique. Community-based South Korean population data were sampled through multistage stratified cluster sampling. Twenty-one variables related to individual factors and various lifestyle patterns were surveyed. QOL was assessed using the Short Form Health Survey (SF-12) and categorized into total QOL, physical component score (PCS), and mental component score (MCS). Seven machine-learning algorithms were used to predict QOL: decision tree, Gaussian Naïve Bayes, k-nearest neighbor, logistic regression, extreme gradient boosting, random forest, and support vector machine. Data imbalance was resolved with the synthetic minority oversampling technique (SMOTE). Random forest was used to compare feature importance and visualize the importance of each variable. RESULTS: For predicting QOL deterioration, the random forest method showed the highest performance. The random forest algorithm using SMOTE showed the highest area under the receiver operating characteristic (AUC) for total QOL (0.822), PCS (0.770), and MCS (0.786). Applying the data, SMOTE enhanced model performance by up to 0.111 AUC. Although feature importance differed across the three QOL indices, stress and sleep quality were identified as the most potent predictors of QOL. Random forest generated the most accurate prediction of QOL among middle-aged adults; the model showed that stress and sleep quality management were essential for improving QOL. CONCLUSION: The results highlighted the need to develop a health management program for middle-aged adults that enables multidisciplinary management of QOL.

Effect of Various Acid Solutions on the CO2 Dissolution Rate, Morphology, and Particle Size of Precipitated Calcium Carbonate Synthesized Using Seashells.

In this study, the influence of acid solutions on the production of precipitated calcium carbonate (PCC) using seashells was investigated. In terms of the Ca dissolution efficiency and atmosphere for dissolving CO32-, the results indicate that HCl, HNO3, CH3COOH, and HCOOH at 1.0 M were the most ideal among the acid solutions. The use of weak acids resulted in the low degree of dissolution of Al and Fe. These impurities could be mostly removed through the pH adjustment process, leading to PCC with a purity of 99% or more. Further, CH3COOH and HCOOH exhibited low CaCO3 carbonation efficiency owing to the hydrogen bonding of the carboxyl group and its hindering effect on the growth of CaCO3 particles. In addition, in the presence of the carboxyl group, the morphology tended to be oval, and the particle size was small. Particularly, when CH3COOH was used, the combined effect of the low initial Ca ion concentration and slow CO2 dissolution rate resulted in minimal changes during the carbonation time and the smallest particle size. However, variations in the degree of Ca concentration with a change in the acid solution concentration influenced the dominance of nucleation and particle growth, leading to variations in the particle size. The results of this study revealed that when manufacturing PCC using seashells, the appropriate acid solution must be selected to obtain the required PCC properties.

Exploring Variations in Sleep Perception: Comparative Study of Chatbot Sleep Logs and Fitbit Sleep Data.

BACKGROUND: Patient-generated health data are important in the management of several diseases. Although there are limitations, information can be obtained using a wearable device and time-related information such as exercise time or sleep time can also be obtained. Fitbits can be used to acquire sleep onset, sleep offset, total sleep time (TST), and wakefulness after sleep onset (WASO) data, although there are limitations regarding the depth of sleep and satisfaction; therefore, the patient's subjective response is still important information that cannot be replaced by wearable devices. OBJECTIVE: To effectively use patient-generated health data related to time such as sleep, it is first necessary to understand the characteristics of the time response recorded by the user. Therefore, the aim of this study was to analyze the characteristics of individuals' time perception in comparison with wearable data. METHODS: Sleep data were acquired for 2 weeks using a Fitbit. Participants' sleep records were collected daily through chatbot conversations while wearing the Fitbit, and the two sets of data were statistically compared. RESULTS: In total, 736 people aged 30-59 years were recruited for this study, and the sleep data of 543 people who wore a Fitbit and responded to the chatbot for more than 7 days on the same day were analyzed. Research participants tended to respond to sleep-related times on the hour or in 30-minute increments, and each participant responded within the range of 60-90 minutes from the value measured by the Fitbit. On average for all participants, the chat responses and the Fitbit data were similar within a difference of approximately 15 minutes. Regarding sleep onset, the participant response was 8 minutes and 39 seconds (SD 58 minutes) later than that of the Fitbit data, whereas with respect to sleep offset, the response was 5 minutes and 38 seconds (SD 57 minutes) earlier. The participants' actual sleep time (AST) indicated in the chat was similar to that obtained by subtracting the WASO from the TST measured by the Fitbit. The AST was 13 minutes and 39 seconds (SD 87 minutes) longer than the time WASO was subtracted from the Fitbit TST. On days when the participants reported good sleep, they responded 19 (SD 90) minutes longer on the AST than the Fitbit data. However, for each sleep event, the probability that the participant's AST was within ±30 and ±60 minutes of the Fitbit TST-WASO was 50.7% and 74.3%, respectively. CONCLUSIONS: The chatbot sleep response and Fitbit measured time were similar on average and the study participants had a slight tendency to perceive a relatively long sleep time if the quality of sleep was self-reported as good. However, on a participant-by-participant basis, it was difficult to predict participants' sleep duration responses with Fitbit data. Individual variations in sleep time perception significantly affect patient responses related to sleep, revealing the limitations of objective measures obtained through wearable devices.

Characteristics of insulin resistance in Korean adults from the perspective of circadian and metabolic sensing genes.

BACKGROUND: The biological clock allows an organism to anticipate periodic environmental changes and adjust its physiology and behavior accordingly. OBJECTIVE: This retrospective cross-sectional study examined circadian gene polymorphisms and clinical characteristics associated with insulin resistance (IR). METHODS: We analyzed data from 1,404 Korean adults aged 30 to 55 with no history of cancer and cardio-cerebrovascular disease. The population was classified according to sex and homeostasis model assessment of insulin resistance (HOMA-IR) values. Demographics, anthropometric and clinical characteristics, and single nucleotide polymorphisms (SNPs) were analyzed with respect to sex, age, and HOMA-IR values. We used association rule mining to identify sets of SNPs from circadian and metabolic sensing genes that may be associated with IR. RESULTS: Among the subjects, 15.0% of 960 women and 24.3% of 444 men had HOMA-IR values above 2. Most of the parameters differed significantly between men and women, as well as between the groups with high and low insulin sensitivity. Body fat mass of the trunk, which was significantly higher in insulin-resistant groups, had a higher correlation with high sensitivity C-reactive protein and hemoglobin levels in women, and alanine aminotransferase and aspartate aminotransferase levels in men. Homozygous minor allele genotype sets of SNPs rs17031578 and rs228669 in the PER3 gene could be more frequently found among women with HOMA-IR values above 2 (p = .014). CONCLUSION: Oxidative stress enhanced by adiposity and iron overload, which may also be linked to NRF2 and PER3-related pathways, is related to IR in adulthood. However, due to the small population size in this study, more research is needed.

Surface Oxidation of Cu2O Nanoparticles by Adsorbed Ammonia.

Copper-based nanoparticles have been intensively studied owing to their superior antibacterial activity. In this study, cuprous oxide (Cu2O) nanoparticles were synthesized using two different methods. In particular, two methods for synthesizing copper oxide from NaOH, namely, with and without the addition of NH3, were used to adjust the morphology of the nanoparticles. The nanoparticles from the NH3 and NaOH samples possessed an octahedral morphology. The crystal structure of the samples was confirmed by X-ray diffraction. The size distribution of the NH3 sample was narrower than that of the NaOH sample. Furthermore, the average size of the NH3 sample was smaller than that of the NaOH sample. Unexpectedly, the antibacterial activity of the NH3 sample was found to be lower than that of the NaOH sample. X-ray photoelectron spectroscopy and Fourier-transform infrared spectroscopy revealed that the adsorbed NH3 caused the surface oxidation of Cu2O nanoparticles with azide (N3) formation on surface.

Segmental bioelectrical impedance analysis for Korean older population with cold pattern.

Objective: This study examined the association of whole-body composition and segmental bioimpedance variables with cold pattern (CP) in different sexes. Methods: We assigned 667 older individuals to a CP group (n = 488) and a non-CP group (n = 179) by using an eight-item self-administered questionnaire. Seven body composition variables and three pairs of segmental bioimpedance variables for the upper and lower extremities, which were obtained from a segmental multifrequency bioimpedance analyzer, were employed to investigate their association with CP. Participants' characteristics were first described. Then we compared the selected body composition and bioimpedance variables between the CP and non-CP groups. Finally, their association with CP was investigated using univariate and multivariate regression analyses. All analyses were performed separately for women and men. Results: Both women and men exhibited a comparable mean age in the CP and non-CP groups; however, women with CP had significantly lower blood pressures, whereas men with CP showed a higher proportion of osteoarthritis than those without CP. Compared with the non-CP group, individuals with CP exhibited significantly smaller body sizes indicated by shorter height and smaller weight, lower body mass index, and smaller volume-to-body surface area ratio in both sexes. After controlling for age, height, weight, and other covariates, we found significant reductions in body lean mass such as fat-free mass and body cell mass, basal metabolic rate per unit mass, total body water, and intra-to-extracellular water ratio in the CP group. With regard to segmental bioimpedance analysis, the resistance ratios and phase angles in the upper and lower extremities yield significant associations with CP incidence, as demonstrated by the odds ratio (95% confidence interval) of 1.72 (1.16-2.57), 1.69 (1.18-2.48), 0.60 (0.40-0.89), and 0.57 (0.39-0.82), respectively. However, these results did not emerge in men. Conclusion: Abnormal cellular water distribution and deterioration in body cell mass and/or cell strength are associated with CP prevalence, regardless of age, height, weight. These findings are similar in the upper and lower extremities and are more pronounced in women. The abovementioned patterns may be considered effective indicators for identifying CP in the older adult population.

Partial sleep restriction-induced changes in stress, quality of life, and lipid metabolism in relation to cold hypersensitivity: A before-and-after intervention study.

Sleep disturbances are associated with cold hypersensitivity (CH) and characterized by excessive cold sensation in specific body parts and cold thermal discomfort. This study investigated the effects of short-term sleep restriction followed by a recovery phase on subjective health status, inflammation, and lipid metabolism in different types of CH. A total of 118 healthy adults aged 35 to 44 years without sleep disturbances were enrolled. Participants underwent 4-hour sleep restrictions per day for 3 days at a hospital and then returned to their daily lives for 4 days of rest. CH was assessed using a structured questionnaire with eight characteristic symptoms. A questionnaire and blood tests were administered baseline, after sleep restriction, and follow-up to assess cortisol, lipid profiles, and self-reported stress and quality of life (QOL). Participants were divided into CH (44.1%) and non-CH (55.9%) groups. The CH group showed increased stress, impaired QOL, and decreased low-density lipoprotein-cholesterol (LDL-C) levels compared to the non-CH group after sleep restriction. The variance for QOL (effect size = 0.07), subjective stress (effect size = 0.053), and LDL-C (effect size = 0.029) among time points depended on the group. Short-term sleep restriction was associated with deterioration of subjective health and reduced lipid metabolism; such changes were more evident in the CH group. Our findings suggest the need to consider an individual's CH status to assess the clinical risk associated with insufficient sleep.

Individual Differences in the Association of Lifestyle with Cardiometabolic Risk in Middle-Aged Koreans Based on Traditional Korean Medicine.

Cardiometabolic (CM) risk differs in morbidity and disease progression depending on lifestyle and individual characteristics. This cross-sectional study aimed to analyze the association between CM risk and lifestyle and examine whether this association varies based on Sasang constitution (SC), a Korean medicine classification. Data were analyzed from 1996 individuals participating in the Korean Medicine Daejeon Citizen Cohort study. Individuals with CM risk had two to five CM risk factors. Lifestyle factors included physical activity, sleep duration, and eating index. SC types were Taeeum-in (TE) and non-TE. We estimated the odds ratio (OR) and 95% confidence interval for CM risk based on SC and lifestyle factors. Among the participants, 33.9% had a CM risk, and the TE and non-TE groups were 26.2% and 7.7%, respectively. In the TE group, CM risk was associated with low physical activity (OR, 1.63) and moderate eating index (OR, 1.41). In the non-TE group, CM risk was associated with ≥8 h of sleep (OR, 1.87) and marginally associated with ≤6 h of sleep. In the TE group alone, CM risk was significantly associated with lifestyle patterns that combined the three lifestyle factors and was approximately two-fold higher in patterns involving less physical activity. The effects of lifestyle on CM risk differed across SC types. To decrease the burden of cardiovascular diseases in middle-aged adults, we recommend an individualized management strategy of healthy lifestyle interventions for cardiovascular risks.

Sasang Constitution Type Combined with General Obesity May Act as a Risk Factor for Prediabetes Mellitus.

Sasang constitutional medicine is a traditional customized medicine in Korea that classifies people into four types: Taeeumin (TE), Taeyangin (TY), Soeumin (SE), and Soyangin (SY). This study explored whether Sasang constitution (SC) types combined with general obesity could be risk factors for prediabetes mellitus (pre-DM). This study was cross-sectional and was conducted from November 2007 to July 2011 in 23 Korean medical clinics. In total, 2185 eligible subjects participated. A t test, one-way ANOVA with Scheffé's post hoc analysis, the chi-square test and multinomial logistic regression were used. Significance was indicated by p < 0.05. The numbers of participants with normal fasting plasma glucose (FPG) levels and pre-DM were 405 (75.3%) and 133 (24.7%) in the SE, 516 (70.3%) and 218 (29.7%) in the SY, and 590 (64.6%) and 323 (35.4%) in the TE (p < 0.001) groups, respectively. There was a significant difference in the proportion of each SC type among people with pre-DM and normal FPG levels in the normal BMI group. The odds ratios (ORs) of the TE type were significantly different from those of the SE type in the crude and Model 1. The distribution of the normal FPG and pre-DM individuals between the obese and normal BMI groups only for the SY type was significantly different. The SY type combined with general obesity had a higher OR (1.846, 95% CI 1.286−2.649) than that combined with normal BMI among participants with pre-DM, and this higher OR remained after adjusting for covariates (OR, 1.604, 95% CI, 1.093−2.354). This study revealed that the TE type might be a risk factor for pre-DM in the normal BMI group, and the SY type with general obesity could be a risk factor for pre-DM compared with the SY type with normal BMI. Accordingly, SC and BMI should be considered when managing pre-DM. To clarify the risk of SC and BMI, further study including epigenetic factors is needed.

A Machine Learning Approach for Recommending Herbal Formulae with Enhanced Interpretability and Applicability.

Herbal formulae (HFs) are representative interventions in Korean medicine (KM) for the prevention and treatment of various diseases. Here, we proposed a machine learning-based approach for HF recommendation with enhanced interpretability and applicability. A dataset consisting of clinical symptoms, Sasang constitution (SC) types, and prescribed HFs was derived from a multicenter study. Case studies published over 10 years were collected and curated by experts. Various classifiers, oversampling methods, and data imputation techniques were comprehensively considered. The local interpretable model-agnostic explanation (LIME) technique was applied to identify the clinical symptoms that led to the recommendation of specific HFs. We found that the cascaded deep forest (CDF) model with data imputation and oversampling yielded the best performance on the training set and holdout test set. Our model also achieved top-1 and top-3 accuracies of 0.35 and 0.89, respectively, on case study datasets in which clinical symptoms were only partially recorded. We performed an expert evaluation on the reliability of interpretation results using case studies and achieved a score close to normal. Taken together, our model will contribute to the modernization of KM and the identification of an HF selection process through the development of a practically useful HF recommendation model.