松永 眞章

Background and Objective: Excellent generalizability is the precondition for the widespread practical implementation of machine learning models. In our previous study, we developed the schizophrenia classification model (SZ classifier) to identify potential schizophrenia patients in the Japanese population. The SZ classifier has exhibited impressive performance during internal validation. However, ensuring the robustness and generalizability of the SZ classifier requires external validation across independent sample sets. In this study, we aimed to present an external validation of the SZ classifier using outpatient data. Methods: The SZ classifier was trained by using online survey data, which incorporate demographic, health-related, and social comorbidity features. External validation was conducted using an outpatient sample set which is independent from the sample set during the model development phase. The model performance was assessed based on the sensitivity and misclassification rates for schizophrenia, bipolar disorder, and major depression patients. Results: The SZ classifier demonstrated a sensitivity of 0.75 when applied to schizophrenia patients. The misclassification rates were 59% and 55% for bipolar disorder and major depression patients, respectively. Conclusions: The SZ classifier currently encounters challenges in accurately determining the presence or absence of schizophrenia at the individual level. Prior to widespread practical implementation, enhancements are necessary to bolster the accuracy and diminish the misclassification rates. Despite the current limitations of the model, such as poor specificity for certain psychiatric disorders, there is potential for improvement if including multiple types of psychiatric disorders during model development.

Abstract Objectives This study aimed to develop an approach to enhance the model precision by artificial images. Materials and Methods Given an epidemiological study designed to predict 1 response using f features with M samples, each feature was converted into a pixel with certain value. Permutated these pixels into F orders, resulting in F distinct artificial image sample sets. Based on the experience of image recognition techniques, appropriate training images results in higher precision model. In the preliminary experiment, a binary response was predicted by 76 features, the sample set included 223 patients and 1776 healthy controls. Results We randomly selected 10 000 artificial sample sets to train the model. Models’ performance (area under the receiver operating characteristic curve values) depicted a bell-shaped distribution. Conclusion The model construction strategy developed in the research has potential to capture feature order related information and enhance model predictability.

Background In Japan, challenges were reported in accurately estimating the prevalence of schizophrenia among the general population. Retrieving previous studies, we investigated that patients with schizophrenia were more likely to experience poor subjective well-being and various physical, psychiatric, and social comorbidities. These factors might have great potential for precisely classifying schizophrenia cases in order to estimate the prevalence. Machine learning has shown a positive impact on many fields, including epidemiology, due to its high-precision modeling capability. It has been applied in research on mental disorders. However, few studies have applied machine learning technology to the precise classification of schizophrenia cases by variables of demographic and health-related backgrounds, especially using large-scale web-based surveys. Objective The aim of the study is to construct an artificial neural network (ANN) model that can accurately classify schizophrenia cases from large-scale Japanese web-based survey data and to verify the generalizability of the model. Methods Data were obtained from a large Japanese internet research pooled panel (Rakuten Insight, Inc) in 2021. A total of 223 individuals, aged 20-75 years, having schizophrenia, and 1776 healthy controls were included. Answers to the questions in a web-based survey were formatted as 1 response variable (self-report diagnosed with schizophrenia) and multiple feature variables (demographic, health-related backgrounds, physical comorbidities, psychiatric comorbidities, and social comorbidities). An ANN was applied to construct a model for classifying schizophrenia cases. Logistic regression (LR) was used as a reference. The performances of the models and algorithms were then compared. Results The model trained by the ANN performed better than LR in terms of area under the receiver operating characteristic curve (0.86 vs 0.78), accuracy (0.93 vs 0.91), and specificity (0.96 vs 0.94), while the model trained by LR showed better sensitivity (0.63 vs 0.56). Comparing the performances of the ANN and LR, the ANN was better in terms of area under the receiver operating characteristic curve (bootstrapping: 0.847 vs 0.773 and cross-validation: 0.81 vs 0.72), while LR performed better in terms of accuracy (0.894 vs 0.856). Sleep medication use, age, household income, and employment type were the top 4 variables in terms of importance. Conclusions This study constructed an ANN model to classify schizophrenia cases using web-based survey data. Our model showed a high internal validity. The findings are expected to provide evidence for estimating the prevalence of schizophrenia in the Japanese population and informing future epidemiological studies.

Abstract Nonpharmaceutical interventions (NPIs) to control COVID‐19 have decreased the incidence of many pediatric infectious diseases. The epidemiology of β‐ and γ‐herpesvirus infections might have been affected by NPIs. The aim of this study was to elucidate changes in trends in β‐ and γ‐herpesvirus infections and complex febrile seizures (cFS) of viral etiology before and during the COVID‐19 pandemic. Between April 2017 and March 2021, febrile children aged ≤5 years were enrolled. Detection of EBV, CMV, HHV‐6B, and HHV‐7 DNA in serum was performed using real‐time PCR. The epidemiology of viral infections and cFS were compared between the prepandemic and pandemic periods. During the observation period, 1432 serum samples were collected. The mean number of febrile children decreased during the pandemic period, but the number of patients with HHV‐6B infection increased from 35 (9.3% of all febrile children) per year before the pandemic to 43 (15.5%) during the pandemic. The change in the proportion of patients with primary HHV‐6B infection was 6.50% (95% confidence interval [CI], 2.05%–11.3%; p = 0.0047). The mean number of patients with cFS decreased during the pandemic period, but the number of patients with HHV‐6B–associated cFS was stable throughout the observation period. Therefore, the change in proportion of patients with cFS caused by primary HHV‐6B infection was 49.5% (95% CI, 12.2%–60.5%; p = 0.0048). The disease burden of primary HHV‐6B infection among patients in the emergency room remained unchanged, with a significant increase in the relative proportion after the COVID‐19 pandemic began.

BACKGROUND: Lung cancer frequently occurs in lungs with background idiopathic interstitial pneumonias (IIPs). Limited resection is often selected to treat lung cancer in patients with IIPs in whom respiratory function is already compromised. However, accurate surgical margins are essential for curative resection; underestimating these margins is a risk for residual lung cancer after surgery. We aimed to investigate the findings of lung fields adjacent to cancer segments affect the estimation of tumor size on computed tomography compared with the pathological specimen. METHODS: This analytical observational study retrospectively investigated 896 patients with lung cancer operated on at Fujita Health University from January 2015 to June 2020. The definition of underestimation was a ≥10 mm difference between the radiological and pathological maximum sizes of the tumor. RESULTS: The lung tumors were in 15 honeycomb, 30 reticulated, 207 emphysematous, and 628 normal lungs. The ratio of underestimation in honeycomb lungs was 33.3% compared to 7.4% without honeycombing (P=0.004). Multivariate analysis showed that honeycombing was a significant risk factor for tumor size underestimation. A Bland-Altman plot represented wide 95% limits of agreement, -40.8 to 70.2 mm, between the pathological and radiological maximum tumor sizes in honeycomb lungs.

The physical, psychiatric, and social comorbidities interfere with the everyday activities of community-dwelling individuals with schizophrenia and increase the risk of their readmission. However, these comorbidities have not been investigated comprehensively in Japan. We conducted a self-reported internet survey in February 2022 to identify individuals aged 20–75 years with and without schizophrenia using a prevalence case-control study. The survey compared physical comorbidities such as being overweight, hypertension, and diabetes; psychiatric comorbidities such as depressive symptoms and sleep disturbances; social comorbidities such as employment status, household income, and social support between participants with and without schizophrenia. A total of 223 participants with schizophrenia and 1776 participants without schizophrenia were identified. Participants with schizophrenia were more likely to be overweight and had a higher prevalence of hypertension, diabetes, and dyslipidemia than participants without schizophrenia. Additionally, depressive symptoms, unemployment, and non-regular employment were more prevalent in participants with schizophrenia than those without schizophrenia. These results highlight the necessity of comprehensive support and interventions addressing physical, psychiatric, and social comorbidities in individuals with schizophrenia in the community. In conclusion, effective interventions for managing comorbidities in individuals with schizophrenia are necessary to enable them to continue to live in the community.

The aim of the present study was to derive dietary patterns to explain variation in a set of nutrient intakes or in the measurements of waist circumference (WC) and fasting blood glucose (FBG) using reduced rank regression (RRR) and to prospectively investigate these patterns in relation to the risk of developing metabolic syndrome (MetS) and its components during the follow-up. The study participants were comprised of 2944 government employees aged 30–59 years without MetS. RRR was applied with 38 food groups as predictors and with two sets of response variables. The first set included intake of putatively beneficial nutrients, and the first factor retained was named the Healthy Dietary Pattern (HDP). The second one included baseline WC and FBG, and the first factor was named the Unhealthy Dietary Pattern (UHDP). Multivariable Cox proportional hazard model was used to estimate hazard ratio and 95% confidence intervals with adjustments for age, sex, total energy consumption and other potential confounders. During the 5-year median follow-up, we ascertained 374 cases of MetS. The HDP score was inversely associated with the incidence of MetS (p-trend = 0.009) and hypertension (p-trend = 0.002) and marginally significantly associated with elevated triglyceride and decreased high-density lipoprotein cholesterol (p-trend = 0.08). The UHDP score was linearly positively associated with the incidence of MetS and all its components (all p-trend < 0.05). Both the HDP and UHDP predicted the development of MetS and its components.

AIMS: The associations between low-density lipoprotein cholesterol (LDL-C) and the risk of cardiovascular disease (CVD) subtypes are not well established among the Japanese population. This study used longitudinal data from the Aichi Workers' Cohort Study to explore the association between LDL-C levels and the risk of coronary heart disease (CHD) and stroke subtypes. METHODS: Pooled data of 8966 adults (7093men and 1903 women) who were recruited between (2002) and (2008) were used for the current analysis. Propensity scores for the LDL-C categories were generated using multinomial logistic regression. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated from the inverse probability weighted Cox proportional hazards model for LDL-C category associations with risks of CHD, stroke subtypes, and CVD. RESULTS: During a median follow-up of 12 years, 122 strokes (57 ischemic strokes, 25 intracerebral hemorrhage, and 40 unknown subtypes) and 82 cases of CHD were observed. LDL-C 160- mg/dL compared to LDL-C 100-119 mg/dL was positively and significantly associated with the risk of CHD (HR: 4.56; 95% CI: 1.91-10.9) but not with ischemic stroke (HR: 0.99; 95% CI: 0.44-2.22). LDL-C was inversely associated with the risk of intracerebral hemorrhage (P for trend=0.009). CONCLUSION: In middle-aged Japanese workers, LDL-C was significantly and positively associated with CHD, but not with ischemic stroke. LDL-C was inversely significantly associated with intracerebral hemorrhage.

AIM: We aimed to examine the association of obesity-related cancer and cardiovascular disease (CVD) with body mass index (BMI) and the estimated population attributable fraction in lean Asians. METHODS: We studied 102,535 participants aged 40-79 years without histories of cancer or CVD at baseline between 1988 and 2009. The cause-specific hazard ratios (csHRs) of BMI categories (＜18.5, 18.5-20.9, 21.0-22.9 [reference], 23.0-24.9, 25.0-27.4, and ≥ 27.5 kg/m2) were estimated for each endpoint. The events considered were mortalities from obesity-related cancer (esophageal, colorectal, liver, pancreatic, kidney, female breast, and endometrial cancer) and those from CVD (coronary heart disease and stroke). Population attributable fractions (PAFs) were calculated for these endpoints. RESULTS: During a 19.2-year median follow-up, 2906 died from obesity-related cancer and 4532 died from CVD. The multivariable-adjusted csHRs (95% confidence interval) of higher BMI categories (25-27.4 and ≥ 27.5 kg/m2) for obesity-related cancer mortality were 0.93 (0.78, 1.10) and 1.18 (0.92, 1.50) in men and 1.25 (1.04, 1.50) and 1.48 (1.19, 1.84) in women, respectively. The corresponding csHRs for CVD mortality were 1.27 (1.10, 1.46) and 1.59 (1.30, 1.95) in men and 1.10 (0.95, 1.28) and 1.44 (1.21, 1.72) in women, respectively. The PAF of a BMI ≥ 25 kg/m2 for obesity-related cancer was -0.2% in men and 6.7% in women and that for CVD was 5.0% in men and 4.5% in women. CONCLUSION: A BMI ≥ 25 kg/m2 is associated with an increased risk of obesity-related cancer in women and CVD in both sexes.

PURPOSE: To investigate whether previously reported seasonal variation and winter-dominant prevalence of acute massive submacular hemorrhages (SMHs) caused by age-related macular degeneration (AMD) disappeared, and those caused by retinal microaneurysms (RMAs) emerged. METHOD: The medical charts of 95 patients (95 eyes) with SMH caused by AMD and 76 patients (76 eyes) with SMH caused by RMAs in 2012-2019 were retrospectively reviewed. For each subject, the month of onset, the mean ambient temperature of that month were recorded. RESULTS: The monthly numbers of cases of SMHs caused by AMD from January to December were 6, 8, 4, 9, 7, 10, 9, 11, 7, 11, 3, and 10. No significant seasonal variation in the monthly incidence was identified (Roger's R = 1.89, p = 0.39). The monthly numbers of SMHs caused by RMAs from January to December were 3, 11, 11, 8, 7, 8, 5, 5, 2, 4, 7, and 5. There was significant seasonal variation in the monthly incidence (Roger's R = 7.67, p = 0.02). There was no significant correlation between the monthly incidence of SMHs caused by RMAs and mean ambient temperature. CONCLUSION: Our previous study conducted for cases obtained in 1998-2005 showed seasonal cyclic trend in the number of SMHs caused by AMD, with the peak in winter. However, that significant seasonal variation disappeared in 2012-2019 in the present study. Common usage of OCT devices and anti-VEGF drugs might be the reason for the lack of seasonal variation in the cases of SMH caused by AMD.

BACKGROUND: Diabetes is a major risk factor for cardiovascular diseases (CVD). This systematic review aims to explore the work-related factors among people with diabetes in developing CVD. METHODS: Four electronic databases were searched on 1 February 2021 using a comprehensive search strategy without any time restriction. Two independent researchers screened the articles and extracted data. The risk of bias was assessed independently using the risk of bias assessment tool for non-randomized studies (RoBANS). A narrative synthesis was conducted considering the heterogeneity of the included articles. RESULTS: A total of five articles incorporating 4 409 810 participants from three geographic regions were included that highlights the research gap. As per the included studies, Occupational drivers with diabetes were at a higher risk of CVD in comparison to the nondrivers, workers with diabetes having a long working hour were at a higher risk of CVD mortality, workers with a lower occupational status were at a higher risk of 10-years stroke risk, and occupational physical activity and occupational commuting lowered the risk of CVD deaths. CONCLUSIONS: This systematic review summarized the available evidence on work-related factors influencing the risk of CVD in people with diabetes. The findings should be interpreted cautiously pondering the limited evidence and imprecision. We identified only five articles related to the topic, and there were no studies from Japan. The scarcity of studies on work-related factors on the prognosis of diabetic patients implies the need for more research in this field. We recommend further exploration of the topic designing primary studies.

BACKGROUND: The aim of this study was to determine the effectiveness of the Smart Life Stay (SLS) program, which is an experience-oriented stayover program, in combination with health tourism and mandatory health guidance on glucose metabolism after 2 years. METHODS: The participants of the SLS program (n = 792) were recruited from a database of 23 medical insurers. They underwent a mandatory health examination termed Specific Health Checkups in 2014. The participants were included if they had diabetes or were at a high risk of diabetes and if they satisfied the following inclusion criteria: (1) body mass index (BMI; kg/m2) > 25, or (2) waist circumference (WC; cm) > 85 for men and > 90 for women, or (3) hemoglobin A1c (HbA1c; %) > 5.6, or (4) fasting plasma glucose (FPG; mg/dl) > 100. Individuals who corresponded to one or more items were included as study participants. The control subjects (n = 3645) were nonparticipants of the program who were selected from the database and met the inclusion criteria. The lifestyle changes and changes in mean BMI, WC, FPG, and HbA1c in both groups from baseline to 2-year follow-up were compared by inverse probability weighting of a propensity score. RESULTS: The percentage of people who exercised regularly increased significantly in the SLS group compared with the control group. In the SLS group, BW, BMI, and WC significantly decreased by 1.75 kg, 0.60 kg/m2, and 1.45 cm, respectively, whereas in the control group, WC, FPG, and HbA1c increased significantly by 0.38 cm, 3.37 mg/dl, and 0.12%, respectively. The comparison between groups revealed that the BW, BMI, WC, FPG, and HbA1c improved significantly in the SLS group. CONCLUSIONS: The SLS program is suggested to help improve glucose metabolism. This program could be a feasible option as a lifestyle intervention program for diabetes.

BACKGROUND: Relatively little evidence exists for type 2 diabetes mellitus (T2DM) prediction models from long-term follow-up studies in East Asians. This study aims to develop a point-based prediction model for 10-year risk of developing T2DM in middle-aged Japanese men. METHODS: We followed 3,540 male participants of Aichi Workers' Cohort Study, who were aged 35-64 years and were free of diabetes in 2002, until March 31, 2015. Baseline age, body mass index (BMI), smoking status, alcohol consumption, regular exercise, medication for dyslipidemia, diabetes family history, and blood levels of triglycerides (TG), high density lipoprotein cholesterol (HDLC) and fasting blood glucose (FBG) were examined using Cox proportional hazard model. Variables significantly associated with T2DM in univariable models were simultaneously entered in a multivariable model for determination of the final model using backward variable selection. Performance of an existing T2DM model when applied to the current dataset was compared to that obtained in the present study's model. RESULTS: During the median follow-up of 12.2 years, 342 incident T2DM cases were documented. The prediction system using points assigned to age, BMI, smoking status, diabetes family history, and TG and FBG showed reasonable discrimination (c-index: 0.77) and goodness-of-fit (Hosmer-Lemeshow test, P = 0.22). The present model outperformed the previous one in the present subjects. CONCLUSION: The point system, once validated in the other populations, could be applied to middle-aged Japanese male workers to identify those at high risk of developing T2DM. In addition, further investigation is also required to examine whether the use of this system will reduce incidence.

OBJECTIVE: To examine whether laughter yoga (LY), i.e., simulated laughter, alters cortisol and dehydroepiandrosterone (DHEA) levels and cortisol/DHEA (C/D) ratios. METHODS: In a randomized controlled trial, 120 healthy university students were allocated to experiencing LY, watching a comedy movie (spontaneous laughter), or reading a book. Salivary cortisol and DHEA levels were measured immediately before, immediately after, and 30 min after the intervention. RESULTS: Cortisol levels and C/D ratios significantly decreased by time in the LY and comedy movie groups. Significant group*time interactions were found between these two groups for cortisol levels and C/D ratios. DHEA levels did not change by time in the LY group. CONCLUSIONS: LY decreased cortisol levels and C/D ratios but did not affect DHEA levels. Simulated and spontaneous laughter differently affected the dynamics of cortisol levels and C/D ratios. Effect of spontaneous laughter on the cortisol dynamics lasted longer than that of simulated laughter. (UMIN000019409).

国際的には低い冠動脈疾患発症率を誇るわが国ではあるが、肥満や不活発な生活習慣の蔓延による耐機能異常、脂質異常の割合の増加により、男性、特に都市部男性における発症率の増加が示唆されている。わが国における冠動脈疾患の危険因子(高血圧、脂質異常症、糖尿病、喫煙、肥満、メタボリックシンドローム、慢性腎臓病、飲酒)について、定義、冠動脈疾患発症、死亡との関連の特徴などについて個々に概説した。

癌サバイバーである日本人労働者の機能的能力(FC)、自己評価健康状態(SRHS)および心理社会的特性(PSC)を評価し、これらの特性を癌サバイバー労働者(癌既往群)と癌無病歴労働者(癌非既往群)との間で比較した。対象は2113年の地方自治体職員5474名(癌既往群112名)で、自己記入質問票を用いてFC、SRHSおよびPSC(社会的支援、自覚ストレス、社会資本、生き甲斐、幸福)における制限を評価し、癌病歴がこれらの制限に関連するか否か検討した。その結果、癌既往群では全年齢の男性と比較的若い女性がいずれも対応する癌非既往群よりFCの制限が多く(全年齢の男性14.5%対2.9%:50歳未満の女性15.2%対1.1%)、全年齢の男性が非既往群に比べてSRHSが悪かった(8.1%対1.5%)。一方、PSCは両群間で有意な差が見られなかった。以上より、男性と比較的若い女性の癌サバイバーである労働者はFCの制限が多く、男性ではSRHSも悪いと考えられた。

BACKGROUND AND AIMS: Coronary heart disease (CHD) and stroke have common risk factors, but some of these differ in the magnitude or direction of associations between CHD and stroke. We assessed whether the impact of each risk factor differed between CHD and stroke mortality in Asians. METHODS: In total, 104 910 subjects aged 40-79 years without histories of cancer, CHD and stroke at baseline were followed between 1988 and 2009. Competing-risks analysis was used to test for differences in the associations of each risk factor with two endpoints (CHD and stroke). Population attributable fractions (PAFs) were also calculated for these endpoints to estimate the population impact of each risk factor. RESULTS: During a median 19.1-year follow-up, 1554 died from CHD and 3163 from stroke. The association of hypertension with CHD was similar to that with stroke in terms of the magnitude and direction (multivariable-adjusted hazard ratio for CHD: 1.63 vs. stroke: 1.73 in men and 1.70 vs. 1.66 in women). Conversely, the magnitude of these associations differed for smoking (CHD: 1.95 vs. stroke: 1.23 in men and 2.45 vs. 1.35 in women) and diabetes (1.49 vs. 1.09 in men and 2.08 vs. 1.39 in women). The highest PAF for CHD was caused by smoking in men and by hypertension in women; that for stroke was caused by hypertension in both sexes. CONCLUSIONS: Hypertension associations and PAFs were consistent between CHD and stroke, but not for other risk factors. These findings may be useful to optimize public health intervention strategies.

We aimed to conduct a cross-sectional study in order to investigate the effect of periodic limb movements during sleep (PLMS) on daytime sleepiness in patients with sleep disordered breathing especially those with obstructive sleep apnea syndrome (OSAS), diagnosed by polysomnography (PSG). Our subjects included 233 male patients who visited our institute during the 25-month period between June 2012 and June 2013. We scored the number of Periodic Limb Movements per hour as the Periodic Limb Movements Index (PLMI) on standard PSG, and considered that periodic limb movements during sleep was present in the patients with PLMI >15/h. Non-PLMS group included those patients with PLMI <= 15/h. To assess the sleepiness of participants, we used Japanese version of the Epworth Sleepiness Scale (JESS) and Japanese version of the Pittsburgh Sleep Quality Index. We compared the sleep parameters and sleepiness indicators between patients with PLMS and age, apnea-hypopnea index (AHI), and body mass index-matched non-PLMS control groups. JESS scores were higher in patients with PLMS than in non-PLMS in both low AHI (<15/h) and high AHI (>= 15/h) strata. However, the difference was statistically significant only in high AHI stratum (P = 0.399 and 0.001, respectively). In conclusion, PLMS was associated with increased JESS especially in patients with moderate-to-severe OSAS. Although the magnitude of the difference in those with AHI <15/h was smaller and statistically insignificant in the present study, this issue warrants further investigation by enrolling more patients with AHI <15/h.

Background: To examine the association of intrahepatic fat with homeostasis model assessment-insulin resistance (HOMA-IR), a marker of insulin resistance, in Japanese adults, and whether intrahepatic fat is associated with insulin resistance independent of waist circumference and other measures of obesity.Methods: Fifty-three individuals aged 37-69 were studied. Spectrum obtained using a 3-T magnetic resonance imager was analysed with LC Model to quantify intrahepatic fat. Blood levels of insulin, glucose and other biochemical markers were obtained after 8 h or more fasting. Percent body fat was estimated by a bioelectrical impedance analyzer. HOMA-IR and intrahepatic fat content were log-transformed in the analysis.Results: We found a positive correlation between intrahepatic fat and HOMA-IR, which was independent of the anthropometric measures of obesity. In contrast, significant and positive correlations of body mass index, percent body fat, and waist circumference with HOMA-IR were largely explained by their associations with intrahepatic fat. Intrahepatic fat was positively associated with alanine transaminase and triglyceride seven after adjustment for HOMA-IR.Conclusion: Intrahepatic fat was associated with insulin resistance independent of age, sex, and measures of obesity in Japanese adults. Hypertriglyceridemia and liver injury may directly occur subsequent to intrahepatic fat accumulation. (C) 2013 Asian Oceanian Association for the Study of Obesity. Published by Elsevier Ltd. All rights reserved.