Objective: Human papillomavirus subtypes are predictive indicators of cervical intraepithelial neoplasia (CIN) progression. While colposcopy is also an essential part of cervical cancer prevention, its accuracy and reproducibility are limited because of subjective evaluation. This study aimed to develop an artificial intelligence (AI) algorithm that can accurately detect the optimal lesion associated with prognosis using colposcopic images of CIN2 patients by utilizing objective AI diagnosis. Methods: We identified colposcopic findings associated with the prognosis of patients with CIN2. We developed a convolutional neural network that can automatically detect the rate of high-grade lesions in the uterovaginal area in 12 segments. We finally evaluated the detection accuracy of our AI algorithm compared with the scores by multiple gynecologic oncologists. Results: High-grade lesion occupancy in the uterovaginal area detected by senior colposcopists was significantly correlated with the prognosis of patients with CIN2. The detection rate for high-grade lesions in 12 segments of the uterovaginal area by the AI system was 62.1% for recall, and the overall correct response rate was 89.7%. Moreover, the percentage of high-grade lesions detected by the AI system was significantly correlated with the rate detected by multiple gynecologic senior oncologists (r=0.61). Conclusion Our novel AI algorithm can accurately determine high-grade lesions associated with prognosis on colposcopic images, and these results provide an insight into the additional utility of colposcopy for the management of patients with CIN2.

BACKGROUND: The burden of dementia is growing rapidly and has become a medical and social problem in Japan. Prospective cohort studies have been considered an effective methodology to clarify the risk factors and the etiology of dementia. We aimed to perform a large-scale dementia cohort study to elucidate environmental and genetic risk factors for dementia, as well as their interaction. METHODS: The Japan Prospective Studies Collaboration for Aging and Dementia (JPSC-AD) is a multisite, population-based prospective cohort study of dementia, which was designed to enroll approximately 10,000 community-dwelling residents aged 65 years or older from 8 sites in Japan and to follow them up prospectively for at least 5 years. Baseline exposure data, including lifestyles, medical information, diets, physical activities, blood pressure, cognitive function, blood test, brain magnetic resonance imaging (MRI), and DNA samples, were collected with a pre-specified protocol and standardized measurement methods. The primary outcome was the development of dementia and its subtypes. The diagnosis of dementia was adjudicated by an endpoint adjudication committee using standard criteria and clinical information according to the Diagnostic and Statistical Manual of Mental Disorders, 3rd Revised Edition. For brain MRI, three-dimensional acquisition of T1-weighted images was performed. Individual participant data were pooled for data analyses. RESULTS: The baseline survey was conducted from 2016 to 2018. The follow-up surveys are ongoing. A total of 11,410 individuals aged 65 years or older participated in the study. The mean age was 74.4 years, and 41.9% were male. The prevalence of dementia at baseline was 8.5% in overall participants. However, it was 16.4% among three sites where additional home visit and/or nursing home visit surveys were performed. Approximately two-thirds of dementia cases at baseline were Alzheimer's disease. CONCLUSIONS The prospective cohort data from the JPSC-AD will provide valuable insights regarding the risk factors and etiology of dementia as well as for the development of predictive models and diagnostic markers for the future onset of dementia. The findings of this study will improve our understanding of dementia and provide helpful information to establish effective preventive strategies for dementia in Japan.
Aged ; Alzheimer Disease/genetics* ; Dementia/genetics* ; Environment ; Female ; Humans ; Incidence ; Japan/epidemiology* ; Male ; Middle Aged ; Prevalence ; Prospective Studies ; Risk Factors

We have previously reported on a predictive model for deficiency-excess pattern diagnosis that was unable to predict the medium pattern. In this study, we aimed to develop predictive models for deficiency, medium,and excess pattern diagnosis, and to confirm whether cutoff values for diagnosis differed between the clinics. We collected data from patients' first visit to one of six Kampo clinics in Japan from January 2012 to February 2015. Exclusion criteria included unwillingness to participate in the study, missing data, duplicate data, under 20 years old, 20 or less subjective symptoms, and irrelevant patterns. In total, 1,068 participants were included. Participants were surveyed using a 153-item questionnaire. We constructed a predictive model for deficiency, medium, and excess pattern diagnosis using a random forest algorithm from training data, and extracted the most important items. We calculated predictive values for each participant by applying their data to the predictive model, and created receiver operating characteristic (ROC) curves with excess-medium and medium-deficiency patterns. Furthermore, we calculated the cutoff value for these patterns in each clinic using ROC curves, and compared them. Body mass index and blood pressure were the most important items. In all clinics, the cutoff values for diagnosis of excess-medium and medium-deficiency patterns was > 0.5 and < 0.5, respectively. We created a predictive model for deficiency, medium, and excess pattern diagnosis from the data of six Kampo clinics in Japan. The cutoff values for these patterns fell within a narrow range in the six clinics.

BACKGROUND: Promotion of oral health in children is recognized as one of the components of health-promoting schools (HPSs). However, few studies have addressed supportive school environments for children's oral health. This study aimed to evaluate the status of dental caries in school children at HPSs, with the objective of examining the impact of a supportive school environment for oral health, considering the lifestyles of individual children and the socioeconomic characteristics of their communities. METHODS: Data of 2043 5th-grade students in 21 elementary schools in Ichikawa city between 2008 and 2013 were analyzed. Children's oral health status was evaluated using the decayed, missing, and filled permanent teeth (DMFT) index. A self-reported lifestyle questionnaire, a survey of the school environment promoting tooth-brushing, and community socioeconomic characteristics derived from the National Census data were included in the analyses. Bivariate analyses were conducted to evaluate the children's DMFT status, and zero-inflated negative binominal (ZINB) regression was used to assess the relationships between DMFT and other variables. RESULTS: Prevalence of dental caries in the permanent teeth of 5th-grade children (aged 10-11 years) was 33.3%, with a mean DMFT score (± SD) of 0.83 ± 1.50. According to multilevel ZINB regression analysis, children from schools with after-lunch tooth-brushing time showed a higher odds ratio (OR) for excess zero DMFT (OR = 1.47, 95% CI = 1.00-2.15, P = 0.049) as compared to those from schools without it. Neither bivariate analysis nor ZINB model analysis revealed any significant influence of children's gender or use of a toothpaste with fluoride. CONCLUSIONS The school-based environment supportive of oral health was significantly associated with a zero DMFT status in children. School-based efforts considering the socioeconomic characteristics of the area warrant attention even with declining prevalence of dental caries.

BACKGROUND/AIMS: Colonic diverticular bleeding can stop spontaneously or be stopped by endoscopic hemostasis. We analyzed the clinical and colonoscopic features of patients with colonic diverticular bleeding to establish the predictive factors for rebleeding. METHODS: A total of 111 patients (median age, 72 years) with colonic diverticular bleeding in Aso Iizuka Hospital between April 2007 and July 2010 were enrolled. Age, sex, body mass index (BMI), comorbidity, medication, location of bleeding, colonoscopic findings and hemostatic methods were analyzed retrospectively from the hospital records. RESULTS: The most common sites of bleeding were the ascending (39.6%) and sigmoid (29.7%) colon. Overt rebleeding occurred in 30 patients (27.0%). Spontaneous hemostasis was seen in 81 patients (73.0%), and endoscopic hemostatic treatment was performed in 30 patients. The BMI in the patients with colonic diverticular rebleeding was significantly higher than in patients without rebleeding. Colonoscopic findings of actively bleeding or nonbleeding visible vessels in the responsible diverticula were more frequent in the group with rebleeding. CONCLUSIONS: A higher BMI and colonoscopic findings of actively bleeding or nonbleeding visible vessels can be used as predictive factors for colonic diverticular rebleeding. Patients with such findings should be carefully followed up after hemostasis of the initial colonic diverticular bleeding.
Body Mass Index ; Colon ; Colon, Sigmoid ; Comorbidity ; Diverticulum ; Hemorrhage ; Hemostasis ; Hemostasis, Endoscopic ; Humans ; Retrospective Studies

BACKGROUNDSeveral studies have suggested that the exposure to cadmium (Cd) increased mortalities from renal diseases, cardiovascular diseases and malignant neoplasm, including lung cancer and prostate cancer among inhabitants living in Cd-polluted areas and factory workers. This study aimed to assess the influence of environmental exposure to Cd on long term outcome of inhabitants living in an area polluted by Cd.

METHODSA 22-year follow-up study was conducted with 3119 inhabitants (1403 men and 1716 women) living in the Cd polluted Kakehashi River basin in Japan. The subjects were divided into 4 groups according to the amount of urinary Cd level (< 3.0 µg/g creatinine (Cr), 3.0 - 4.9 µg/g Cr, 5.0 - 9.9 µg/g Cr, and ≥ 10.0 µg/g Cr). Mortality was calculated by the person-years method. Hazards ratios (HR) and 95% confidence intervals (CI) were assessed by the Cox's proportional hazard model.

RESULTSCompared with urinary Cd < 3.0 µg/g Cr group, the HR of 5.0 - 9.9 µg/g Cr and ≥ 10.0 µg/g Cr groups were significantly increased after adjustment for age in both sexes: 1.24 (95%CI 1.01 - 1.51) and 1.48 (95%CI 1.17 - 1.90) for men; 1.64 (95%CI 1.17 - 2.28) and 1.78 (95%CI 1.27 - 2.50) for women. The most frequent cause of death was malignant neoplasm in men and cardiovascular diseases in women. The significant increase in mortality risk for cardiovascular diseases was observed in the subjects with ≥ 10 µg/g Cr in both sexes: 1.79 for men (95%CI 1.02 - 3.12) and 2.38 for women (95%CI 1.11 - 5.07). When the subjects were divided into 2 categories (< 20 µg/g Cr and ≥ 20 µg/g Cr), the HR of the urinary Cd ≥ 20 µg/g Cr group for nephritis and nephrosis were 4.82 (95%CI 1.07 - 21.61) in men and 7.92 (95%CI 1.77 - 35.33) in women, respectively. The significant increase was not observed for malignant neoplasm.

CONCLUSIONThese results suggest a dose-response relationship between Cd body burden and mortality for cardiovascular diseases, cerebrovascular diseases and nephritis and nephrosis.

Aged ; Aged, 80 and over ; Cadmium ; toxicity ; Cardiovascular Diseases ; mortality ; Cerebrovascular Disorders ; mortality ; Environmental Exposure ; adverse effects ; Female ; Follow-Up Studies ; Hazardous Substances ; Humans ; Japan ; Male ; Middle Aged ; Nephritis ; mortality ; Nephrosis ; mortality ; Proportional Hazards Models ; Risk Factors

  The narcotic drug management system using drug information stored in electronic medical cards has been introduced by many medical facilities. The recording of the history of narcotics handling, such as delivery and returning, is essentialfor the management of narcotic drugs. However, the hitherto used system was hard to use for recording the handling history in detail. The new system developed by us was designed so as to record the handling history including the information on “drug execution (actualadministration)” and “drug request (prescription)” in the electronic card with ease. Using the order number and bar code of narcotic drugs in the card, the new systemmade it possible to prepare the narcotic book precisely and efficiently. The introduction of this system in our hospital has brought about a significant improvement in the management of narcotic drugs.

A study was conducted to examine the relationship between percent body fat (%fat) and medical examination parameters. The subjects were 250 women aged 17 to 68 years, who neither drank nor smoked. The results obtained were as follows:
After controlling for the effects of age and maximal oxygen intake per kilogram body weight (VO₂max/wt), %fat showed a significant correlation with high-density lipoprotein cholesterol (HDL-C), total cholesterol/HDL-C ratio (TC/HDL-C), β-lipoprotein (β-L), glutamic oxaloacetic transaminase (GOT), glutamic pyruvic tansaminase (GPT), uric acid (UA), white blood cell count (WBC), red blood cell count (RBC), systolic blood pressure (SBP), and diastolic blood pressure (DBP) . Values of %fat calculated from each regression equation when HDL-C=38mg/dl, TC/HDL-C=4.9, β-L=500 mg/dl, GOT=41mu/ml, GPT=46 mu/ml, UA=5.7mg/dl, WBC=8500/mm3, RBC=520×104/mm3, SBP=159 mmHg, and DBP=94 mmHg were 32.2%, 31.9%, 30.8%, 35.4%, 36.4%, 31.8%, 30.7%, 35.0%, 33.8% and 32.6%, respectively.
The subjects were then divided into 2 groups, above (above group) or below (below group) 30%, 31%, 32%, 33%, 34% or 35% body fat. Wherever the subjects were divided into 2 groups at these values of %fat, the above group showed a significantly higer occurrence of abnormal values for medical parameters than the below group.
The subjects were further divided into 3 groups: 154 subjects with %fat below 29.9% (N group), 47 subjects with %fat between 30% and 34.9% (MO group), and 49 subjects with %fat above 35% (O group) . The O group showed significantly higher occurrence of abnormal values for medical parameters than the MO and N groups. The O group also showed a significantly higher mean value of DBP than the MO and N groups. Furthermore, the O group showed a significantly lower mean value of HDL-C and significantly higher mean values of TC/HDL-C, UA, GOT, GPT, SBP, FBS than the N group. There was no significant difference between the N and MO groups in the occurrence of abnormal values for medical parameters. However, the MO group showed a significantly lower mean value of HDL-C and significantly higher mean values of TC/HDL-C, UA and β-L.
These results suggest that the values of medical parameters become poorer when %fat exceeds 30%, a level widely used in Japan as a criterion for evaluation of obesity.
The subjects were further divided into 3 groups: 50 subjects who were non-obese and fit (NF group) ; 104 subjects who were non-obese and unfit (NU group) ; 88 subjects who were obese and unfit (OU group) . There were significant differences in the occurrence of abnormal values for medical parameters among these groups. The NF group showed the lowest values, followed by the NU and OU groups. Thus, it seems preferable to evaluate individuals using a combination of %fat and VO₂max/wt.

A study was conducted to derive a criterion for evaluation of obesity based on relationships between percent body fat (%fat) and clinical parameters. The subjects were 457 men aged 18 to 73 years. The results obtained were as follows:
After controlling for the effects of age, maximal oxygen intake per kilogram body weight (VO₂max/wt), alcohol intake and cigarette smoking, %fat showed significant correlations with high-density lipoprotein cholesterol (HDL-C), total cholesterol/HDL-C ratio (TC/HDL-C), fasting blood sugar (FBS), glutamic oxaloacetic transaminase (GOT), glutamic pyruvic transaminase (GPT), γ-glutamyl transpeptidase (γ-GTP), white blood cell count (WBC), red blood cell count (RBC), systolic blood pressure (SBP), and diastolic blood pressure (DBP) . Values of %fat estimated from borderline values of these clinical parameters ranged from 17.9% to 20.4%.
When the subjects were divided into 11 groups from a below 12% fat group to an above 22% fat group, the 18%, 20%, 21% and above 22% fat groups showed significantly higher occurrence of abnormal values for clinical parameters than the below 12% fat group.
When the subjects were divided into 2 groups, i, e, above (above group) or below (below group) 18%, 19%, 20%, or 21% body fat, the above group always showed a significantly higher occurrence of abnormal values for clinical parameters than the below group.
The subjects were further divided into 3 groups: 250 subjects with %fat below 17.9% (N group), 63 subjects with %fat between 18% and 19.9% (MO group), and 144 subjects with %fat above 20% (O group) . The O group showed the highest occurrence of abnormal values for clinical parameters, followed in order by the MO and N groups. These differences were significant. The O group showed a significantly lower mean value of HDL-C and significantly higher mean values of TC, TC/HDL-C, TG, GPT, SBP, DBP and FBS than the N group. The O group also showed significantly higher mean values for WBC and RBC than the MO and N groups. The O and MO groups showed significantly higer mean values of β-L, γ-GTP and UA than the N group.
From these results and the fact that a value above 20% fat is widely used as a criterion for evaluation of obesity, men with a %fat range of 18.0% to 19.9% are defined as mildly obese, whereas those with above 20% fat are defined as obese.
The subjects were further divided into 3 gooups: %fat below (non-obese) or above (obese) 20%, and VO₂max/wt above (fit) or below (unfit) values recommended by the Ministry of Public Welfare for each age group. One hundred fifty-one subjects were non-obese and fit (NF group), 142 subjects were non-obese and unfit (NU group) and 107 subjects were obese and unfit (OU group) . The NF group showed the lowest occurrence of abnormal values for clinical parameters, followed in order by the NU and OU groups. These differences were significant. Thus, it seems appropriate to evaluate obesity by using a combination of %fat and VO₂max/wt.