Prevalence and Risk Factors Comparison of Anterior and Posterior Intracranial Arterial Stenosis

The prevalence and risk factors of intracranial atherosclerotic stenosis (ICAS) located in the anterior circulation (AC) and posterior circulation (PC) has been scarcely noted in the general population. We aimed to determine ICAS prevalence and risk factor profile of AC and PC in a representative population. Data were from the China Hypertension Survey of Beijing. In total, 4800 people aged 35 years or older were enrolled in this subsurvey for ICAS, and 3954 participants were eligible for analysis. ICAS was assessed by transcranial Doppler. The prevalence of ICAS in AC was much greater than that in PC (11.9% vs. 4.2%), and subjects with ICAS in PC were 3.9 years older than those with ICAS in AC. Multivariable logistics regression showed that the odds of hypertension and diabetes increased by 79% (OR: 1.79, 95% CI: 1.40–2.27) and 35% (OR: 1.35, 95% CI: 1.04–1.75) in those with AC vascular lesions and by 3.35 times (OR: 3.35, 95% CI: 2.49–4.50) and 71% (OR: 1.71, 95% CI: 1.19–2.46) in those with PC vascular lesions compared with those without vascular lesions. Most modifiable vascular risk factors for ICAS appeared to exert similar magnitudes of risk for PC to AC lesions.

Optimal strategy for supplier selection in a global supply chain using machine learning technique

This paper proposes an optimization strategy for the best selection process of suppliers. Based on recent literature reviews, the paper assumes a selection of commonly used variables for selecting suppliers, and using Logistic regression algorithm technique, to build a model of optimization that learns from customer’s requirements and supplier’s data, and then make predictions and recommendations for best suppliers. The supplier selection process can quickly at times, turn into a complex task for decision-makers, to dealing with the growing number of supplier base list. But Logistics regression technique makes the process easier in the ability to efficiently fetch customer’s requirements with the entire supplier base list and determine by predicting a list of potential suppliers meeting the actual requirements. The selected suppliers make up the recommendation list for the best suppliers for the requirements. And finally, graphical representations are given to showcase the framework analysis, variable selection, and other illustrations about the model analysis

Computer-Aided Medical Diagnosis System Using Logistics Regression Algorithms (LRA) Supervised Learning Approach

This work focused on the designing of medical diagnosis system using Supervised Machine Learning. Logistics Regression Algorithms (LRA) was adopted, the label inputs for the data set which the symptoms were trained and mapped with the input of the user. Diagnosis of malaria was considered in this work; the system verified the value of the logical regression in the medical decision support system. Medical practitioners and other health workers can use this system to make better decisions in medical diagnosis for malaria. Adoption of this system will reduce stress of diagnoses malaria from patient and reduce congestion in our hospitals.

Information and Husband’s Support on Implementation Postpartum Gymnastics during COVID 19 Pandemic in Indonesia

Background : The practice of postnatal gymnastics in Indonesia is still little done by postpartum mothers both before and during the covid pandemic 19. Some of the contributing factors include lack of knowledge about postnatal gymnastics, fatigue, and the mother's mindset about the importance of postnatal gymnastics for health which is still lacking. Post-delivery exercise can help mothers to get back to their normal health as they were before pregnancy. Objective : This study aims to analyze the effect of information and husband's support on the implementation postnatal gymnastics during the covid 19 pandemic. Methods : Non-experimental research design: cross sectional analytic, the independent variable is husband's information and support and the dependent variable is postpartum exercise. This research was conducted in April – August 2021. The population of postpartum mothers was 102, the sampling technique was incidental sampling in Wonokromo Village, Surabaya, East Java, Indonesia. Research instrument with a questionnaire. The data were analyzed using the Multiple Logistics Regression test. Results : The results showed that partially there was a significant effect of husband's support (p = 0.013) on postnatal exercise, but for information during the pandemic there was no effect (p = 0.998) on postpartum exercise. The results of the analysis showed that the OR value of the information was 1.615E9 (95%CI: 0.000 – 0.000) and the OR of husband's support was 3.385 (95%CI: 1.289 – 8.886). Conclusion : Husband's support during the covid 19 pandemic affects mothers in doing postnatal exercise. Obtaining information about postnatal exercise during the COVID-19 pandemic does not affect the mother in its implementation.

Neighborhood Greenspace and Cognition: The Cardiovascular Health Study

Access to greenspace has been positively associated with cognition among older adults, however prior research has been limited in temporal and geographic scope. We evaluated associations between neighborhood greenspace and incidence of dementia and change in cognitive functioning using a longitudinal sample of non-demented adults (n=2,465) from the Cardiovascular Health and Cognition Study. Percent greenness (1-km radial buffers) was derived from the National Land Cover Dataset. Cognitive function was measured using the Mini-Mental State Examination (3MSE) and dementia status was clinically adjudicated. Cox proportional hazard and logistics regression analyses were used to examine associations of baseline greenness with risk of incident dementia and risk of mild cognitive impairment, respectively. Generalized linear mixed models accounting for within-subject correlations were used to examine the association between greenspace in the neighborhood at baseline and 3MSE score (1991-1999). Ongoing results will be presented, along with modifiers and mediators of associations.

Clinical features and major bleeding predictors for 161 fatal cases of COVID-19: A retrospective observational study

The aim of this study was to investigate the patient characteristics and laboratory parameters for COVID-19 non-survivors as well as to find risk factors for major bleeding complications. For this retrospective study, the data of patients who died with COVID-19 in our intensive care unit were collected in the period of March 20 - April 30, 2020. D-dimer, platelet count, C-reactive protein (CRP), troponin, and international normalized ratio (INR) levels were recorded on the 1st, 5th, and 10th days of hospitalization in order to investigate the possible correlation of laboratory parameter changes with in-hospital events. A total of 161 non-survivors patients with COVID-19 were included in the study. The median age was 69.8±10.9 years, and 95 (59%) of the population were male. Lung-related complications were the most common in-hospital complications. Patients with COVID-19 had in-hospital complications such as major bleeding (39%), hemoptysis (14%), disseminated intravascular coagulation (13%), liver failure (21%), ARDS (85%), acute kidney injury (40%), and myocardial injury (70%). A multiple logistics regression analysis determined that age, hypertension, diabetes mellitus, use of acetylsalicylic acid (ASA) or low molecular weight heparin (LMWH), hemoglobin, D-dimer, INR, and acute kidney injury were independent predictors of major bleeding. Our results showed that a high proportion of COVID-19 non-survivors suffered from major bleeding complications.

Environmental Risk Factors for Talaromycosis Hospitalizations of HIV-Infected Patients in Guangzhou, China: Case Crossover Study

Talaromycosis is a fatal opportunistic infection prevalent in human immunodeficiency virus (HIV)-infected patients, previous studies suggest environmental humidity is associated with monthly talaromycosis hospitalizations of HIV-infected patients, but the acute risk factor remains uncertain. In this study, we evaluated the associations between talaromycosis hospitalizations of HIV-infected patients (n = 919) and environmental factors including meteorological variables and air pollutants at the event day (assumed “lag 0” since the exact infection date is hard to ascertain) and 1–7 days prior to event day (lag 1–lag 7) in conditional logistics regression models based on a case crossover design. We found that an interquartile range (IQR) increase in temperature at lag 0–lag 7 (odds ratio [OR] [95% CI] ranged from 1.748 [1.345–2.273] to 2.184 [1.672–2.854]), and an IQR increase in humidity at lag 0 (OR [95% CI] = 1.192 [1.052–1.350]), and lag 1 (OR [95% CI] = 1.199 [1.056–1.361]) were significantly associated with talaromycosis hospitalizations of HIV-infected patients. Besides, temperature was also a common predictor for talaromycosis in patients with co-infections including candidiasis (n = 386), Pneumocystis pneumonia (n = 183), pulmonary tuberculosis (n = 141), and chronic hepatitis (n = 158), while humidity was a specific risk factor for talaromycosis in patients with candidiasis, and an air pollutant, SO2, was a specific risk factor for talaromycosis in patients with Pneumocystis pneumonia. In an age stratified evaluation (cutoff = 50 years old), temperature was the only variable positively associated with talaromycosis in both younger and older patients. These findings broaden our understanding of the epidemiology and pathogenesis of talaromycosis in HIV-infected patients.

Towards the Development of a Substance Abuse Index (SEI) through Informatics

Substance abuse or drug dependence is a prevalent phenomenon, and is on the rise in United States. Important contributing factors for the prevalence are the addictive nature of certain medicinal/prescriptive drugs, individual dispositions (biological, physiological, and psychological), and other external influences (e.g., pharmaceutical advertising campaigns). However, currently there is no comprehensive computational or machine learning framework that allows systematic studies of substance abuse and its factors with majority of the works using subjective surveys questionnaires and focusing on classification techniques. Lacking standardized methods and/or measures to prescribe medication and to study substance abuse makes it difficult to advance through collective research efforts. Thus, in this paper, we propose to test the feasibility of developing a, objective substance effect index, SEI, that can measure the tendency of an individual towards substance abuse. To that end, we combine the benefits of Electronics Medical Records (EMR) with machine learning technology by defining SEI as a function of EMR data and using logistics regression to obtain a closed form expression for SEI. We conduct various evaluations to validate the proposed model, and the results show that further work towards the development of SEI will not only provide researchers with standard computational measure for substance abuse, but may also allow them to study certain attribute interactions to gain further insights into substance abuse tendencies.

Abstract 12051: Impact of COVID-19 on Barriers to Dispatcher-Assisted Cardiopulmonary Resuscitation in Adult Out-of-Hospital Cardiac Arrests in Singapore

Objective: Reduced rates of bystander cardiopulmonary resuscitation (BCPR) in out-of-hospital cardiac arrest (OHCA) were observed during the Coronavirus Disease-2019 (COVID-19) pandemic in many regions. We investigated the impact of COVID-19 on barriers to Dispatcher-Assisted Cardiopulmonary Resuscitation (DA-CPR) in Singapore. Methods: This nationwide retrospective cohort study involved all calls to our national 995 call center for adult (≥ 18 years old) OHCA not witnessed by Emergency Medical Services. We reviewed audio recordings during the pandemic (January-June 2020) and pre-pandemic (January-June 2019) periods to compare the OHCA characteristics, and the types of “barriers” to DA-CPR — the reason why DA-CPR was not performed. Our primary outcome was the presence/absence of barriers to DA-CPR. Multivariable logistics regression was used to estimate the adjusted odds ratio (aOR) for the likelihood of barriers to DA-CPR accounting for patient and event characteristics. The effect of COVID-19 on DA-CPR rates was evaluated using interrupted time series analysis. Results: There were 1481 OHCA during the pandemic (median age 73 years, 62.7% male), and 1400 prior to the pandemic (median age 72 years, 63.6% male). Residential OHCA and witnessed OHCA increased during the pandemic (78.9% vs 75.5%, p=0.03 and 56.1% vs 39.9%, p<0.01 respectively), but not BCPR and DA-CPR (64.3% vs 65.6%, p=0.44, and 49.1% vs 48.1%, p=0.57 respectively). There were increased barriers to DA-CPR during the pandemic — ‘patient status changed’ (difficulty with recognition) and ‘caller not with patient’ (witnesses calling family rather than 995) doubled in proportion during COVID-19. ‘afraid to do CPR’ fell to 3.8% during the pandemic, while the fear of COVID-19 transmission made up 0.5% of the barriers. Logistic regression showed that females and OHCAs occurring in home residences were more likely to have barriers to DA-CPR (aOR 1.27 and 2.63 respectively). COVID-19 did not have an impact on the trend of DA-CPR rates (p=0.49). Conclusion: COVID-19 did not affect callers’ willingness to perform DA-CPR. Distancing measures led to more residential arrests with an increase in proportion with barriers to DA-CPR, highlighting opportunities for public education and intervention.