scholarly journals Post-stroke Anxiety Analysis via Machine Learning Methods

2021 ◽  
Vol 13 ◽  
Author(s):  
Jirui Wang ◽  
Defeng Zhao ◽  
Meiqing Lin ◽  
Xinyu Huang ◽  
Xiuli Shang
Keyword(s):  
Diabetes Mellitus ◽  
Machine Learning ◽  
Risk Factors ◽  
Logistic Regression ◽  
Ischemic Stroke ◽  
Multivariate Logistic Regression ◽  
Learning Methods ◽  
Anxiety Group ◽  
Post Stroke ◽  
Machine Learning Methods

Post-stroke anxiety (PSA) has caused wide public concern in recent years, and the study on risk factors analysis and prediction is still an open issue. With the deepening of the research, machine learning has been widely applied to various scenarios and make great achievements increasingly, which brings new approaches to this field. In this paper, 395 patients with acute ischemic stroke are collected and evaluated by anxiety scales (i.e., HADS-A, HAMA, and SAS), hence the patients are divided into anxiety group and non-anxiety group. Afterward, the results of demographic data and general laboratory examination between the two groups are compared to identify the risk factors with statistical differences accordingly. Then the factors with statistical differences are incorporated into a multivariate logistic regression to obtain risk factors and protective factors of PSA. Statistical analysis shows great differences in gender, age, serious stroke, hypertension, diabetes mellitus, drinking, and HDL-C level between PSA group and non-anxiety group with HADS-A and HAMA evaluation. Meanwhile, as evaluated by SAS scale, gender, serious stroke, hypertension, diabetes mellitus, drinking, and HDL-C level differ in the PSA group and the non-anxiety group. Multivariate logistic regression analysis of HADS-A, HAMA, and SAS scales suggest that hypertension, diabetes mellitus, drinking, high NIHSS score, and low serum HDL-C level are related to PSA. In other words, gender, age, disability, hypertension, diabetes mellitus, HDL-C, and drinking are closely related to anxiety during the acute stage of ischemic stroke. Hypertension, diabetes mellitus, drinking, and disability increased the risk of PSA, and higher serum HDL-C level decreased the risk of PSA. Several machine learning methods are employed to predict PSA according to HADS-A, HAMA, and SAS scores, respectively. The experimental results indicate that random forest outperforms the competitive methods in PSA prediction, which contributes to early intervention for clinical treatment.

Why so GLUMM? Detecting depression clusters through graphing lifestyle-environs using machine-learning methods (GLUMM)

2017 ◽  
Vol 39 ◽  
pp. 40-50 ◽  
Author(s):  
J.F. Dipnall ◽  
J.A. Pasco ◽  
M. Berk ◽  
L.J. Williams ◽  
S. Dodd ◽  
...  
Keyword(s):  
Machine Learning ◽  
Risk Factors ◽  
Heterogeneous Data ◽  
Multivariate Logistic Regression ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Health And Nutrition ◽  
Unhealthy Eating ◽  
Health Factors ◽  
Mental Health Factors

AbstractBackgroundKey lifestyle-environ risk factors are operative for depression, but it is unclear how risk factors cluster. Machine-learning (ML) algorithms exist that learn, extract, identify and map underlying patterns to identify groupings of depressed individuals without constraints. The aim of this research was to use a large epidemiological study to identify and characterise depression clusters through “Graphing lifestyle-environs using machine-learning methods” (GLUMM).MethodsTwo ML algorithms were implemented: unsupervised Self-organised mapping (SOM) to create GLUMM clusters and a supervised boosted regression algorithm to describe clusters. Ninety-six “lifestyle-environ” variables were used from the National health and nutrition examination study (2009–2010). Multivariate logistic regression validated clusters and controlled for possible sociodemographic confounders.ResultsThe SOM identified two GLUMM cluster solutions. These solutions contained one dominant depressed cluster (GLUMM5-1, GLUMM7-1). Equal proportions of members in each cluster rated as highly depressed (17%). Alcohol consumption and demographics validated clusters. Boosted regression identified GLUMM5-1 as more informative than GLUMM7-1. Members were more likely to: have problems sleeping; unhealthy eating; ≤ 2 years in their home; an old home; perceive themselves underweight; exposed to work fumes; experienced sex at ≤ 14 years; not perform moderate recreational activities. A positive relationship between GLUMM5-1 (OR: 7.50, P < 0.001) and GLUMM7-1 (OR: 7.88, P < 0.001) with depression was found, with significant interactions with those married/living with partner (P = 0.001).ConclusionUsing ML based GLUMM to form ordered depressive clusters from multitudinous lifestyle-environ variables enabled a deeper exploration of the heterogeneous data to uncover better understandings into relationships between the complex mental health factors.

scholarly journals Development and validation of prognosis model of mortality risk in patients with COVID-19

2020 ◽  
Vol 148 ◽  
Author(s):  
Xuedi Ma ◽  
Michael Ng ◽  
Shuang Xu ◽  
Zhouming Xu ◽  
Hui Qiu ◽  
...  
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Clinical Features ◽  
Mortality Risk ◽  
Operating Characteristics ◽  
Multivariate Logistic Regression ◽  
Learning Methods ◽  
Data Set ◽  
Machine Learning Methods ◽  
Better Than

Abstract This study aimed to identify clinical features for prognosing mortality risk using machine-learning methods in patients with coronavirus disease 2019 (COVID-19). A retrospective study of the inpatients with COVID-19 admitted from 15 January to 15 March 2020 in Wuhan is reported. The data of symptoms, comorbidity, demographic, vital sign, CT scans results and laboratory test results on admission were collected. Machine-learning methods (Random Forest and XGboost) were used to rank clinical features for mortality risk. Multivariate logistic regression models were applied to identify clinical features with statistical significance. The predictors of mortality were lactate dehydrogenase (LDH), C-reactive protein (CRP) and age based on 500 bootstrapped samples. A multivariate logistic regression model was formed to predict mortality 292 in-sample patients with area under the receiver operating characteristics (AUROC) of 0.9521, which was better than CURB-65 (AUROC of 0.8501) and the machine-learning-based model (AUROC of 0.4530). An out-sample data set of 13 patients was further tested to show our model (AUROC of 0.6061) was also better than CURB-65 (AUROC of 0.4608) and the machine-learning-based model (AUROC of 0.2292). LDH, CRP and age can be used to identify severe patients with COVID-19 on hospital admission.

scholarly journals Factors associated with depression in the patients with diabetes mellitus type 2

2019 ◽  
Vol 76 (11) ◽  
pp. 1178-1183 ◽  
Author(s):  
Admir Sabanovic ◽  
Natasa Maksimovic ◽  
Mirjana Stojanovic-Tasic ◽  
Marijan Bakic ◽  
Anita Grgurevic
Keyword(s):  
Diabetes Mellitus ◽  
Risk Factors ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Logistic Regression Analysis ◽  
Diabetes Mellitus Type 2 ◽  
Multivariate Logistic Regression Analysis ◽  
Multivariate Logistic Regression ◽  
Factors Associated

Background/Aim. The assessment of association of depression and diabetes mellitus type 2 using the Patient Health Questionaire (PHQ-9) has not been done in Montenegro. The aim of this study was to assess the prevalence of depression in the patients with type 2 diabetes mellitus, and to identify the risk factors associated with the presence of depression. Methods. A cross-sectional study was conducted at the General Hospital in Bijelo Polje, from July to September, 2015. It included 70 patients over 35 years of age with the diagnosis of diabetes for at least six months. For the assessment of depression presence and intensity PHQ?9 was used. All variables associated with the presence of depression at a significance level of p < 0.05 were included into the final method of the multivariate logistic regression analysis. Results. Comorbidities were statistically significant more frequent among patients with depression (?2 = 5.40; p = 0.020). Duration of diabetes over five years was significantly associated with depression (?2 = 12.48; p < 0.001). Depression occurred more frequently among physically inactive subjects (?2 = 10.74; p = 0.005). The presence of diabetic polyneuropathy (?2 = 6.04; p = 0.014) and cataract (?2 = 5.351; p = 0.021) were also significantly associated with depression. A multivariate logistic regression analysis showed that the duration of diabetes over five years and presence of cataract were independently associated with depression. Conclusion. The risk factors for depression among the subjects with diabetes were disease duration more than five years and the presence of cataract. Since depression is a serious disease and can be a risk factor for many chronic diseases, the best way of prevention is its early detection and treatment.

scholarly journals Decreased eGFR Is Associated With Ischemic Stroke in Patients With Dilated Cardiomyopathy

2019 ◽  
Vol 25 ◽  
pp. 107602961986690 ◽  
Author(s):  
Yuqing Deng ◽  
Zhiqing Chen ◽  
Lili Hu ◽  
Zhenyan Xu ◽  
Jinzhu Hu ◽  
...  
Keyword(s):  
Risk Factors ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Ischemic Stroke ◽  
Dilated Cardiomyopathy ◽  
Logistic Regression Analysis ◽  
Multivariate Logistic Regression Analysis ◽  
Analysis Model ◽  
Multivariate Logistic Regression ◽  
Increased Risk

Dilated cardiomyopathy (DCM) is increasingly indicated as a cause of cardioembolic syndrome, in particular, cardioembolic ischemia stroke. However, the potential risk factors for stroke among DCM patients remain under investigated. DCM patients hospitalized from June 2011 to June 2016 were included. The cases were defined as the group of DCM patients with stroke compared with those without stroke. Clinical characteristic data were collected and compared between the two groups including demographic data, complicated diseases, echocardiography index, and laboratory parameters and estimated glomerular filtration rate (eGFR). A multivariate logistic regression analysis model adjusted by sex and age was used to explore the related risk factors for stroke in DCM patients. A total of 779 hospitalized patients with DCM were included. Of these, 55 (7.1%) had experienced a stroke. Significantly lower eGFR levels (68.03 ± 26.22 vs 79.88 ± 24.25 mL/min/1.73 m2, P = .001) and larger left atrial diameters (45.32 ± 7.79 vs 43.25 ± 7.11 mm, P = .04) were found in the group of patients having DCM with stroke compared to those without stroke. When the eGFR was categorized as eGFR >60, 30<eGFR≤ 60 and eGFR ≤ 30, there were more patients with 30<eGFR≤ 60 (30.9% vs 17.7%) and eGFR≤ 30 (9.1% vs 3.3%) in the ischemic stroke group ( P = 0.003). A multivariate logistic regression analysis model adjusted by sex and age showed that 30 <eGFR≤60 (odds ratio [OR]: 2.07, 95% confidence interval [CI]: [1.05-4.07], P = .035) and eGFR≤30 (OR: 4.04, 95% CI: [1.41-11.62], P = .009) were statistically associated with ischemic stroke in patients with DCM. It is concluded that decreased eGFR is significantly associated with an increased risk of ischemic stroke in patients with DCM.

scholarly journals Risk factors for periprocedural ischemic stroke following endovascular treatment of intracranial aneurysms

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Yisen Zhang ◽  
Chao Wang ◽  
Zhongbin Tian ◽  
Wei Zhu ◽  
Wenqiang Li ◽  
...  
Keyword(s):  
Risk Factors ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Ischemic Stroke ◽  
Treatment Modality ◽  
Intracranial Aneurysms ◽  
Univariate Analysis ◽  
Multivariate Logistic Regression Analysis ◽  
Flow Diverter ◽  
Multivariate Logistic Regression

Abstract Background The aim of this study was to comprehensively evaluate the risk factors of periprocedural ischemic stroke associated with endovascular treatment of intracranial aneurysms using a real-world database. Methods From August 2016 to March 2017, 167 patients were enrolled. Univariate analysis and multivariate logistic regression analysis were used to examine the risk factors for periprocedural ischemic stroke. Results Among the 167 cases, periprocedural ischemic stroke occurred in 20 cases (11.98%). After univariate analysis, the ischemic group had a higher proportion of large (≥ 10 mm) aneurysms than the control group (45.0% vs. 23.1%, p = 0.036). The incidence of periprocedural ischemic stroke was higher in cases treated by flow diverter (21.6%) or stent-assisted coiling (11.8%) than in cases treated by coiling only (2.7%), and the differences were statistically significant (p = 0.043). After multivariate logistic regression analysis, treatment modality was the independent risk factor for periprocedural ischemic stroke. Compared with the coiling-only procedure, flow diverter therapy was associated with a significantly higher rate of periprocedural ischemic stroke (OR 9.931; 95% CI 1.174–84.038; p = 0.035). Conclusions Aneurysm size and treatment modality were associated with periprocedural ischemic stroke. Larger aneurysms were associated with increased risk of periprocedural ischemic stroke. Flow diverter therapy was associated with significantly more periprocedural ischemic stroke than the coiling procedure alone.

Abstract TP176: Risk Factors and Clinical Significances of the Quantitative CTA Vascular Calcification in Ischemic Stroke Patients

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Takashi Shimoyama ◽  
Sibaji Gaj ◽  
Kunio Nakamura ◽  
Shivakrishna Kovi ◽  
Ken Uchino
Keyword(s):  
Risk Factors ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Ischemic Stroke ◽  
Vascular Calcification ◽  
Logistic Regression Analysis ◽  
Multivariate Logistic Regression Analysis ◽  
Multivariate Logistic Regression ◽  
Stroke Patients ◽  
Age Category

Background and Purpose: Intracranial arterial calcification is a marker of atherosclerosis burden in the general population. The aim of the study is to investigate risk factor profiles of vascular calcification in ischemic stroke patients. Methods: We identified ischemic stroke patients who underwent complete CTA from a prospective single-hospital stroke registry in 2018. Automatic artery and calcification segmentation method measured calcification volumes in the intracranial, extracranial, and aortic arteries using deep-learning U-net model and region-grow algorithms. Severe vascular calcification was defined as patients in the upper quartile calcification volume. The prevalence of severe vascular calcification and mean calcification volume were investigated by age category (<60 years, 60-70 years, 70-80 years, 80 years ≥). The relation between each potential risk factors and severe vascular calcification was assessed using the multivariate logistic regression analysis adjusted for age, sex, NIHSS score, and TOAST stroke subtypes. Results: Of the 558 consecutive acute ischemic stroke patients, 388 patients (212 males; mean age 66.6±14.2 years) met inclusion and with quantitative CTA calcification. The prevalence of severe vascular calcification (CTA calcification volume> 812 mm 3 ) increased with increasing age category (<60 years: 6.8% (7/103), 60-70 years: 15.7% (18/115), 70-80 years: 39.6% (38/105), 80 years ≥: 45.9% (34/74), P<0.001 for χ 2 test). Over age 80 years subsets had significantly higher mean calcification volume with 1213 mm 3 than other age category (<60 years: 225 mm 3 , P<0.001; 60-70 years: 462 mm 3 , P<0.001; 70-79 years: 817 mm 3 , P=0.020 for t-test). In the multivariate logistic regression analysis, age (OR 1.096, 95% CI 1.066-1.128, P<0.001), smoking (OR 3.430, 95% CI 1.833-6.419, P<0.001), and large artery atherosclerosis (LAA) (OR 4.260, 95% CI 1.963-9.247, P<0.001) were independently associated with severe vascular calcification. Conclusion: In the quantitative CTA analysis of calcification volume, older age and smoking were high risk for severe atherosclerotic calcium burden in ischemic stroke patients. Moreover, severe vascular calcification may differentiate LAA from other stroke etiology.

Abstract 147: The Association Between Hematocrit and Mortality among Ischemic Stroke Patients

Stroke ◽  
2012 ◽  
Vol 43 (suppl_1) ◽  
Author(s):  
Jason J Sico ◽  
Laura J Myers ◽  
Dede Ordin ◽  
Linda S Williams ◽  
Dawn M Bravata
Keyword(s):  
Logistic Regression ◽  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Stroke Mortality ◽  
Multivariate Logistic Regression ◽  
P Values ◽  
Time Points ◽  
Post Stroke ◽  
One Year ◽  
The Relationship

Introduction: Anemia is associated with higher mortality among patients with such non-stroke vascular conditions as heart failure and myocardial infarction. Less is known regarding the relationship between anemia and mortality among patients with acute ischemic stroke. Methods: Medical records were abstracted for a sample of 3965 veterans from 131 Veterans Health Administration (VHA) facilities who were admitted for a confirmed diagnosis of ischemic stroke (fiscal year 2007). Hematocrit (Hct) values from 24-hours of admission were categorized into 6-tiers (≤27%, 28-32%, 33-37%, 38-42%, 43-47%, ≥48%). We excluded patients with: female gender (n=95), incomplete Hct data (n=94), thrombolysis (n=32), and inconsistent death dates (n=6). We used multivariate logistic regression to examine the relationship between anemia and in-hospital, 30-day, 60-day and one-year mortality using multivariate logistic regression models for each time point, adjusting for age, NIHSS, comorbidity (including pneumonia), and Acute Physiology and Chronic Health Evaluation (APACHE)-III scores. The discrimination (c-statistics) and calibration (Hosmer-Lemeshow goodness of fit [HLGOF]) statistics were generated to gauge model performance and fit. Results: Approximately 2.1% of the N=3750 patients presented with Hcts ≤27%, 6.2% were 28-32%, 17.9% were 33-37%, 36.4% were 38-42%, 28.2% were 43-47%, and 9.1% were ≥48%. Adjusted mortality odds at all time points were 2.5 to 3.5 times higher for those with ≤Hct 27% (p values < 0.013 for in-hospital and 30-day mortality; p values at 6 months and one year were 0.002 and 0.001, respectively). Mortality risk at 6 months and 1 year showed a significant and dose-response relationship to Hct for all Hct groups <38%. High Hcts were independently associated only with in-hospital mortality and only in those with Hct ≥48 (OR 2.9, p=0.004). Models performed well across time points (C=0.813, HLGOF=0.9684 [in-hospital]; C=0.832, HLGOF=0.8186 [30-day]; C=0.863, HLGOF=0.7307 [60-day]; C=0.880, HLGOF=0.4313 [one-year]). Conclusions: Even a moderate level of anemia is independently associated with an increased risk of death during the first year following acute ischemic stroke. Very low or very high Hct is associated with early post-stroke mortality. Further work is required to evaluate whether interventions that treat anemia, its complications and underlying etiologies may also reduce post-stroke mortality.

scholarly journals A comparison of logistic regression models with alternative machine learning methods to predict the risk of in-hospital mortality in emergency medical admissions via external validation

2018 ◽  
Vol 26 (1) ◽  
pp. 34-44 ◽  
Author(s):  
Muhammad Faisal ◽  
Andy Scally ◽  
Robin Howes ◽  
Kevin Beatson ◽  
Donald Richardson ◽  
...  
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Hospital Mortality ◽  
Receiver Operating Characteristic Curve ◽  
Operating Characteristic ◽  
Characteristic Curve ◽  
External Validation ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Operating Characteristic Curve

We compare the performance of logistic regression with several alternative machine learning methods to estimate the risk of death for patients following an emergency admission to hospital based on the patients’ first blood test results and physiological measurements using an external validation approach. We trained and tested each model using data from one hospital ( n = 24,696) and compared the performance of these models in data from another hospital ( n = 13,477). We used two performance measures – the calibration slope and area under the receiver operating characteristic curve. The logistic model performed reasonably well – calibration slope: 0.90, area under the receiver operating characteristic curve: 0.847 compared to the other machine learning methods. Given the complexity of choosing tuning parameters of these methods, the performance of logistic regression with transformations for in-hospital mortality prediction was competitive with the best performing alternative machine learning methods with no evidence of overfitting.

Development of a prediction model for 1-year poor prognosis in patients with acute ischemic stroke

2019 ◽  
Vol 67 (6) ◽  
pp. 957-963 ◽  
Author(s):  
Xia Ling ◽  
Bo Shen ◽  
Kangzhi Li ◽  
Lihong Si ◽  
Xu Yang
Keyword(s):  
Risk Factors ◽  
Logistic Regression ◽  
Ischemic Stroke ◽  
Prediction Model ◽  
Acute Ischemic Stroke ◽  
Poor Prognosis ◽  
Multivariate Logistic Regression ◽  
Lemeshow Test ◽  
Independent Risk Factors ◽  
Auc Value

The goals of this study were to develop a new prediction model to predict 1-year poor prognosis (death or modified Rankin scale score of ≥3) in patients with acute ischemic stroke (AIS) and to compare the performance of the new prediction model with other prediction scales. Baseline data of 772 patients with AIS were collected, and univariate and multivariate logistic regression analyses were performed to identify independent risk factors for 1-year poor prognosis in patients with AIS. The area under the receiver operating characteristics curve (AUC) value of the new prediction model and the THRIVE, iScore and ASTRAL scores was compared. The Hosmer-Lemeshow test was used to assess the goodness of fit of the model. We identified 196 (25.4%) patients with poor prognosis at 1-year follow-up, and of these 68 (68/196, 34.7%) had died. Multivariate logistic regression and receiver operating characteristic curve analyses showed that age ≥70 years, consciousness (lethargy or coma), history of stroke or transient ischemic attack, cancer, abnormal fasting blood glucose levels ≥7.0 mmol/L, and National Institutes of Health Stroke Scale score were independent risk factors for 1-year poor prognosis in patients with AIS. Scores were assigned for each variable by rounding off β coefficient to the integer score, and a new prediction model with a maximum total score of 9 points was developed. The AUC value of the new prediction model was higher than the THRIVE score (p<0.05). The χ2 value for the Hosmer-Lemeshow test was 7.337 (p>0.05), suggesting that the prediction model had a good fit. The new prediction model can accurately predict 1-year poor prognosis in Chinese patients with AIS.

Sign in / Sign up

Export Citation Format

Share Document