scholarly journals 230 Machine Learning Models for Predicting Ischemic Stroke and Major Bleeding Risk in Patients with Atrial Fibrillation

2020 ◽  
Vol 29 ◽  
pp. S137-S138
Author(s):  
J. Lu ◽  
G. Dwivedi ◽  
F. Sanfilippo ◽  
M. Bennamoun ◽  
J. Hung ◽  
...  
Keyword(s):  
Machine Learning ◽  
Atrial Fibrillation ◽  
Ischemic Stroke ◽  
Major Bleeding ◽  
Bleeding Risk ◽  
Learning Models ◽  
Machine Learning Models

Abstract 140: Comparison of Major Complication of Oral Anticoagulants in Non-Valvular Atrial Fibrillation: Systematic Review and Meta-Analyses

2017 ◽  
Vol 37 (suppl_1) ◽  
Author(s):  
Hong Seok Lee
Keyword(s):  
Atrial Fibrillation ◽  
Ischemic Stroke ◽  
Major Bleeding ◽  
Oral Anticoagulant ◽  
Stroke Risk ◽  
Oral Anticoagulants ◽  
Bleeding Risk ◽  
Gi Bleeding ◽  
Ischemic Stroke Risk ◽  
Meta Analyses

Background: Oral anticoagulants known as a novel oral anticoagulant have been used for the management of non -valvular atrial fibrillation. There was no enough study regarding the efficacy and safety of three major new oral anticoagulants. We assessed major three oral anticoagulants in terms of major bleeding complication and stroke prevention by meta-analyses studies comparing those drugs. Method: Relevant studies were identified through electronic literature searches of MEDLINE, EMBASE, Cochrane library, and clinicaltrials.gov (from inception to February 24, 2016). RevMan and ITC software were used for direct comparisons, respectively. Results: Apixaban (N=6020), versus dabigatran(N=12038), apixaban versus rivaroxaban(N=8503) and rivaroxaban versus dabigatran were analyzed directly. There was significantly higher major bleeding risks in apixaban compared to dabigatran (both 110mg and 150mg) after adjusting baseline bleeding risk (Relative risk 3.41, 95% confidence interval(2.61 to 4.47) in 110mg, (5.62, 4.83 to 6.54) in 150mg. Intracranial bleeding risk in apixaban was significantly higher than in dabigatran (10.5, 6.10 to18.01). However, apixaban had less GI bleeding risk compared to dabigatran (0.80 , 0.65 to 0.98) and also had less ischemic stroke risk (0.31,0.22 to 0.42). Rivaroxaban showed higher major bleeding risk than dabigatran 110mg (2.34 , 1.81 to 3.03), however, Rivaroxaban had less bleeding risk compared to dabigatran 150mg (0.41, 0.35 to 0.46). Dabigatran 110mg and 150mg had less GI bleeding risk compared to rivaroxaban (0.31 , 0.24 to 0.39) and (0.23,0.17 to 0.29) respectively. Ischemic stroke risk was also decreased in dabigatran110mg (0.46, 0.38 to 0.57). and 150mg (0.66 ,0.52 to 0.83). Conclusion: Observed oral anticoagulants were associated with various complications. Overall, apixaban had higher intracranial bleeding risk than dabigatran. The highest GI bleeding risk in rivaroxaban compared to apixaban and dabigatran. Ischemic stroke risk was the highest in dabigatran. In conclusion, we may use those oral anticoagulant based on risks rates, however, a larger study with longer follow-up is needed to corroborate findings.

Abstract WP312: Identifying Acute Ischemic Stroke by Analyzing Icd-10 Claims Data Using Machine Learning Models

Stroke ◽  
2017 ◽  
Vol 48 (suppl_1) ◽  
Author(s):  
Charles Esenwa ◽  
Jorge Luna ◽  
Benjamin Kummer ◽  
Hojjat Salmasian ◽  
David Vawdrey ◽  
...  
Keyword(s):  
Machine Learning ◽  
Logistic Regression ◽  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Claims Data ◽  
Administrative Claims ◽  
Learning Models ◽  
Icd 10 ◽  
New Diagnosis ◽  
Machine Learning Models

Introduction: Retrospective identification of patients hospitalized with new diagnosis of acute ischemic stroke is important for administrative quality assurance, post-discharge clinical management, and stroke research. The benefit of using administrative claims data is its widespread availability, but the disadvantage is in the inability to accurately and consistently identify the clinical diagnosis of interest. Hypothesis: We hypothesized that decision tree and logistic regression models could be applied to administrative claims data coded using International Classification of Diseases, version 10 (ICD-10) to create algorithms that could accurately identify patients with acute ischemic stroke. Methods: We used hospital records from our institution to develop a gold standard list of 243 patients, continuously hospitalized with a new diagnosis of stroke from 10/1/2015 to 3/31/2016. We used 1,393 neurological patients without a diagnosis of stroke as negative controls. This list was used to train and test two machine learning methods of diagnosis and procedure codes analysis, for the purpose of ischemic stroke identification: one using classification and regression tree (CART) and another using regularized logistic regression. We trained the models using 75% of the data and performed the evaluation using the remaining 25%. Results: The CART model had a κ=0.78, sensitivity of 96%, specificity of 90%, and a positive predictive value of 99%. The regularized logistic regression model had a κ=0.73, sensitivity of 97%, specificity of 81%, and a positive predictive value of 98%. Conclusion: Both the decision tree and logistic regression machine based learning models showed very high accuracy in identifying patients with a new diagnosis of ischemic stroke, using ICD-10 code claims data, when compared to our gold standard. Applying these machine learning models to identify patients with ischemic stroke has widespread applications, especially in this period where national billing data has transitioned from ICD-9 to ICD-10 codes.

Immune-inflammatory, coagulation, adhesion, and imaging biomarkers combined in machine learning models improve the prediction of death 1 year after ischemic stroke

2021 ◽  
Author(s):  
Ana Lucia Cruz Fürstenberger Lehmann ◽  
Daniela Frizon Alfieri ◽  
Maria Caroline Martins de Araújo ◽  
Emanuelle Roberto Trevisani ◽  
Maisa Rocha Nagao ◽  
...  
Keyword(s):  
Machine Learning ◽  
Ischemic Stroke ◽  
Imaging Biomarkers ◽  
Learning Models ◽  
Machine Learning Models

scholarly journals Localized prediction of tissue outcome in acute ischemic stroke patients using diffusion- and perfusion-weighted MRI datasets

PLoS ONE ◽  
2020 ◽  
Vol 15 (11) ◽  
pp. e0241917
Author(s):  
Malte Grosser ◽  
Susanne Gellißen ◽  
Patrick Borchert ◽  
Jan Sedlacik ◽  
Jawed Nawabi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Outcome Prediction ◽  
Hybrid Approach ◽  
Learning Models ◽  
Stroke Patients ◽  
Local Approach ◽  
Prediction Approach ◽  
Machine Learning Models

Background An accurate prediction of tissue outcome in acute ischemic stroke patients is of high interest for treatment decision making. To date, various machine learning models have been proposed that combine multi-parametric imaging data for this purpose. However, most of these machine learning models were trained using voxel information extracted from the whole brain, without taking differences in susceptibility to ischemia into account that exist between brain regions. The aim of this study was to develop and evaluate a local tissue outcome prediction approach, which makes predictions using locally trained machine learning models and thus accounts for regional differences. Material and methods Multi-parametric MRI data from 99 acute ischemic stroke patients were used for the development and evaluation of the local tissue outcome prediction approach. Diffusion (ADC) and perfusion parameter maps (CBF, CBV, MTT, Tmax) and corresponding follow-up lesion masks for each patient were registered to the MNI brain atlas. Logistic regression (LR) and random forest (RF) models were trained employing a local approach, which makes predictions using models individually trained for each specific voxel position using the corresponding local data. A global approach, which uses a single model trained using all voxels of the brain, was used for comparison. Tissue outcome predictions resulting from the global and local RF and LR models, as well as a combined (hybrid) approach were quantitatively evaluated and compared using the area under the receiver operating characteristic curve (ROC AUC), the Dice coefficient, and the sensitivity and specificity metrics. Results Statistical analysis revealed the highest ROC AUC and Dice values for the hybrid approach. With 0.872 (ROC AUC; LR) and 0.353 (Dice; RF), these values were significantly higher (p < 0.01) than the values of the two other approaches. In addition, the local approach achieved the highest sensitivity of 0.448 (LR). Overall, the hybrid approach was only outperformed in sensitivity (LR) by the local approach and in specificity by both other approaches. However, in these cases the effect sizes were comparatively small. Conclusion The results of this study suggest that using locally trained machine learning models can lead to better lesion outcome prediction results compared to a single global machine learning model trained using all voxel information independent of the location in the brain.

scholarly journals Prior ischemic strokes are non-inferior for predicting future ischemic strokes than CHA2DS2-VASc score in hemodialysis patients with non-valvular atrial fibrillation

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Anat Bel-Ange ◽  
Shani Zilberman Itskovich ◽  
Liana Avivi ◽  
Kobi Stav ◽  
Shai Efrati ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Ischemic Stroke ◽  
Major Bleeding ◽  
Large Scale ◽  
Stroke Risk ◽  
Bleeding Risk ◽  
Maintenance Hemodialysis ◽  
Bleeding Events ◽  
Study Population ◽  
Ischemic Stroke Risk

Abstract Background We tested whether CHA2DS2-VASc and/or HAS-BLED scores better predict ischemic stroke and major bleeding, respectively, than their individual components in maintenance hemodialysis (MHD) patients with atrial fibrillation (AF). Methods A retrospective cohort study of a clinical database containing the medical records of 268 MHD patients with non-valvular AF (167 women, mean age 73.4 ± 10.2 years). During the median follow-up of 21.0 (interquartile range, 5.0–44.0) months, 46 (17.2%) ischemic strokes and 24 (9.0%) major bleeding events were reported. Results Although CHA2DS2-VASc predicted ischemic stroke risk in the study population (adjusted HR 1.74 with 95% CI 1.23–2.46 for each unit of increase in CHA2DS2-VASc score, and HR of 5.57 with 95% CI 1.88–16.49 for CHA2DS2-VASc score ≥ 6), prior ischemic strokes/transient ischemic attacks (TIAs) were non-inferior in both univariate and multivariate analyses (adjusted HR 8.65 with 95% CI 2.82–26.49). The ROC AUC was larger for the prior ischemic stroke/TIA than for CHA2DS2-VASc. Furthermore, the CHA2DS2-VASc score did not predict future ischemic stroke risks in study participants who did not previously experience ischemic strokes/TIAs (adjusted HR 1.41, 95% CI: 0.84–2.36). The HAS-BLED score and its components did not have predictive abilities in discriminating bleeding risk in the study population. Conclusions Previous ischemic strokes are non-inferior for predicting of future ischemic strokes than the complete CHA2DS2-VASc score in MHD patients. CHA2DS2VASc scores are less predictive in MHD patients without histories of CVA/TIA. HAS-BLED scores do not predict major bleeding in MHD patients. These findings should redesign approaches to ischemic stroke risk stratification in MHD patients if future large-scale epidemiological studies confirm them.

Improving XGBoost with Imagination Sampling

2020 ◽  
Vol 2 (1) ◽  
pp. 3-6
Author(s):  
Eric Holloway
Keyword(s):  
Machine Learning ◽  
General System ◽  
Learning Models ◽  
Starting Point ◽  
Machine Learning Models

Imagination Sampling is the usage of a person as an oracle for generating or improving machine learning models. Previous work demonstrated a general system for using Imagination Sampling for obtaining multibox models. Here, the possibility of importing such models as the starting point for further automatic enhancement is explored.

Sign in / Sign up

Export Citation Format

Share Document