Cross-cultural validation of the stroke riskometer using generalizability theory

Medvedev, Oleg N.; Truong, Quoc Cuong; Merkin, Alexander; Borotkanics, Robert; Krishnamurthi, Rita; Feigin, Valery

doi:10.1038/s41598-021-98591-8

Cited by 13 publications

(17 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although all apps were properly validated, the majority of the apps (18/20 provided mainly stroke/CVD prediction estimates, and only 2 apps (HeartScore and Stroke Riskometer) and one Web app (PreventS-MD) met most of the requirements for an ideal mHealth primary stroke prevention tool. Of those 3, only the Stroke Riskometer 80,[86][87][88][89] was tested in an RCT that demonstrated the feasibility, acceptability, and preliminary efficacy of the app, 90 and only 2 apps (Stroke Riskometer and PreventS-MD) are specifically designed for both primary and secondary stroke prevention and applied "motivational primary prevention strategy" 91 regardless of the level of stroke/CVD risk, 6,39 thus bridging the gap between high absolute CVD risk and population-wide primary prevention strategies. 50,56…”

Section: Currently Available Digital Technologies For Primary Stroke ...mentioning

confidence: 99%

Digital Health in Primordial and Primary Stroke Prevention: A Systematic Review

Feigin

Owolabi

Hankey

et al. 2022

Stroke

View full text Add to dashboard Cite

The stroke burden continues to grow across the globe, disproportionally affecting developing countries. This burden cannot be effectively halted and reversed without effective and widely implemented primordial and primary stroke prevention measures, including those on the individual level. The unprecedented growth of smartphone and other digital technologies with digital solutions are now being used in almost every area of health, offering a unique opportunity to improve primordial and primary stroke prevention on the individual level. However, there are several issues that need to be considered to advance development and use this important digital strategy for primordial and primary stroke prevention. Using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines we provide a systematic review of the current knowledge, challenges, and opportunities of digital health in primordial and primary stroke prevention.

show abstract

Section: Currently Available Digital Technologies For Primary Stroke ...mentioning

confidence: 99%

Digital Health in Primordial and Primary Stroke Prevention: A Systematic Review

Feigin

Owolabi

Hankey

et al. 2022

Stroke

View full text Add to dashboard Cite

show abstract

“…Participants were asked for informed consent and both the Stroke Riskometer questionnaire along with Framingham Stroke Profile was filled, risk percentage from both the tools were recorded using the App for each individual. The Stroke Riskometer demonstrated strong reliability in measuring stroke risks across the countries with coefficients G relative and absolute of 0.84, 95%CI (0.79; 0.89) and 0.82, 95%CI (0.76; 0.88) respectively [9]. Classifying the range into < 5% as low, 6-10% as moderate and > 10% as high risk in Stroke Riskometer tool (Fig.…”

Section: Methodsmentioning

confidence: 99%

Classification of Stroke Riskometer AppTM into low, moderate and high stroke risk range

Neelam¹,

P.²,

Swathi³

et al. 2022

Fiz Pol

View full text Add to dashboard Cite

Background. Stroke is one of the leading of the causes of disability and according to the global burden of disease (GBD) study in 2010 it is the second leading cause of death worldwide. Developing countries reported 85% global burden of stroke enduring 80 percent of death due to hemorrhagic strokes. Stroke-related hospitalization rate in India is 46 per 100,000 persons. However Stroke Riskometer App can be used to significantly improve stroke and Non- Communicable disease prevention, since it operates on latest expansions in risk allocution/communication, International guidelines on stroke and Cardio-Vascular Disease prevention. However, the app does not classify the risk individuals into low, moderate and high categories which will not create any impact over the high risk individual, therefore it calls for the need to classify the tool into low, moderate and high risk ranges. Aim. To classify the Stroke Riskometer App into low, moderate and high stroke risk ranges among adult and elderly population. Methods. Participants were 250 samples selected from Chettinad Academy of Research and Education and S.A. Poly Clinic, Chennai, for over a_period of 6 months (April 2021-September 2021) and the Stroke Riskometer questionnaire along with Framingham Stroke Profile was filled, risk percentage from both the tools were recorded using the App for each individual. Results. Frequency analysis and Chi-square tests were performed, classifying the range into < 5% as low, 6-10% as moderate and > 10% as high risk in Stroke Riskometer tool. In Chi-square test it showed n = 104 as low risk, n = 11 as moderate risk,n = 14 as high risk in both FRS and Stroke Riskometer respectively. The chi-square value is 24.224 and the significant p value is < 0.0001showing that the value is performing well with FRS score. Conclusion. Though the App was classified into ranges it need to be continually developed and validated with larger sample size, heterogenous population and robust ethnic groups.

show abstract

“…It estimates personal stroke risk level and additional risk factors, including family medical history, cognitive impairment, memory loss and dementia. The Stroke Riskometer app was cross-culturally validated using Generalizability Theory and its risk assessment scores were found generalizable across countries and languages [54 ▪ ].…”

Section: Artificial Intelligence On the Marketmentioning

confidence: 99%

Machine learning, artificial intelligence and the prediction of dementia

Merkin

Krishnamurthi

Medvedev

2021

Current Opinion in Psychiatry

Self Cite

View full text Add to dashboard Cite

Purpose of reviewArtificial intelligence and its division machine learning are emerging technologies that are increasingly applied in medicine. Artificial intelligence facilitates automatization of analytical modelling and contributes to prediction, diagnostics and treatment of diseases. This article presents an overview of the application of artificial intelligence in dementia research. Recent findingsMachine learning and its branch Deep Learning are widely used in research to support in diagnosis and prediction of dementia. Deep Learning models in certain tasks often result in better accuracy of detection and prediction of dementia than traditional machine learning methods, but they are more costly in terms of run times and hardware requirements. Both machine learning and Deep Learning models have their own strengths and limitations. Currently, there are few datasets with limited data available to train machine learning models. There are very few commercial applications of machine learning in medical practice to date, mostly represented by mobile applications, which include questionnaires and psychometric assessments with limited machine learning data processing.

show abstract

Cross-cultural validation of the stroke riskometer using generalizability theory

Cited by 13 publications

References 26 publications

Digital Health in Primordial and Primary Stroke Prevention: A Systematic Review

Digital Health in Primordial and Primary Stroke Prevention: A Systematic Review

Classification of Stroke Riskometer AppTM into low, moderate and high stroke risk range

Machine learning, artificial intelligence and the prediction of dementia

Contact Info

Product

Resources

About