Changing stroke rehab and research worldwide now.Time is Brain! trillions and trillions of neurons that DIE each day because there are NO effective hyperacute therapies besides tPA(only 12% effective). I have 523 posts on hyperacute therapy, enough for researchers to spend decades proving them out. These are my personal ideas and blog on stroke rehabilitation and stroke research. Do not attempt any of these without checking with your medical provider. Unless you join me in agitating, when you need these therapies they won't be there.

What this blog is for:

My blog is not to help survivors recover, it is to have the 10 million yearly stroke survivors light fires underneath their doctors, stroke hospitals and stroke researchers to get stroke solved. 100% recovery. The stroke medical world is completely failing at that goal, they don't even have it as a goal. Shortly after getting out of the hospital and getting NO information on the process or protocols of stroke rehabilitation and recovery I started searching on the internet and found that no other survivor received useful information. This is an attempt to cover all stroke rehabilitation information that should be readily available to survivors so they can talk with informed knowledge to their medical staff. It lays out what needs to be done to get stroke survivors closer to 100% recovery. It's quite disgusting that this information is not available from every stroke association and doctors group.

Thursday, June 12, 2025

Mining the risk factors for stroke occurrence and dietary protective factors based on the NHANES database: Analysis using SHAP

 Useless piece of shit! NO DIET PROTOCOLS CREATED! What the fuck do you think stroke research is for? Just getting published; obviously don't care about stroke survivors or definitively preventing stroke?

, ,
https://doi.org/10.1016/j.jad.2025.119671
Get rights and content

Highlights

  • A predictive model was created through a large - scale retrospective study.
  • Based on the model, a scoring scale was constructed to assess the risk of stroke.
  • Both the model and the scale show high clinical accuracy.

Abstract

Background

Stroke has a high morbidity and mortality rate and its incidence is rising. This study aimed to identify risk factors for stroke, develop an accurate prediction model and scoring system, thereby providing novel primary prevention strategies for stroke and post-stroke depression.

Methods

Data from 49,491 participants in the National Health and Nutrition Examination Survey (NHANES) (1999–2016) were analyzed. A training cohort of 3972 individuals was created. Logistic regression was applied for univariate and multivariate analyses to identify risk factors. SHAP values analyzed the impact of these factors, leading to the development of a prediction model and scoring scale. Model performance was validated using DCA, ROC, and calibration curves.

Results

A prediction model with an AUC of 82 % was developed, alongside a scoring scale with an AUC of 79 %. Key risk factors were identified as “age(OR = 1.061, 95%CI:1.055–1.066, P<0.001)”, “income level(OR= 0.812, 95%CI:0.777–0.85, P<0.001)”, “cholesterol(OR = 0.845, 95%CI:0.793–0.901, P<0.001)”, “albumin(OR = 0.614, 95%CI:0.497–0.757, P<0.001)”, “lymphocyte percentage(OR= 1.045, 95%CI:1–1.091, P=0.048)”, “alkaline phosphatase(OR= 1.005, 95%CI:1.002–1.002, P<0.001)”, “ Glucose(OR= 1.059, 95%CI:1.026–1.093, P<0.001)”, “triglycerides(OR = 1.061, 95%CI:1.004–1.123, P = 0.037)”, and “neutrophil percentage(OR= 1.058, 95%CI:1.015–1.102, P=0.008)”.

Conclusion

In this study, a simple scoring scale was successfully developed based on the research findings. The scale demonstrates excellent clinical usability and operational convenience, providing a novel self-screening tool for the occurrence of stroke and post-stroke depression.

Introduction

Stroke, characterized by its extremely high mortality and disability rates, has already become one of the major disease types that pose a great threat to human life and health worldwide (Qin et al., 2022; Campbell et al., 2019; Hilkens et al., 2024). Based on the results of epidemiological surveys, taking the United States as a typical example for analysis, the prevalence of stroke has shown a significant upward trend, and it is expected to climb from 3 % in 2012 to 4 % in 2030. Along with this, the annual costs in the medical and nursing fields will approach as high as nearly 200 billion US dollars (Ovbiagele et al., 2013). Post-stroke depression (PSD), one of the most common neuropsychiatric complications of stroke, has an incidence rate of 30–50 % and significantly increases mortality and disability rates in patients (Lanctôt et al., 2020; Frank et al., 2022; Chen et al., 2024). In view of this severe situation, if high-risk stroke patient groups can be accurately identified in the early stage of the disease and primary prevention strategies can be implemented in advance, it is expected to significantly reduce the heavy medical burden borne by society and the families of patients. Driven by this background, prediction models and scoring scales have been constructed and gradually developed and improved as effective countermeasures, aiming to provide scientific basis and decision support for the early warning and precise prevention and control of stroke and PSD.
In the field of neurosurgery, numerous scoring scales have been widely recognized. Representative ones among them include the Glasgow Coma Scale (GCS) (Teasdale and Jennett, 1974), the Mini-Mental State Examination (MMSE) (Folstein et al., 1975), and the National Institutes of Health Stroke Scale (NIHSS) (Kwah and Diong, 2014), etc. These scales, with their remarkable advantages such as simple structure and convenient operation, have been widely popularized and applied in clinical practice and have become indispensable important tools in the diagnosis and treatment process of neurosurgery. However, in terms of predicting the occurrence of stroke, there is a relative lack of effective prediction models available at present. Among them, although the Framingham Stroke Risk Assessment Scale is somewhat well-known, it has many limitations itself. For example, it has deficiencies in aspects such as the scope of risk factors covered, the performance of prediction accuracy in specific populations, and the adaptability to complex clinical situations, which has led to its failure to be widely used in predicting the occurrence of stroke in clinical practice.
In addition, although some research teams have been committed to developing a series of prediction models (Sonoda et al., 2019; Arafa et al., 2022; Liu et al., 2020; Yu et al., 2021; Li et al., 2023), from the overall effect evaluation, these models have not reached an ideal state in key indicators such as prediction accuracy, practicability, and universality, and it is difficult for them to meet the actual clinical needs. Based on this, both clinicians when formulating precise diagnosis and treatment plans and patients during the process of self-health management and risk prediction are urgently in need of a stroke prediction model that combines the characteristics of simplicity and high accuracy to fill this crucial gap in current clinical practice and provide more powerful support and guarantee for the early warning and effective prevention and control of stroke.
SHAP (SHapley Additive exPlanations), as an effective method specifically used to interpret the output of machine learning models (Ali et al., 2023), has its theoretical foundation deeply rooted in the concept of Shapley values in game theory. By precisely calculating the contribution degree of each feature in the model to the prediction results, SHAP endows machine learning models with a high degree of interpretability. Through this way, it is possible to gain in-depth insights into the specific roles and contribution magnitudes played by each feature in the process of generating the model's prediction results. The core advantages of this method are remarkable. It can accurately and stably measure and evaluate the importance of features. The obtained measurement values have a wide range of application values. They can not only be applied to the in-depth analysis of individual prediction results to clarify the mechanism of each feature behind specific predictions but also help with the systematic analysis of the overall behavioral characteristics of the model at the macro level (Giacobbe et al., 2023), thus providing a solid and reliable basis and guidance for the optimization, performance improvement, and result interpretation of the model.
The core objective of this study is to explore the potential mechanisms of stroke occurrence and construct an effective prediction system. It aims to systematically screen the risk factors closely related to stroke occurrence using the logistics regression analysis method, and then build a prediction model and a prediction scoring table with high reliability and accuracy. To enhance the interpretability and transparency of the model, the SHAP method is introduced. By calculating the contribution degree of each feature to the model's prediction results, the complex relationships and action mechanisms within the model are deeply analyzed, so that the output results of the model are more interpretable and credible. Another core objective of our stroke risk prediction model is to identify high-risk populations — individuals not only threatened by stroke but also characterized by underlying pathological mechanisms such as neuroinflammation and oxidative stress, which have been demonstrated to be closely associated with the development of depression and anxiety. It is expected that the results of this study can provide a more solid and valuable scientific reference basis for the formulation of early prevention strategies for stroke and PSD, fill some gaps in the current research in this field, and promote the further development and improvement of preventive medicine for stroke and PSD.
This study distinguishes itself from previous research through two key innovations. First, it constructs a predictive scoring system exclusively using readily available demographic and routine blood markers, ensuring ease of implementation in primary care settings and enabling layperson self-assessment without specialized medical training. Second, the application of SHAP provides transparent, quantifiable rankings of risk factors, offering mechanistic insights unattainable in traditional black-box machine learning models.

More at link.


No comments:

Post a Comment