4 papers3 findings
NIHSS
upin humans1OR 1.16
NIHSS raises the risk of post-stroke dementia in humans (OR 1.16, 95% CI 1.01–1.13).
“higher national institutes of health stroke scale scores were associated with a higher risk of dementia (OR: 1.16, 95% CI: 1.01–1.13, p = 0.02).”
Humansin post-stroke patientsn = 81
Brain Resilience and Its Association with Post‐Stroke Dementia: A Neuroimaging‐Based Study
AAIC 2025 abstract24 Dec 2025
Brain age gap
upin humans1OR 1.07
Brain age gap raises the risk of post-stroke dementia in humans (OR 1.07, 95% CI 1.01–1.13).
“Brain resilience was significantly associated with post‐stroke dementia (OR: 1.07, 95% CI: 1.01–1.13, p = 0.02).”
Humansin post-stroke patientsn = 81
Brain Resilience and Its Association with Post‐Stroke Dementia: A Neuroimaging‐Based Study
AAIC 2025 abstract24 Dec 2025
Extreme gradient-boosting tree model
upin humans1
extreme gradient-boosting tree model predicts post-stroke dementia in humans (accuracy 92.78%, AUC 0.836).
“The XGBoost model achieved the highest performance metrics (accuracy: 92.78%, AUC: 0.836), closely followed by LightGBM (92.78%, 0.834).”
Humanspost-strokeone yearn = 55,888
Comparing Machine Learning and Traditional Models to Predict One‐Year Post‐Stroke Dementia Risk
AAIC 2025 abstract23 Dec 2025
In 7,047 UK stroke survivors, lower muscle quality and higher blood metabolite risk scores were linked to higher risks of all-cause, Alzheimer's and vascular dementia.In 7,047 UK stroke survivors, lower muscle quality and higher blood metabolite risk scores were linked to higher risks of all-cause, Alzheimer's and vascular dementia.