AI-Powered Stroke Diagnosis System: Methodological Framework and Implementation

Narigina, Marta; Vindecs, Agris; Bošković, Dušanka; Merkuryev, Yuri; Romanovs, Andrejs

doi:10.3390/fi17050204

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

AI-Powered Stroke Diagnosis System: Methodological Framework and Implementation

by

Marta Narigina

¹

,

Agris Vindecs

¹

,

Dušanka Bošković

²

,

Yuri Merkuryev

¹ and

Andrejs Romanovs

^1,*

¹

Institute of Information Technology, Riga Technical University, 6A Kipsalas Street, LV-1048 Riga, Latvia

²

Faculty of Electrical Engineering, University of Sarajevo, Zmaja od Bosne bb, 71000 Sarajevo, Bosnia and Herzegovina

^*

Author to whom correspondence should be addressed.

Future Internet 2025, 17(5), 204; https://doi.org/10.3390/fi17050204 (registering DOI)

Submission received: 9 March 2025 / Revised: 20 April 2025 / Accepted: 29 April 2025 / Published: 2 May 2025

(This article belongs to the Special Issue Artificial Intelligence for Smart Healthcare: Methods, Applications, and Challenges)

Download Versions Notes

Abstract

This study introduces an AI-based framework for stroke diagnosis that merges clinical data and curated imaging data. The system utilizes traditional machine learning and advanced deep learning techniques to tackle dataset imbalances and variability in stroke presentations. Our approach involves rigorous data preprocessing, feature engineering, and ensemble techniques to optimize the predictive performance. Comprehensive evaluations demonstrate that gradient-boosted models outperform in accuracy, while CNNs enhance stroke detection rates. Calibration and threshold optimization are utilized to align predictions with clinical requirements, ensuring diagnostic reliability. This multi-modal framework highlights the capacity of AI to accelerate stroke diagnosis and aid clinical decision making, ultimately enhancing patient outcomes in critical care.

Keywords: stroke diagnosis; eICU database; machine learning; deep learning; convolutional neural networks; ensemble methods; calibration; threshold optimization

Share and Cite

MDPI and ACS Style

Narigina, M.; Vindecs, A.; Bošković, D.; Merkuryev, Y.; Romanovs, A. AI-Powered Stroke Diagnosis System: Methodological Framework and Implementation. Future Internet 2025, 17, 204. https://doi.org/10.3390/fi17050204

AMA Style

Narigina M, Vindecs A, Bošković D, Merkuryev Y, Romanovs A. AI-Powered Stroke Diagnosis System: Methodological Framework and Implementation. Future Internet. 2025; 17(5):204. https://doi.org/10.3390/fi17050204

Chicago/Turabian Style

Narigina, Marta, Agris Vindecs, Dušanka Bošković, Yuri Merkuryev, and Andrejs Romanovs. 2025. "AI-Powered Stroke Diagnosis System: Methodological Framework and Implementation" Future Internet 17, no. 5: 204. https://doi.org/10.3390/fi17050204

APA Style

Narigina, M., Vindecs, A., Bošković, D., Merkuryev, Y., & Romanovs, A. (2025). AI-Powered Stroke Diagnosis System: Methodological Framework and Implementation. Future Internet, 17(5), 204. https://doi.org/10.3390/fi17050204

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

AI-Powered Stroke Diagnosis System: Methodological Framework and Implementation

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI