A retrospective study on machine learning-assisted stroke recognition for medical helpline calls

Sep 17, 2026

Upload by Oliver

Stroke outcomes depend on rapid treatment, and prehospital telehealth relies on accurate call-taker recognition to enable timely ambulance transport and hospital admission. This retrospective study uses Copenhagen Emergency Medical Services and the Danish Stroke Registry call recordings from 2015–2020 to build and evaluate machine learning models that classify helpline calls as stroke versus nonstroke using automatic speech recognition and an ensemble text classifier, then compares performance with call-takers using MH-1813 data from 2021.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Introduction

Results

Population characteristics

Aalborg Universitet A retrospective study on machine learning-assisted stroke recognition for medical helpline calls Wenstrup, Jonathan; Havtorn, Jakob Drachmann; Borgholt, Lasse; Blomberg, Stig Nikolaj; Maaloe, Lars; Sayre, Michael R. ; Christensen, Hanne; Kruuse, Christina; Christensen, Helle Collatz Published in: npj Digital Medicine DOI (link to publication from Publisher): 10.1038/s41746-023-00980-y Creative Commons License CC BY 4.0 Publication date: 2023 Document Version Publisher's PDF, also known as Version of record Link to publication from Aalborg University Citation for published version (APA): Wenstrup, J. , Havtorn, J. D. , Borgholt, L. , Blomberg, S. N. , Maaloe, L. , Sayre, M. R. , Christensen, H. , Kruuse, C. , & Christensen, H. C. (2023) . A retrospective study on machine learning-assisted stroke recognition for medical helpline calls. npj Digital Medicine, 6(1), Article 235. [https://doi.org/10.1038/s41746-023-00980-y](https://doi.org/10.1038/s41746-023-00980-y) General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. -Users may download and print one copy of any publication from the public portal for the purpose of private study or research. -You may not further distribute the material or use it for any profit-making activity or commercial gain -You may freely distribute the URL identifying the publication in the public portal- ARTICLE OPEN [www.nature.com/npjdigitalmed](www.nature.com/npjdigitalmed) A retrospective study on machine learning-assisted stroke recognition for medical helpline calls Jonathan Wenstrup 1,2,11, Jakob Drachmann Havtorn 3,4,11, Lasse Borgholt 3,5,6,11, Stig Nikolaj Blomberg7, Lars Maaloe 3,4, Michael R. Sayre 8, Hanne Christensen 9,10, Christina Kruuse1,10,12 and Helle Collatz Christensen 7,10,12 ✉ INTRODUCTION Stroke is a leading cause of disability and death worldwide1–3. Effective treatment is time-sensitive, and an optimal outcome is more likely when treatment is administered within the ﬁrst four and a half hours from stroke onset4,5. The gateway to ambulance transport and hospital admittance is through prehospital telehealth services, including emergency medical call centres, nurse advice call lines, and out-of-hours health services. In the prehospital setting, the use of mobile stroke units has made it possible to deliver advanced treatment faster6,7. As the mobile stroke unit is only dispatched to patients with a suspected stroke, the impact of the mobile stroke unit is directly inﬂuenced by accurate call-taker recognition of stroke6,7. Call-takers who can rapidly and accurately recognise stroke are therefore crucial in facilitating prompt care in both prehospital and in-hospital settings. Despite initiatives to improve stroke recognition8,9, approximately half of all patients with stroke do not receive the correct triage for their condition from call-takers 10–12. Most initiatives aim to improve stroke recognition by call-takers via introducing more speciﬁc assessment tools8,9 or providing specialised training 13. Recent advances in machine learning technology might be applied to improve stroke recognition without requiring changes to the triaging approach, and machine learning-aided identiﬁcation of stroke has been suggested as a means of improving mobile stroke unit effectiveness7. Real-time feedback from a machine learning model can improve the recognition of out-of-hospital cardiac arrest14, 15. Therefore, this study aimed to develop and assess the potential of machine learning in improving prehospital stroke recognition during medical helpline calls. In this study, we use call recordings and registry data from the Copenhagen Emergency Medical Services (CEMS) and the Danish Stroke Regis

FAQs

What was the study aiming to achieve?

The study aimed to develop and assess whether machine learning can improve prehospital stroke recognition during medical helpline calls.

How were the machine learning models built from the call data?

Calls were transcribed using an automatic speech recognition model, then classified with a text classification model trained as an ensemble of five individual models.

How was model performance evaluated against call-takers?

The model was compared with call-takers using MH-1813 data from 2021, and test data without diagnostic category codes were also analyzed to assess potential bias.

Sep 17, 2026

Upload by Oliver