Clinical validation of an artificial intelligence-based decision support system for diagnosis and risk stratification of heart failure (STRATIFYHF) : a protocol for a prospective, multicentre longitudinal study

Charman, Sarah Jane (author), Okwose, Nduka C. (author), Bosnić, Zoran (author), Vračar, Petar (author), Bano, Fatima (research coworker), Pičulin, Matej (research coworker), Flis, Borut (research coworker)

Introduction Heart failure (HF) is a complex clinical syndrome. Accurate risk stratification and early diagnosis of HF are challenging as its signs and symptoms are non-specific. We propose to address this global challenge by developing the STRATIFYHF artificial intelligence-driven decision support system (DSS), which uses novel analytical methods in determining the risk, diagnosis and prognosis of HF. The primary aim of the present study is to collect prospective clinical data to validate the STRATIFYHF DSS (in terms of diagnostic accuracy, sensitivity and specificity) as a tool to predict the risk, diagnosis and progression of HF. The secondary outcomes are the demographic and clinical predictors of risk, diagnosis and progression of HF. Methods and analysis STRATIFYHF is a prospective, multicentre, longitudinal study that will recruit up to 1600 individuals (n=800 suspected/at risk of HF and n=800 diagnosed with HF) aged ≥45 years old, with up to 24 months of follow-up observations. Individuals suspected of HF will be divided into two categories based on current definitions and predefined inclusion criteria. All participants will have their medical history recorded, along with data on physical examination (signs and symptoms), blood tests including serum natriuretic peptides levels, ECG and echocardiogram results, as well as demographic, socioeconomic and lifestyle data, and use of complete novel technologies (cardiac output response to stress test and voice recognition biomarkers). All measurements will be recorded at baseline and at 12-month follow-up, with medical history and hospitalisation also recorded at 24-month follow-up. Cardiovascular MRI assessment will be completed in a subset of participants (n=20–40) from eligible clinical centres only at baseline. Each clinical centre will recruit a subset of participants (n=30) who will complete a 6-month home-based monitoring of clinical characteristics and accelerometry (wrist-worn monitor) to determine the feasibility and acceptability of the STRATIFYHF mobile application. Focus groups and semistructured interviews will be conducted with up to 15 healthcare professionals and up to 20 study participants (10 at risk of HF and 10 diagnosed with HF) to explore the needs of patients and healthcare professionals prior to the development of the STRATIFYHF DSS and to evaluate the acceptability of this mobile application. Ethics and dissemination Ethical approval has been granted by the East Midlands - Leicester Central Research Ethics Committee (24/EM/0101). Dissemination activities will include journal publications and presentations at conferences, as well as development of training materials and delivery of focused training on the STRATIFYHF DSS and mobile application. We will develop and propose policy guidelines for integration of the STRATIFYHF DSS and mobile application into the standard of care in the HF care pathway. Trial registration number NCT06377319.

• Language: English
• Keywords: heart failure, decision support, risk stratification, clinical validation, artificial intelligence
• Work type: Article
• Typology: 1.01 - Original Scientific Article
• Organization: FRI - Faculty of Computer and Information Science
• Publication status: Published
• Publication version: Version of Record
• Year: 2025
• Number of pages: 9 str.
• Numbering: Vol. 15, iss. 1
• PID: 20.500.12556/RUL-171504
• UDC: 004.8:616.12-008.46
• ISSN on article: 2044-6055
• DOI: 10.1136/bmjopen-2024-091793
• COBISS.SI-ID: 222988547
• Publication date in RUL: 27.08.2025

Record is a part of a journal

• Title: BMJ open
• Publisher: BMJ Publishing Group
• ISSN: 2044-6055
• COBISS.SI-ID: 30480601

Licences

• License: CC BY-NC 4.0, Creative Commons Attribution-NonCommercial 4.0 International
• Link: http://creativecommons.org/licenses/by-nc/4.0/
• Description: A creative commons license that bans commercial use, but the users don’t have to license their derivative works on the same terms.

Secondary language

• Language: Slovenian
• Keywords: srčna odpoved, podpora odločanju, napovedovanje rizika, klinična validacija, umetna inteligenca

Projects

• Funder: EC - European Commission
• Funding programme: HE
• Project number: 101080905
• Name: Artificial intelligence-based decision support system for risk stratification and early detection of heart failure in primary and secondary care
• Acronym: STRATIFYHF

• Funder: UKRI - UK Research and Innovation
• Funding programme: Horizon Europe Guarantee
• Project number: 10073472
• Name: STRATIFYHF: Artificial intelligence-based decision support system for risk stratification and early detection of heart failure in primary and secondary care