AI-powered early diagnosis for heart attacks

 

The challenge

Cardiovascular diseases (CVDs) are the leading cause of death globally, with approximately 18 million yearly fatalities from CVDs, representing 32% of all global deaths, according to WHO.[1] Each year CVD causes 4 million deaths in Europe, accounting for 45% of all deaths. CVDs are also the single largest cost driver in the global healthcare system.[2]

Amongst CVDs, ischemic heart disease (IHD) is particularly concerning, being the leading cause of death for men and women in developed countries.[3] IHD, also called coronary heart disease, refers to heart problems caused by narrowed heart (coronary) arteries that supply blood to the heart muscle. IHD, either chronic or acute, are responsible for 20% of deaths in Europe, which translates to 1.74 million people annually.[4]

As a leading cause of death and disability worldwide, IHD poses a significant economic burden. In the EU, the total cost of IHD is estimated to be around €59 billion per year, including healthcare costs (32%), lost productivity (33%), and informal care (35%), being 28% of the overall cost of CVD.[5] These costs are expected to continue to rise as the population ages, and the prevalence of risk factors such as obesity and diabetes is increasing.

Acute and chronic coronary diseases are nowadays a significant patient burden for hospital EDs, resulting in healthcare professionals’ workload and stress, which triggers diagnosis and intervention delays and inadequate activations. Current EU healthcare systems and economy cost strategy are unprepared to overcome the growing prevalence of acute cardiovascular events. Governments are the main bodies involved in defining and shortcutting this problem.[6]

The solution

The ASSIST project consortium proposes a digital, end-to-end solution for the triage and diagnosis of acute coronary syndromes, with a focus on the accurate diagnosis of ST-segment elevation myocardial infarction (STEMI), non-ST-segment elevation myocardial infarction (NSTEMI) and false positive activations.

The ASSIST project consortium will develop and validate the first AI-based solution for the early detection and accurate triage of AMI. The solution has two components:

1) Willem-ST will be built on Idoven’s AI-powered platform (Willem™). Willem™ currently has a sensitivity of 70% to detect ST changes. The ASSIST project will increase this sensitivity to >95% by leveraging existing databases to increase the diagnostic sensitivity of AMI with AI algorithms.

Willem™ can analyse ECG data from any lead, wearable, mobile and implantable device in an automatic and standardised way, augmenting both the cardiologist and ED clinician’s ability to identify, triage and diagnose patients at scale.

2) A cloud medical platform that facilitates a real-time alert of suspected STEMI diagnosis as soon as the first ECG is performed and interpreted by Willem-ST. A specific alert system to call out cases that Willem-ST has identified as potential STEMI patients, as well as a prioritisation view to ensure time-sensitive cases are reviewed first by cardiology experts.

The cloud software will optimise diagnosis and prognosis with a single platform to synchronise all the clinicians involved in AMI care protocols (e.g., ambulances, general practitioners and cardiologists). This cloud connectivity platform will be able to capture ECG signals from both proprietary and standard ECG signal formats through hardware, cardiovascular information system (CVIS) and electronic medical records (EMR) integration.

Expected impact

Willem-ST will fill the unmet gap for end-to-end, AI-assisted solutions for AMI triage. Willem-ST will be a first-of-its-kind solution to improve the sensitivity of triaging patients with suspected ACS at first medical contact and reduce the delays in diagnosis time with the aid of AI.

As a result of the project, Willem-ST will deliver a sensitivity to detect ST change of >95% in the triage of patients with suspected ACS and a reduction of time from first medical contact (FMC) to diagnosis by 67%.

Willem-ST would also reduce false positive activations of cardiac catheterisation clinic laboratory (CCL) as a consequence of improving the positive predictive value of the triage done by the first medical contact, leading to cost savings in avoiding costly, unnecessary in-hospital interventions.

Willem-ST solution will benefit patients needing urgent care, improving the early identification and triage of AMI and chronic ischemic heart disease. It will also benefit citizens and institutions by reducing the burden of coronary diseases and by optimising the spending and costs of the healthcare system.

External Partners
  • Interamerican Society of Cardiology (SIAC)
References

[1] WHO (2021). Cardiovascular Diseases (CVDs), World Health Organisation. Accessible at: https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds).

[2] Eurstat (2020). Deaths Due to Coronary Heart Diseases in the EU, Eurostat. Accessible at: https://ec.europa.eu/eurostat/web/products-eurostat-news/-/edn-20200928-1

[3] Khan, M. A., et al. (2020). Global Epidemiology of Ischemic Heart Disease: Results from the Global Burden of Disease Study. Cureus, 12(7), e9349. https://doi.org/10.7759/cureus.9349

[4] Wilkins, E., et al. (2017), European Cardiovascular Disease Statistics 2017, European Heart Network, Brussels, https://ehnheart.org/images/CVD-statistics-report-August-2017.pdf

[5] V Gorasso, et al. (2021). Cost of hospitalisation for ischaemic heart and cerebrovascular diseases in Belgium, European Journal of Public Health, Volume 31, ckab164.111, https://doi.org/10.1093/eurpub/ckab164.111

[6] WHO (2022). Towards a beating cardiovascular disease plan for Europe, Eurohealth, 27 (2), 37 – 40. World Health Organisation. Regional Office for Europe. https://apps.who.int/iris/handle/10665/352273

Manuel Marina
| CEO | Idoven
Contact