Development and validation of an AI decision-support tool to improve the equitable diagnosis and risk stratification of patients presenting to the Emergency Department

Background

Cardiovascular disease frequently presents with symptoms such as acute chest pain or breathlessness. Together, individuals with these symptoms account for 1 in 4 presentations to the Emergency Department and 1 in 10 unscheduled hospital admissions in the United Kingdom.¹ Prompt and accurate diagnosis of the cause of these symptoms is crucial, particularly for conditions like acute myocardial infarction, acute heart failure or pulmonary embolism, where delays in diagnosis and treatment may be associated with sudden death. Clinical assessment is not straightforward, as symptoms and signs from these conditions are frequently indistinguishable from other common and less serious conditions. Key biomarkers are often elevated in other conditions and can be challenging to interpret even for experienced clinicians.

Current pathways for the diagnosis of myocardial infarction, acute heart failure and pulmonary embolism have important limitations. First, the majority of care pathways rely on fixed biomarker thresholds that do not account for age, sex, or comorbidities, resulting in variable diagnostic performance and inequalities in care. Second, they are inefficient, with most patients requiring serial biomarker measurements or additional imaging with a low diagnostic yield. Third, with some notable exceptions, current care pathways do not incorporate important clinical features that have a major influence on pre-test probability. Whether these pathways perform equally well in socioeconomically deprived populations, who carry the greatest burden of cardiovascular disease, remains largely unknown.

Artificial intelligence (AI) has potential to improve patient care by helping clinicians make more accurate and timely diagnoses. Our previous AI algorithms have demonstrated superior performance compared to traditional methods for the diagnosis of myocardial infarction and heart failure.²⁻⁴ However, current models have three gaps. They are limited to single conditions. They fail to account for the fact that patients often require evaluation for multiple conditions at once. And they have not been evaluated across socioeconomic groups. Without this, AI risks reproducing, or even widening, existing inequalities in care.

Aims

This project aims to develop and validate an AI-based decision-support tool for the Emergency Department that integrates multi-modal data to improve the diagnosis of acute cardiovascular conditions, and to ensure that it performs equitably across all patient groups. Specifically, the AI-based decision-support tool will:

• Integrate diverse data sources by combining patient demographics, clinical history, cardiac biomarkers and the electrocardiogram (ECG) to improve diagnostic accuracy.
• Provide real-time diagnostic support by offering individualised diagnostic probabilities for myocardial infarction, heart failure and pulmonary embolism, helping clinicians make faster and more precise decisions and reducing unnecessary investigations and admissions.
• Ensure equitable performance by evaluating the tool across sex, age, ethnicity and socioeconomic deprivation, and addressing any disparities before implementation.

We will use machine learning and deep learning approaches in consecutive patients with suspected acute cardiovascular disease from the Heart Disease Registry (https://dataloch.org/) as the training set. External validation will be conducted in patients from NHS Greater Glasgow and Clyde, a population with a high burden of socioeconomic deprivation. It will also be conducted in patients from the United States, Australia and New Zealand, to ensure model calibration and generalisability across diverse populations.

Training Outcomes

• Clinical and data science expertise: The student will gain experience in applying machine and deep learning to real-world healthcare problems at the interface of clinical practice and data analytics. 
  • Advanced statistical methodology and programming skills: The student will learn cutting-edge techniques in machine learning and deep learning, including the analysis of physiological signals such as the ECG. 
  • Health equality and responsible AI: The student will learn how to assess and mitigate bias in AI tools, ensuring that they benefit all patients. 
  • Interdisciplinary collaboration: Close work with clinicians and data scientists across Edinburgh and Glasgow will enhance the student's skills in interdisciplinary communication and the translation of AI tools into clinical practice.

References: 

¹ British Heart Foundation Heart and Circulatory Diseases Statistics, British Heart Foundation, 2023. (www.bhf.org.uk).

² Doudesis D, Lee KK, Yang J et al. Validation of the myocardial-ischaemic-injury-index machine learning algorithm to guide the diagnosis of myocardial infarction in a heterogenous population: a prespecified exploratory analysis. The Lancet Digital Health 2022;4(5):e300-e308.

³ Lee KK, Doudesis D, Anwar M, et al. Development and validation of a decision support tool for the diagnosis of acute heart failure: systematic review, meta-analysis, and modelling study. BMJ. 2022;377:e068424.

⁴ Doudesis D, Lee KK, Boeddinghaus J, et al. Machine learning for diagnosis of myocardial infarction using cardiac troponin concentrations. Nature Medicine. 2023;29(5):1201-10.