A study led by King’s College London (KCL) has revealed that Black patients are less likely to be underdiagnosed with a common type of heart failure when using artificial intelligence (AI), compared to in routine practice.
The research, funded by the British Heart Foundation, could help researchers understand the extent of heart failure with preserved ejection fraction (HFpEF) underdiagnosis across ethnicities, as well as reduce bias and improve diagnoses.
Heart failure is estimated to affect more than one million people in the UK, 50% of whom have HFpEF, which occurs when the heart pumps out blood normally but cannot fill up as well, leading to signs and symptoms of failure such as breathlessness, fatigue and dizziness.
Using an AI algorithm called Natural Language Processing (NLP), which reads and understands medical text and analyses electronic medical records, researchers identified nearly 1,973 patients who met the current European Society of Cardiology guidelines for a diagnosis of HFpEF – 64% of whom were white, 29% were Black and 7% were Asian.
Aiming to see if these same patients would be effectively diagnosed in routine care without NLP, researchers found that Black and Asian patients were less likely to be underdiagnosed using the AI.
The team believes that this is due to HFpEF being diagnosed partly by using scores from a H2FPEF test, which is not used in the algorithm, while the NLP considers other possible contributing factors, specifically atrial fibrillation, which was more common in people with white and Asian backgrounds, compared to hypertension, which was more common in Black patients.
Ultimately, researchers believe that the HFpEF diagnostic tool could have led to more Black patients being missed and emphasise the need to improve the diagnosis of HPpEF while also analysing AI utilisation to bring about a more accurate diagnosis.
Study co-lead Dr Kevin O’Gallagher, clinician scientist and honorary consultant, interventional cardiology, KCL, commented: “It is vital [that] clinicians are aware of how heart failure presents in patients of all ethnicities if we are to effectively tackle inequalities within the condition.
“More research still needs to be done to improve diagnostic tools. It is crucial that everyone has the same chance of accessing life-enhancing treatment when they need it the most.”

https://pharmatimes.com/news/study-finds-ai-reduces-underdiagnosis-of-black-patients-with-common-heart-failure/