AI tools rule out heart attacks and diagnoses lung cancer early

Two new artificial intelligence (AI) tools have been shown to speed up diagnoses and improve care.

Rapid-RO can rapidly rule out heart attacks in people attending A&E, a new study has found. The research, carried out by Imperial College London with funding from the British Heart Foundation (BHF), showed that the tool successfully ruled out heart attacks in more than a third of patients in the four UK hospitals taking part in the trial.

Professor James Leiper, associate medical director at the BHF, said: “This research demonstrates the important role AI could play in guiding treatment decision for heart patients. By quickly identifying patients who are safe to be discharged, this technology could help people avoid unnecessary hospital stays, allowing valuable NHS time and resource to be redirected to where it could have the greatest benefit.”

Normally hospitals use blood tests to confirm diagnosis when a patient arrives at hospital with a suspected heart attack. These tests measure levels of a protein called troponin which rises when there is damage to the heart muscle.

This increase may not be reliably seen until hours later, however, so patients are often kept in hospital for further troponin tests and monitoring. Some will eventually be discharged without needing treatment after a heart attack is ruled out.

Combines blood test and patient information data

PhD student Dario Sesia developed Rapid-RO as a means to identify patients who are at very low risk of heart attack. Sesia trained the tool using data from over 60,000 patients across the UK and then tested on more than 35,000 patients.

It works by combining the data from the initial troponin blood test with other patient information collected during hospital admission, which is then analysed by the algorithm.

Patients are then identified as either being in a very low risk group for having a heart attack, or not. Rapid-RO was able to successfully rule out heart attacks in 36% of patients, compared to 27% ruled out by troponin blood testing alone. It was also found to be more accurate at identifying heart attack cases. Troponin tests missed four times as many heart attacks (108 cases) as the AI tool (27 cases), the researchers said.

AI tool can cut waiting times for lung cancer treatment

While the tool is yet to be made commercially available, 40 NHS trusts have adopted an AI chest X-ray tool to support early lung cancer diagnosis.

The Annalise.ai  tool will be available to radiologists across six imaging networks: Greater Manchester North East and North Cumbria, Cheshire and Merseyside, East Midlands, Surrey, Sussex and Frimley and Yorkshire. It is being rolled out in a staged process with full implementation expected by November 2024.

The tool will support the analysis of approximately 2.5 million chest X-rays a year. According to a study by Annalise.ai, published in The Lancet in July 2021, it is capable of identifying up to 124 different findings on chest X-rays and can help facilitate quicker diagnosis of serious conditions like lung cancer.

It can also be used in intensive care settings to identify acute conditions such as chest sepsis and incorrectly placed lines and tubes, helping to reduce the occurrence of “never events” in hospitals.

NHS Grampian has been using Annalise.ai since 2022. It has found that the tool shortened the interval between chest X-ray assessment and the start of lung cancer treatment by an average of nine days and increased the detection of cancers at earlier stages by 27%.

The roll-out is being paid for by NHS England’s AI Diagnostic Fund (AIDF).

The £21 million fund, announced by NHS England in June 2023, aims to improve the deployment of AI and machine learning tools that can speed up the diagnosis and treatment of serious illness including cancer, stroke and heart conditions.