AI assisted pacing

---

A pacemaker is a small electrical device implanted in the chest that uses electrical signals to keep the heart beating in a regular rhythm and prevent it from beating too slowly. However, traditional pacemakers achieve this by stimulating the heart muscle, which produces a very abnormal heartbeat. Over time, these abnormal heart beats can damage the heart, potentially causing heart failure.

Conduction system pacing (CSP) is an innovation in pacemaker treatment. In CSP the heart’s natural electrical pathways are stimulated producing a heartbeat very similar to a normal heartbeat. This means that patients with CSP pacemakers are much less likely to develop heart failure.

Whilst this could be very beneficial for people who need pacemakers, CSP is a new procedure and requires highly specialised expertise to carry it out. One key reason for this is that it can be very difficult for doctors who are not experts in CSP to identify whether CSP has been successful or not. It is even harder to make it possible for the pacemaker itself to identify if CSP has happened successfully. If this were possible it would be a major advance in pacemaker treatments.

This project, led by Dr Ahran Arnold at Imperial College London, will use artificial intelligence to teach a computer program to identify successful CSP from electrocardiograms (ECGs) which trace the electrical activity in the heart.

To start, they will first confirm what successful CSP looks like, by measuring electrical signals in the hearts of patients undergoing the procedure. This information, combined with ECGs of successful and unsuccessful CSP will be used to train a computer programme to recognise a successful procedure. The programme will then be tested to establish whether it can be used as a tool for detecting successful CSP. This will allow the powerful treatment of CSP to be confidently applied in hospitals without CSP ECG experts, allowing many more patients to benefit from the treatment and avoid heart failure.

The team will then take this process one step further and teach the pacemakers themselves to identify successful CSP. This will mean that if more electrical power is required to produce successful CSP for any heartbeat this can be adjusted on-the-go by the pacemaker, without the patient coming to hospital. This will remove barriers for patients to access this kind of treatment, allowing more people to benefit from the latest pacing technologies, both improving quality of life and reducing the risk of heart failure.