Safety Analysis of Machine Learning Software to Assess the Risk of Sudden Cardiac Death

点击翻译按钮获取中文摘要

**More people die from cardiovascular diseases than from cancer**. FEN EP Ltd (FEN EP) is a UK Medical Devices SME that is dedicated to saving patients at risk of sudden cardiac death (SCD) in the UK and globally.FEN EP is building and developing a novel and disruptive medical device with embedded software called **PEFA** (Paced Electrogram Fractionation Analysis) that will transform healthcare pathways, while still fitting seamlessly into routine diagnostic pathways in hospitals. FEN EP's patented and disruptive PEFA solution involves correctly diagnosing and dividing patients into high-risk patients who need treatment and low-risk patients who do not. By improving patient selection, FEN EP aims to save patient lives and significantly reduce healthcare costs in the National Health Service (NHS) and beyond. The understanding of arrhythmias that cause SCD is an electrophysiological problem that can be analysed using FEN EP's proprietary technology and is fundamental to improving patient care and improving their quality of life. The core objectives of the project are to build ten devices for further clinical trials in the NHS and are consistent with regulatory requirements in the UK, Europe and the USA. FEN EP's objective to enable the translation of its device into routine clinical care within the NHS as rapidly as possible. The concept, rationale and implementation of this software-driven device into healthcare systems is set out in multiple peer reviewed publications and a recently accepted review in an internationally recognised cardiology journal.**Artificial intelligence** (AI) in healthcare is an overarching term used to describe the use of machine-learning algorithms and software to mimic human cognition in the analysis, presentation and comprehension of complex medical and healthcare data. More specifically, AI is the ability of computer algorithms such as those embedded within PEFA to approximate conclusions based solely on input data, based on machine learning training of previously acquired clinical or synthetically-derived data. This innovate project seeks to ensure compliance to protect both patients and the medical professionals treating them. Thereby, not only protecting the National Health Service and the health of the patients that it serves but also increasing confidence in AI.Government funding is essential to enable the further development of this application, which will also benefit the UK economy and will create new jobs. This application aligns with the UK Government's objectives of "onshoring" its AI and medical diagnostics capabilities and building its high-value manufacturing capabilities, leading to increased exports internationally.