Recipient organisationLighthearted Ai Health Limited
Funding£210K
PeriodApr 2025 — Apr 2026
In plain English
AI plain-English summary
A contactless optical stethoscope called LightScope aims to diagnose heart disease by capturing heart sounds, ECG, respiration, and blood pressure from a single body location in 10 seconds, using laser technology and AI. This matters because heart disease causes 25% of preventable deaths in the UK, yet is often detected too late. At-risk groups—women and ethnic minorities—face higher misdiagnosis rates and poorer outcomes due to limited access to early diagnosis. Current diagnostic tools for aortic stenosis, a heart valve disease affecting over 1.5 million people in the UK, miss many cases, leading to heart failure and death. If the clinical validation trial involving 400 patients confirms safety and effectiveness, LightScope could transform cardiac diagnostics. It would enable rapid, accurate detection in community settings rather than hospitals, reducing unnecessary hospitalisations and improving outcomes for marginalised groups. The device is designed for CE marking as a Class IIA medical device, paving the way for NHS adoption and integration with remote patient monitoring and telehealth companies. Beyond aortic stenosis, the company plans to expand AI algorithms to diagnose congenital heart disease, prosthetic valve issues, ischaemic heart disease, and heart failure using data from over 3,000 patients.
View original technical description
Heart disease, responsible for 25% of preventable deaths, is often detected too late.(1) At-risk groups such as women and ethnic minorities face higher misdiagnosis rates and poorer outcomes due to limited access to early diagnosis.(2-9) The NHS Darzi review emphasized the importance of prevention, early diagnosis, and utilizing new technologies to shift care closer to home. The Minister for Public Health reinforced this by stating, "We know so many deadly diseases can be avoided if we seek help in enough time."(10) To address this critical issue, LightHearted AI has developed the LightScope---a contactless optical stethoscope that uses laser technology and AI to rapidly capture heart sounds, ECG, respiration rates, and blood pressure from one single body location in 10 seconds. This device then provides an instantaneous AI-assisted diagnosis for several conditions, offering greater accuracy than traditional methods. Our initial focus is on diagnosing aortic stenosis (AS), a growing global health issue. In the UK, heart valve diseases like AS affect over 1.5 million people, with many cases going undiagnosed. Without timely intervention, AS can lead to heart failure and death.(11) LightScope's innovative approach offers a non-invasive, fast, and reliable method to detect AS, addressing the limitations of current diagnostic tools. We will conduct a UK-based clinical validation trial involving 400 patients to validate the safety and effectiveness of LightScope in diagnosing AS. The trial will generate the final piece of evidence for CE marking (Class IIA) submission and pave the way for commercialisation within the NHS, as well as adoption by remote patient monitoring and telehealth companies. Beyond AS, LightHearted AI plans to develop and expand its AI algorithms to diagnose other heart conditions including congenital heart disease, prosthetic heart valve monitoring, ischaemic heart disease and heart failure using a proprietary dataset from over 3,000 patients. Securing funding will help advance this mission, to improve early detection and hence outcomes, reduce healthcare disparities, and accelerate LightScope's route to market. LightScope's potential to transform cardiac care is vast. By enabling early diagnosis this technology will reduce unnecessary hospitalisations and improve patient outcomes, particularly for marginalised groups. Additionally, LightScope supports the decentralisation of care by enabling diagnosis in community settings, making it a cost-effective solution for overburdened healthcare systems.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know