Active Heart, Stroke & Blood Computing & AI

Advancing Endoscopic Cardiac Surgery through an AI-Enabled Guided Prediction System

In plain English

AI plain-English summary

Fewer than one in five heart surgeries in the UK currently avoids splitting the breastbone. This project builds an AI system that reads CT scans before surgery, then generates a detailed 3D map of a patient’s heart and surrounding structures, so surgeons can plan and perform endoscopic cardiac procedures through small incisions instead of opening the chest. The problem is practical and widespread. Open-heart surgery requires a sternotomy—cutting through the breastbone—which causes significant pain, long recovery times, and risk of complications. Endoscopic approaches exist but are technically demanding and used in less than 20% of UK cases. Surgeons lack a reliable, automated tool to visualise critical anatomy and plan the safest route for instruments through the chest wall. If the system works, it could shift the default approach for many cardiac procedures. More surgeons would adopt endoscopic methods, leading to smaller scars, less postoperative pain, shorter hospital stays, and lower costs for the NHS. The technology integrates directly into existing surgical planning workflows, so adoption would not require a complete overhaul of hospital processes. The impact is on surgical practice itself—making a safer, less invasive option the routine choice rather than the exception.

View original technical description
Endoscopic cardiac surgery offers patients a minimally invasive alternative to traditional open-heart procedure. This approach involves small incisions and use of an endoscope equipped with a high-definition 3D camera, allowing surgeons to perform complex heart surgeries with enhanced precision. Benefits of endoscopic cardiac surgery include smaller incision resulting in minimal scarring, reduced recovery time, less postoperative pain and reduced complications and better surgical outcomes. Patients undergoing heart surgery fear the sternotomy or chest splitting that is involved. This product will help more surgeons to move away from it and adopt endoscopic approach, and hopefully benefit many patients and the NHS. At present less than 20% of procedures are done with a non sternotomy approach in the UK. Our aim is to empower cardiac surgeons move away from sternotomy approach into endoscopic approach as this not only benefits the patients in terms of having fewer complications and early discharge but also benefits the hospitals in terms of cost savings. Our innovation is a AI-enabled CT-guided prediction system for safe and reproducible methods of performing non-sternotomy cardiac surgery by providing detailed visualisation of critical structures, automating the process and integrating seamlessly into surgical planning workflows.

View the original record at the funder ↗

Related Research

Grants with similar aims, by meaning.

The application of Artificial Intelligence to patients undergoing endoscopic transsphenoidal surgery: a single-site prospective feasibility study.
AI-CARE - Artificial Intelligence for CAtaract surgery Risk Estimation and reduction
AI-enabled ECG to reduce demand for echocardiography and shorten diagnostic waiting lists
Mixed Reality and Computer Vision for Endoscopic Spinal Surgery
Self-propelled soft robotic endoscopes for next-generation gastrointestinal surgery and beyond

Original classification

Collaborative R&D

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.