Science Tech
A software platform driven by artificial intelligence, named PreSize, has demonstrated the potential to revolutionize the treatment of brain aneurysms. According to researchers, PreSize exhibited greater accuracy in placing a stent compared to experienced neurosurgeons.
A brain aneurysm is a vulnerable or thin area on a blood vessel wall in the brain that becomes filled with blood, posing a life-threatening risk if it ruptures. Stents, mesh tubes that spring open to redirect blood flow and prevent excess pressure, are commonly used in treatment.
Currently, there is a significant amount of uncertainty in stent placement, involving a degree of “guesswork,” as explained by Dr. Katerina Spranger, the founder of Oxford Heartbeat, the company behind PreSize. Complications can arise during the placement of stents using a catheter in the groin and X-ray guidance, such as detachment or vessel rupture.
PreSize offers an accurate representation of the patient’s brain, along with “digital twins” that mimic the behavior of a stent in real-life conditions, predicting the outcome of placing a specific device in a particular location. This allows surgeons to experiment with various stents in the operating theatre to determine the most effective option.
The software has already been used in over 375 operations, including those in seven hospitals in England and Scotland, as well as some in Germany, Finland, and Ukraine.
The National Institute for Health and Care Research (NIHR) acknowledges the substantial potential of AI technologies in expediting diagnosis and treatment, thereby enhancing patient outcomes. NIHR is supporting various AI projects, including PreSize, in collaboration with the NHS AI Lab, aiming to gather evidence for wider implementation across the NHS.
Initial research results indicate a significant advantage for PreSize. A study published in the BMJ Journal of Neurointerventional Surgery revealed that PreSize achieved a 95% accuracy rate in predicting the stent’s placement location in the brain, surpassing the 81% accuracy rate of human clinicians.
While recognizing the transformative potential of AI in healthcare, Dr. Spranger emphasized the importance of rigorous testing before widespread adoption. She highlighted the need for medical device software companies to be held accountable to address any concerns or distrust surrounding AI in healthcare.
