Cloud-based AI technology propels cardiovascular medicine forward

The UK government has recognised the potential of technology to improve patient care and save lives. Technology is considered central to the future of healthcare, and Matt Hancock recently outlined a digital strategy that defined tech innovations as key to helping prevent, diagnose and treat different diseases.

The med-tech sector and the NHS have become close partners, with private and public organisations working together to ensure innovation and science continue to improve the patient experience. Some of the technologies that have been adopted by the NHS in recent years include Microsoft’s InnerEye, which assists physicians to mark-up scans for prostate cancer patients and Babylon Health, whose app uses artificial intelligence (AI) to assess medical symptoms and to help ‘triage’ patients with very basic queries.

Another area where AI is helping to advance healthcare is in cardiovascular medicine. Cardiologists can now use AI’s capabilities to collate data, and as more and more of it is gathered and analysed with deep learning technology, doctors can better understand how severe a condition may be, accurately diagnose the disease or explain to patients why they are experiencing certain symptoms. In turn, this can help to formulate the most effective treatment plan possible.

HeartFlow is the result of research I began while a doctoral student at Stanford University more than 20 years ago. I worked with Christopher Zarins, MD, former chief of vascular surgery at Stanford University and an expert in vascular biology, and Thomas J.R. Hughes, PhD, former professor of mechanical engineering at Stanford University and a leading expert in computational fluid dynamics, to develop computing technology to model blood flow in arteries from medical imaging data.  As a professor in the departments of Bioengineering and Surgery at Stanford, I focused on developing this technology for more than a decade before Dr. Zarins and I went on to found HeartFlow in 2007.

At HeartFlow, we brought experts in bioengineering, computational fluid dynamics, cloud computing, computer vision, and artificial intelligence together to revolutionise the diagnosis of heart disease. Supported by robust clinical evidence, including numerous clinical trials and hundreds of peer-reviewed scientific publications, the HeartFlow Analysis received FDA clearance in 2014. HeartFlow has also made its way across the Atlantic where it has caught the attention of the NHS.

The condition

Coronary heart disease (CHD) is the UK’s biggest killer – responsible for more than 66,000 deaths a year across the country. It affects more people than breast cancer and prostate cancer combined. The wide variety of symptoms exhibited by patients with this condition makes it particularly challenging to diagnose – sufferers can experience everything from feelings of indigestion and jaw pain to breathlessness and tightness in the chest.

A lack of public awareness of CHD symptoms may also contribute to the problem. A recent YouGov survey found that two-thirds of respondents could name chest pain or tightness as a symptom of heart disease, while just 16% knew that similar feelings in the abdomen are also symptoms. Even when patients do seek out medical treatment, they often have to wait several weeks to be seen by a doctor and undergo multiple tests before a diagnosis can be made. All in all, it’s tricky to swiftly and effectively diagnose CHD.

Technology goes beyond diagnosis

HeartFlow is helping to streamline the CHD evaluation process. The technology uses deep learning and artificial intelligence to help cardiologists better understand blood flow in a patient’s arteries.

Under the NHS’s Innovation and Technology Payment Programme, the HeartFlow Analysis is being fast-tracked in the UK and is now available in more than 30 hospitals.

The process starts with the hospital securely uploading a CT scan into HeartFlow’s cloud-based software system hosted on Amazon Web Services (AWS). HeartFlow leverages deep learning and highly trained analysts to create a personalised digital 3D model of the patient’s coronary arteries. Next, powerful algorithms solve millions of complex equations that simulate blood flow within the model, so that the impact of any blockages on the arteries can be assessed.

The completed analysis is securely transferred back to the hospital and allows cardiologists to interact with the 3D digital model: selecting specific areas to investigate, zooming in and out, and rotating the image. This enables a level of examination of blood flow in the arteries ordinarily only possible through an invasive procedure.

The benefits of such an approach are clear. Patients can receive a diagnosis without undergoing a surgical procedure, and physicians can have greater confidence in their patients’ treatment and management plans.

The future of AI in healthcare

Technology is constantly developing, and recent years have seen its application in healthcare accelerate rapidly. HeartFlow is leading the way in this, applying artificial intelligence to help make the diagnostic process for heart disease more accurate and efficient.

There is still so much room for further development, though. Technology will most certainly play a vital role in helping to tackle our health challenges. And in the hands of the medical professionals who interpret the data and put it to work, it has the potential to improve and save countless lives.

+ posts

Cloud Industry Forum presents TWF! 2024 state of the cloud report


Related articles

Start Bridging the Cloud Skill Gap Today

In today's rapidly evolving digital landscape, cloud computing has...

How Data Fabric Helps Address Multi-Cloud Sprawl

The abundance of data facilitates good decision-making, but too...

CIF Presents TWF – Dean & Sarah-Jane Gratton

In this episode of the 2024 season of our...

The Quantum Leap in Internet Technology

Envision a world where the internet, as we currently...

Translating ancient code for the generative AI era

Having easy access to data is critically important in...

Subscribe to our Newsletter