Researchers Use Artificial Intelligence to Analyze 3D Virtual Hearts and Predict Cardiac Patients’ Survival Rates

Share this Article

imperial-college-london-logoIf this study I’m writing about could have a theme song, I personally think it should be Céline Dion’s timeless earworm from Titanic, “My Heart Will Go On.” I know that sounds pretty cheesy, but hear me out: the results of the study, published in the journal Radiology, reveal that, based on analysis of virtual 3D hearts, artificial intelligence (AI) can actually help doctors identify which of their pulmonary hypertension patients have the greatest risk of heart deterioration. That’s right – AI that can predict the survival rates of heart disease patients! Clinicians from the MRC London Institute of Medical Sciences (LMS), together with Imperial College London scientists, developed software that is able to analyze MRI scan images, and use them to build a smart 3D heart that predicts patient survival rates. It’s no 3D printed heart model, but a virtual one.

This could have a major impact on the future of medical technology, as explored in the paper, “Machine Learning of Three-dimensional Right Ventricular Motion Enables Outcome Prediction in Pulmonary Hypertension: A Cardiac MR Imaging Study.”

“This is the first time computers have interpreted heart scans to accurately predict how long patients will live. It could transform the way doctors treat heart patients,” said Dr. Declan O’Regan, Research Lead, LMS. “The computer is up to 80% accurate at predicting survival at one year. A doctor equipped with this new cardiac imaging approach would therefore be able to make more informed judgments about outcome than if they were relying only on current ways to investigate patient data.”


Computer software creates a 3D virtual heart from MRI scans, then learns to predict when patients will die [Image: Imperial College London]

According to Imperial College London, the technology was tested on patients suffering from pulmonary hypertension, which can lead to heart failure if not treated. Data from more than 250 patients who were referred to the National Pulmonary Hypertension Service at the Imperial College Healthcare NHS Trust between May of 2004 and October of 2013 were included in the software. By using the MRI scans, and information like patients’ blood tests, the software was used to create 3D virtual hearts that basically replicated how over 30,000 points in the heart contract with every beat.

The study states that, “Computational modeling provides a platform for improving or understanding of the heart, and the integration of experimental and clinical data is now bringing computational models closer to use in routine clinical practice.”

Using the 3D virtual hearts, the AI can quickly determine which features of the simulated cardiac function were the earliest predictors of heart failure and eventual death. The results show that this machine learning program is faster, and more accurate, at making these types of predictions than currently available methods. Co-author of the study Dr. Tim Dawes developed the algorithms for the software.

“The computer performs the analysis in seconds and simultaneously interprets data from imaging, blood tests and other investigations without any human intervention,” said Dr. Dawes, LMS. “It could help doctors to give the right treatments to the right patients, at the right time.”


From the paper: Example of computational modeling for a patient with idiopathic pulmonary arterial hypertension. A, Cine MR images were segmented by using prior knowledge from a set of disease-specific atlases. Here, the intensity image in the short-axis of the heart is overlaid with labels for left ventricular blood pool (red), myocardium (green), RV blood pool (yellow), and free wall (blue). B, A 3D model at end-diastole (gray) and end-systole (blue, right ventricle; and red, left ventricle) was used to determine the direction and magnitude of systolic excursion at each corresponding anatomic point in the mesh by using a deformable motion model. C, A statistical model of RV endocardial motion was used for feature selection to determine functional patterns associated with survival (relative weightings shown for the RV free wall).

Up until this point, radiologists had to take lengthy heart function measurements, by hand, to determine which patients were at the greatest risk. But now, that could all change. The researchers say that this innovative technology could potentially be used on patients with other types of heart disease. The software will be tested using data from Hammersmith Hospital next, to verify the findings and figure out if it’s really as smart as it thinks it is, and can predict the type of treatment that would work best for those patients.

Dr. Dawes said, “We would like to develop the technology so it can be used in many heart conditions to complement how doctors interpret the results of medical tests. The goal is to see if better predictions can guide treatment to help people to live longer.”

According to the team of researchers, although AI has been used to help research cancer and brain diseases, this is the first study to use AI to predict heart disease outcomes. If they can verify their findings, it would enable physicians to more quickly determine which pulmonary hypertension patients are the most at-risk, and start treating them fast, so they are able to live longer.

heart-scan-computers“Pulmonary arterial hypertension (PAH) is a rare but devastating condition which can substantially reduce your quality of life or lead to death at a young age,” said Dr. Mike Knapton, Associate Medical Director at the British Heart Foundation, which partially funded this AI study. “It is currently incurable, and it can be difficult for doctors to know which treatments will work best for individual patients. This exciting use of computer software in clinical practice will help doctors in the future to make sure that patients are receiving the correct treatment before the condition deteriorates and leaves them needing a lung-transplant. The next step is to test this technology in more hospitals.”

The study was authored by Timothy J. W. Dawes, FRCA, PhD; Antonio de Marvao, MRCP, PhD; Wenzhe Shi, PhD; Tristan Fletcher, PhD, MSc; Geoffrey M. J. Watson, MRCP; John Wharton, PhD; Christopher J. Rhodes, PhD; Luke S. G. E. Howard, FRCP, DPhil; J. Simon R. Gibbs, MD, FRCP; Daniel Rueckert, PhD; Stuart A. Cook, FRCP, PhD; Martin R. Wilkins, MD, FMedSci; and Declan P. O’Regan, FRCR, PhD. To see one of the 3D heart models in action, take a look at this short video:

Discuss in the Patient Survival Rate forum at

[Sources: British Journal-Healthcare / Imperial College London]


Share this Article

Recent News

3D Printing News Briefs, July 13, 2024: Metal 3D Printer, AFWERX Award, & More

3D Printing Markets Grows 8% Year over Year


3D Design

3D Printed Art

3D Printed Food

3D Printed Guns

You May Also Like

Vision Miner Acquires its 3D Printer Supplier AddWise

Vision Miner, a provider of industrial 3D printing solutions, has announced the acquisition of AddWise, a manufacturer of 3D printers and related products, in a deal valued that the companies...

“Auto Repair Needs 3D Printing” – Harold Sears Weighs in on Auto Additive’s Launch

Despite the automotive sector’s long-time adoption of additive manufacturing (AM), the use of the technology for end parts in consumer vehicles is only just now beginning to take off. And,...


Formlabs Buys Nascent SLS 3D Printer Competitor Micronics

Formlabs, maker of accessible yet professional 3D printers, has acquired Micronics, which recently debuted with a claim of making a $2,999 3D printer. I, for one, was pretty incredulous about...

The Producers: HP’s President of 3D Printing Savi Baveja Explains How the Company is Addressing Scalability

HP (NSYE: HPQ) and the additive manufacturing (AM) industry in the US need each other. In the long run, I believe that what’s good for one will be good for...