- A new algorithm
can accurately predict the life expectancy of a patient with heart
- The algorithm is
based on a machine learning approach that uses 53 data points including
age, gender, body mass index, blood type and blood chemistry.
- The tool can help
determine which patients will survive heart failure, for how long and
whether they need a heart transplant.
machine-learning-based algorithm can help save lives by accurately predicting
the life expectancy of patients who have had a heart failure. The algorithm was
developed by a research team at UCLA led by Mihaela van der Schaar,
Chancellor's Professor of Electrical and Computer Engineering at the UCLA Samueli School of Engineering.
The study is published in the journal PLOS One.
called Trees of Predictors, uses machine
learning, which means that computers effectively "learn" from
additional new data over time. Fifty three data points are taken into account
to address differences between potential heart donors and recipients. 33 of the
data points relate to information about the potential recipients, 14 pertain to
the donors and 6 apply to the compatibility between donor and recipient. The
data points include analysis of age, gender, body mass index, blood type and
What does the
uses the 53 data points to predict how long people with heart failure
will live, based on whether they
receive a transplant or not. The algorithm may allow doctors to personalize
treatments for people awaiting heart transplants.
‘A new algorithm called “Trees of Predictors” uses machine learning to accurately predict life expectancy after heart failure.’
"Our work suggests
that more lives could be saved with the application of this new
machine-learning-based algorithm," said Van der Schaar, a Turing Fellow at
the Alan Turing Institute in London, and the Man Professor at University of
Oxford. "It would be especially useful for determining which patients need heart
most urgently and which patients are good candidates for
bridge therapies such as implanted mechanical-assist devices."
How the algorithm performed compared to
team tested the new algorithm on 30 years of data of people registered with the
United Network for Organ Sharing, an organization that matches donors and
transplant recipients in the U.S.
observed that the algorithm provided significantly better predictions for how
long a patient would live for compared to current methods that health care
Trees of Predictors was compared with currently used prediction models to
project which transplant recipients would live for at least three years after a
transplant, the new algorithm outperformed the other models by 14%. The
algorithm correctly predicted that 2,442 more heart transplant recipients of
the 17,441 who received transplants and lived at least that long after the
method, we are able to identify a significant number of patients who are good
transplant candidates but were not identified as such by traditional
approaches," said Dr. Martin Cadeiras, a cardiologist at the David
Geffen School of Medicine at UCLA. "This
methodology better resembles the human thinking process by allowing multiple
alternative solutions for the same problem but taking into consideration the
variability of each individual."
- Matthew Chin and Amy Akmal, "New algorithm more accurately predicts life expectancy after heart failure" (2018), UCLA