AI to help find cardiac rhythm devices
Thursday March 28th, 2019
A new AI system can help identify implanted electronic devices in emergency situations, developers have announced.
A project at Hammersmith Hospital has tested the system, showing that it can identify the type and model of device implanted in a patient "quickly and accurately."
Currently staff have to use flowchart algorithms to identify devices when failure puts a patient at risk. These are now "out-dated" and inaccurate, according to the developers.
The new system has been "trained" to identify more than 1,600 cardiac devices.
It was tested on radiographic images of more than 1,500 patients from Hammersmith.
The tests, reported in JACC: Clinical Electrophysiology, showed it to be 99% accurate in identifying manufacturers – compared with a 72% success rate for cardiologists armed with the flow-charts.
* A second project reported yesterday suggests that AI is now able to predict the risk of premature death among middle-aged people.
Researchers at Nottingham University developed their new system based on records of more than half a million people on the UK Biobank.
Reporting in PLoS One, they say the system is "very accurate" in its predictions.
Assistant Professor of Epidemiology and Data Science Dr Stephen Weng, said: "Preventative healthcare is a growing priority in the fight against serious diseases so we have been working for a number of years to improve the accuracy of computerised health risk assessment in the general population. Most applications focus on a single disease area but predicting death due to several different disease outcomes is highly complex, especially given environmental and individual factors that may affect them.
"We have taken a major step forward in this field by developing a unique and holistic approach to predicting a person's risk of premature death by machine-learning. This uses computers to build new risk prediction models that take into account a wide range of demographic, biometric, clinical and lifestyle factors for each individual assessed, even their dietary consumption of fruit, vegetables and meat per day.
"We mapped the resulting predictions to mortality data from the cohort, using Office of National Statistics death records, the UK cancer registry and 'hospital episodes' statistics. We found machine learned algorithms were significantly more accurate in predicting death than the standard prediction models developed by a human expert."
Cardiac Rhythm Device Identification Using Neural Networks. JACC: Clinical Electrophysiology in press
Prediction of premature all-cause mortality: A prospective general population cohort study comparing machine-learning and standard epidemiological approaches. PLoS One 27 March 2019
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0214365
Tags: A&E | Heart Health | UK News
Printer
friendly page